WEBVTT

1
00:00:06.631 --> 00:00:08.466
Hi everyone, I'm Kali Bateman

2
00:00:08.466 --> 00:00:09.426
 here for Mixing Light.

3
00:00:09.634 --> 00:00:11.761
And today I've got Bryn Morrow here,

4
00:00:12.053 --> 00:00:13.930
who's the virtual production supervisor

5
00:00:14.222 --> 00:00:15.515
for Steel Bridge Studios.

6
00:00:16.182 --> 00:00:17.809
It's really exciting to have you here

7
00:00:17.892 --> 00:00:20.645
to have a chat today about volume work.

8
00:00:21.438 --> 00:00:24.691
We met grading some car commercials together,

9
00:00:24.858 --> 00:00:27.736
which actually I didn't know were shot on a volume

10
00:00:27.819 --> 00:00:30.155
until about halfway through the session.

11
00:00:30.488 --> 00:00:32.532
And I turned off a resize and went,

12
00:00:32.699 --> 00:00:34.200
oh, look, there's the edge of the universe.

13
00:00:35.034 --> 00:00:37.328
And it was really impressive stuff.

14
00:00:37.704 --> 00:00:39.122
Once you started to look for it,

15
00:00:39.205 --> 00:00:40.790
you could start to say, oh yeah,

16
00:00:40.874 --> 00:00:42.751
maybe that's something to do with the volume,

17
00:00:42.834 --> 00:00:44.461
but it was pretty seamless stuff.

18
00:00:45.295 --> 00:00:48.381
And we got to talking about how it all works.

19
00:00:48.465 --> 00:00:50.675
And I realized how little I knew

20
00:00:50.759 --> 00:00:53.470
about how volume production works.

21
00:00:54.137 --> 00:00:55.930
So very excited to have you here

22
00:00:56.014 --> 00:00:58.016
to shed a bit of light on that today.

23
00:00:58.475 --> 00:00:59.726
Thanks so much for coming.

24
00:01:00.518 --> 00:01:01.060
- Oh, thank you.

25
00:01:01.352 --> 00:01:02.896
I'm very happy to be here.

26
00:01:02.979 --> 00:01:05.648
So thanks for the nice introduction, Kali.

27
00:01:06.858 --> 00:01:08.026
- Such a pleasure.

28
00:01:08.943 --> 00:01:11.070
So can you just tell me,

29
00:01:11.821 --> 00:01:13.448
basically from the very beginning,

30
00:01:14.449 --> 00:01:18.369
you need to have a image to go on these volumes.

31
00:01:18.453 --> 00:01:20.663
So just to try to explain what a volume is,

32
00:01:20.747 --> 00:01:22.499
and you'll probably do a far better job than me,

33
00:01:23.291 --> 00:01:26.836
but a large LED wall or walls

34
00:01:27.504 --> 00:01:31.633
that serve as a background for motion pictures.

35
00:01:32.550 --> 00:01:34.260
So cameras in front of the volume

36
00:01:34.344 --> 00:01:36.221
and usually actors or props,

37
00:01:36.805 --> 00:01:39.766
some kind of integration of physical sets

38
00:01:40.475 --> 00:01:42.352
in between the volume and the camera.

39
00:01:42.852 --> 00:01:45.021
And then a whole bunch of data is recorded

40
00:01:45.772 --> 00:01:49.776
and the backgrounds are happening in real time

41
00:01:50.109 --> 00:01:51.110
through Unreal Engine.

42
00:01:51.611 --> 00:01:53.321
Have I kind of got the basics down there?

43
00:01:53.863 --> 00:01:56.366
- Yeah, actually, I think you did very well there.

44
00:01:56.449 --> 00:02:00.453
I mean, there's obviously variations

45
00:02:00.829 --> 00:02:04.249
to how virtual production is done,

46
00:02:04.457 --> 00:02:07.210
but that is certainly one way of doing it.

47
00:02:07.293 --> 00:02:11.548
And in the project that you were grading for us,

48
00:02:12.924 --> 00:02:15.134
that was essentially how it was done.

49
00:02:15.218 --> 00:02:16.678
Like the backgrounds were

50
00:02:16.761 --> 00:02:20.598
projected onto an LED panel,

51
00:02:20.682 --> 00:02:23.017
like a large sort of screen,

52
00:02:23.101 --> 00:02:26.354
which is a combination of a

53
00:02:26.437 --> 00:02:28.439
bunch of very small panels

54
00:02:28.523 --> 00:02:30.275
as well that we combine together

55
00:02:30.483 --> 00:02:33.444
and that particular one,

56
00:02:35.029 --> 00:02:37.031
from if memory serves me right,

57
00:02:37.115 --> 00:02:40.827
that was about, I think it was about 15 meters

58
00:02:41.035 --> 00:02:46.291
by four and a half meters high for the main wall.

59
00:02:46.875 --> 00:02:49.210
But yeah, so I mean, essentially a volume

60
00:02:49.669 --> 00:02:53.006
is a combination of panels

61
00:02:53.089 --> 00:02:57.302
that are sort of stacked together to make a wall.

62
00:02:57.719 --> 00:02:59.554
And then whether you have a

63
00:02:59.637 --> 00:03:01.681
ceiling LED panel in there

64
00:03:01.764 --> 00:03:04.642
and a side, a couple of sides

65
00:03:04.767 --> 00:03:06.728
or a reflection sort of panels as well

66
00:03:07.145 --> 00:03:10.857
that can combine to create, as we call it a volume.

67
00:03:11.941 --> 00:03:14.819
And you don't necessarily have to have

68
00:03:15.320 --> 00:03:17.030
digital assets in there.

69
00:03:17.989 --> 00:03:20.116
So that particular example, yes,

70
00:03:20.199 --> 00:03:22.577
those were all virtually created

71
00:03:23.328 --> 00:03:24.662
from our virtual art department.

72
00:03:25.580 --> 00:03:27.832
And so those were full 3D environments,

73
00:03:28.082 --> 00:03:31.336
but we've done many projects

74
00:03:31.419 --> 00:03:33.796
where we have live action plates that we filmed

75
00:03:34.005 --> 00:03:36.341
and then those are projected onto the panels.

76
00:03:36.424 --> 00:03:38.509
But essentially that's what's happening.

77
00:03:38.843 --> 00:03:43.598
We're projecting the digital environment

78
00:03:44.223 --> 00:03:47.185
onto the back, onto the LED panels.

79
00:03:47.769 --> 00:03:51.773
And we have sort of a 3D space recognition

80
00:03:52.065 --> 00:03:55.401
that we understand where everything should be

81
00:03:55.485 --> 00:03:58.780
in the volume in terms of its 3D positioning.

82
00:03:59.322 --> 00:04:03.368
And the idea is for us to translate that

83
00:04:03.451 --> 00:04:06.663
to the real world set that we've created as well.

84
00:04:06.746 --> 00:04:08.706
And then that needs to exist

85
00:04:08.790 --> 00:04:09.874
in our digital environment.

86
00:04:09.958 --> 00:04:13.628
So we have specific distances that correlate

87
00:04:13.878 --> 00:04:15.964
to distances to the panels.

88
00:04:16.506 --> 00:04:18.800
And essentially you're able to then get parallax

89
00:04:19.676 --> 00:04:22.512
purely because the camera that we have,

90
00:04:22.929 --> 00:04:24.889
which is obviously a production camera

91
00:04:25.223 --> 00:04:27.100
and can go into all the different sorts of cameras

92
00:04:27.183 --> 00:04:29.143
that we would use for virtual production,

93
00:04:29.227 --> 00:04:32.730
but that camera is being tracked in real time.

94
00:04:32.814 --> 00:04:36.401
So we have a bunch of motion capture cameras

95
00:04:36.484 --> 00:04:39.529
that we create our own little volume for.

96
00:04:40.154 --> 00:04:42.073
And those are constantly tracking

97
00:04:43.449 --> 00:04:47.412
the physical production camera in space.

98
00:04:47.495 --> 00:04:48.621
And then that is translated

99
00:04:48.705 --> 00:04:50.498
into our digital environment.

100
00:04:51.207 --> 00:04:54.335
And we have a digital replica camera

101
00:04:54.627 --> 00:04:55.545
in the digital environment.

102
00:04:56.129 --> 00:04:59.757
And those two essentially are matched together.

103
00:05:00.133 --> 00:05:02.885
So that's how when the practical camera moves,

104
00:05:03.011 --> 00:05:06.139
then the background is moved in relation to that.

105
00:05:06.681 --> 00:05:11.185
And it's field of view changes based on lensing.

106
00:05:11.269 --> 00:05:12.895
Well, once we adjust those,

107
00:05:12.979 --> 00:05:14.105
and so just the parallax

108
00:05:14.188 --> 00:05:15.064
shifting in the background,

109
00:05:15.148 --> 00:05:16.357
that's where that all sort of happens.

110
00:05:16.941 --> 00:05:17.608
But that would happen more

111
00:05:17.692 --> 00:05:18.735
for digital environments.

112
00:05:19.610 --> 00:05:20.903
But if it's a live action

113
00:05:20.987 --> 00:05:23.614
shoot that we've captured plates,

114
00:05:24.407 --> 00:05:25.575
it doesn't exist the same way.

115
00:05:25.825 --> 00:05:27.201
Of course, that then is not,

116
00:05:27.618 --> 00:05:29.037
it's not a digital environment anymore.

117
00:05:29.454 --> 00:05:31.289
So the parallax is still

118
00:05:31.372 --> 00:05:32.790
sort of there in some senses

119
00:05:32.874 --> 00:05:33.916
because we do a sort of a two

120
00:05:34.000 --> 00:05:37.754
and a half D sort of workflow.

121
00:05:38.254 --> 00:05:40.840
But in that sense, it's not quite as 3D

122
00:05:40.923 --> 00:05:42.925
as the 3D virtual ones.

123
00:05:43.760 --> 00:05:44.844
- Wow, okay.

124
00:05:44.927 --> 00:05:47.555
I think for me, it kind of comes together

125
00:05:47.680 --> 00:05:51.100
when you tell us how the camera itself

126
00:05:51.392 --> 00:05:52.977
and the motion capture in the camera

127
00:05:53.603 --> 00:05:55.438
is linked to a 3D camera.

128
00:05:55.980 --> 00:05:56.689
I think that's the bit

129
00:05:56.773 --> 00:05:58.232
that I didn't quite understand

130
00:05:58.483 --> 00:05:59.609
before our discussion,

131
00:06:00.735 --> 00:06:02.945
because it all seems quite magical to me.

132
00:06:03.571 --> 00:06:05.239
So the process is that you

133
00:06:05.323 --> 00:06:07.241
would get those environments

134
00:06:07.742 --> 00:06:10.453
digitized prior to the shoot, right?

135
00:06:10.661 --> 00:06:13.498
So if it's a fully 3D environment,

136
00:06:14.040 --> 00:06:16.417
that's quite a bit of work that's gone in

137
00:06:16.501 --> 00:06:19.170
prior to actually cameras rolling.

138
00:06:20.046 --> 00:06:20.880
Can you tell me a little bit

139
00:06:20.963 --> 00:06:22.381
about how you take an image?

140
00:06:23.007 --> 00:06:25.384
And it can be a 3D one or it can be a,

141
00:06:26.010 --> 00:06:27.637
something that you've sourced in the real world.

142
00:06:28.096 --> 00:06:30.431
How do you then get that into the volume?

143
00:06:32.391 --> 00:06:33.726
- Yeah, I mean, if we could start

144
00:06:33.810 --> 00:06:35.603
from sort of the 3D side of it,

145
00:06:36.354 --> 00:06:37.772
our virtual art department,

146
00:06:38.815 --> 00:06:40.566
obviously we've been getting a brief

147
00:06:40.775 --> 00:06:44.737
as to what that environment wants to be

148
00:06:44.821 --> 00:06:48.658
and how realistic it also needs to be.

149
00:06:48.741 --> 00:06:50.118
Because of course, it's subjective,

150
00:06:52.662 --> 00:06:57.041
working in an artistic sort of realm,

151
00:06:57.708 --> 00:06:59.377
there could be multiple sort of environments

152
00:06:59.752 --> 00:07:00.461
that you can make,

153
00:07:01.254 --> 00:07:04.048
but in certain, I mean, in most cases,

154
00:07:04.132 --> 00:07:07.135
it is realistic as we can make it.

155
00:07:08.261 --> 00:07:11.013
And then of course, we very

156
00:07:11.097 --> 00:07:14.392
much create the environments,

157
00:07:14.851 --> 00:07:17.395
sort of we like to Hollywood the environment.

158
00:07:17.645 --> 00:07:21.774
So in that sense, we build a full 360 environment,

159
00:07:22.650 --> 00:07:24.360
but we do Hollywood it out

160
00:07:24.443 --> 00:07:26.195
to a specific field of view,

161
00:07:26.529 --> 00:07:28.322
at least for a range of motion for the camera

162
00:07:28.406 --> 00:07:29.323
that we're expecting,

163
00:07:29.907 --> 00:07:31.284
so that we don't create too

164
00:07:31.367 --> 00:07:32.785
huge an environment because--

165
00:07:33.369 --> 00:07:35.580
- What does that mean, Hollywood it out?

166
00:07:35.830 --> 00:07:36.664
I've never heard that.

167
00:07:36.998 --> 00:07:40.668
- It's the term, you know, like back in the day,

168
00:07:42.712 --> 00:07:45.631
in production, you know, you're building a set to,

169
00:07:46.048 --> 00:07:49.552
you don't have to build in full environment,

170
00:07:49.886 --> 00:07:52.138
you know, the whole world if you're building sets.

171
00:07:52.430 --> 00:07:55.933
And so we consider it just Hollywooding it

172
00:07:56.100 --> 00:07:57.685
to the environment, I mean, to the camera.

173
00:07:58.686 --> 00:07:59.770
I don't know, it's a term

174
00:07:59.854 --> 00:08:02.231
that I picked up way back.

175
00:08:02.607 --> 00:08:03.024
So I don't know.

176
00:08:03.232 --> 00:08:03.649
- I love it.

177
00:08:05.109 --> 00:08:07.987
- So anyway, we ultimately build the environment

178
00:08:08.237 --> 00:08:11.699
to the camera's, you know, like the range of motion

179
00:08:11.782 --> 00:08:13.826
that we're expecting it to sort of go along

180
00:08:14.243 --> 00:08:17.288
and we're able to obviously deduce that

181
00:08:17.371 --> 00:08:19.207
by doing previs or techvis

182
00:08:20.458 --> 00:08:22.543
and also obviously, you know, creating boards

183
00:08:22.710 --> 00:08:24.128
and we have a, you know, we

184
00:08:24.212 --> 00:08:26.130
get quite far into our previs

185
00:08:26.672 --> 00:08:29.592
prior to, you know,

186
00:08:29.675 --> 00:08:31.052
projecting it up onto the screen.

187
00:08:31.135 --> 00:08:33.554
So we already sort of have a pretty clear idea

188
00:08:33.763 --> 00:08:36.474
of how much of the environment's gonna be seen

189
00:08:37.475 --> 00:08:38.517
during the production.

190
00:08:38.601 --> 00:08:40.811
I mean, that doesn't mean that we're not giving

191
00:08:40.978 --> 00:08:42.605
the director as much

192
00:08:42.688 --> 00:08:45.316
opportunity to change the camera

193
00:08:45.691 --> 00:08:47.818
as, you know, they generally would like

194
00:08:48.194 --> 00:08:50.404
and also the DP for that matter, obviously.

195
00:08:51.239 --> 00:08:53.366
So, but we certainly give

196
00:08:53.449 --> 00:08:54.533
them a bit of a constraint

197
00:08:56.285 --> 00:08:59.121
and we've already built it to sort of, you know,

198
00:08:59.205 --> 00:09:01.791
we're expecting, for instance, you know,

199
00:09:01.874 --> 00:09:02.833
we wanna see behind us,

200
00:09:02.917 --> 00:09:04.210
we're gonna be shooting reverses,

201
00:09:04.293 --> 00:09:06.170
we're gonna be shooting sort of POVs

202
00:09:06.254 --> 00:09:08.297
and sort of wide vistas or whatever.

203
00:09:08.381 --> 00:09:11.050
And so we'll build it to those constraints.

204
00:09:12.760 --> 00:09:15.346
But again, there's many sort of factors

205
00:09:15.554 --> 00:09:17.723
with regards to optimization and things like that,

206
00:09:17.807 --> 00:09:20.601
which the reason why we don't build everything out

207
00:09:21.018 --> 00:09:23.396
is purely because of optimization

208
00:09:23.688 --> 00:09:26.190
and being able to run at the required frame rate

209
00:09:27.024 --> 00:09:28.276
that we're gonna wanna roll at.

210
00:09:29.527 --> 00:09:32.822
And so again - go down the rabbit hole.

211
00:09:33.531 --> 00:09:34.156
We could sort of get into

212
00:09:34.240 --> 00:09:35.741
that later if you want to, but--

213
00:09:35.992 --> 00:09:36.867
- Yeah, sure, I'd love to.

214
00:09:37.410 --> 00:09:40.413
- But ideally, you know, the idea would be

215
00:09:41.205 --> 00:09:43.332
once we've pre-vised out those environments

216
00:09:43.582 --> 00:09:44.208
and, you know, we can have

217
00:09:44.292 --> 00:09:45.626
multiple environments in a day

218
00:09:45.710 --> 00:09:46.502
that we wanna shoot,

219
00:09:47.712 --> 00:09:49.880
we'll then have a

220
00:09:49.964 --> 00:09:51.716
pre-light at least with the camera

221
00:09:51.966 --> 00:09:54.427
and we'll project those onto the wall.

222
00:09:54.510 --> 00:09:56.345
Now, when I say project those onto the wall,

223
00:09:57.138 --> 00:09:59.265
we, within Unreal Engine, for instance,

224
00:10:00.182 --> 00:10:01.142
we have something called

225
00:10:01.225 --> 00:10:02.893
nDisplay, which is a plugin,

226
00:10:04.228 --> 00:10:07.231
which essentially is a projection plugin

227
00:10:07.356 --> 00:10:10.401
where you create planers within the scene

228
00:10:10.651 --> 00:10:12.570
which represent the panels,

229
00:10:13.112 --> 00:10:15.656
represent the position of the panels.

230
00:10:17.241 --> 00:10:18.951
And that nDisplay node

231
00:10:19.035 --> 00:10:20.911
essentially is what we call it,

232
00:10:21.037 --> 00:10:25.333
that we then have the environment all around

233
00:10:25.499 --> 00:10:26.542
and then with the camera,

234
00:10:26.625 --> 00:10:29.754
it projects onto a very specific position

235
00:10:30.463 --> 00:10:33.716
of where the panels should exist in space.

236
00:10:34.842 --> 00:10:37.762
And that becomes, that's kind

237
00:10:37.845 --> 00:10:39.555
of where the trick really lies

238
00:10:39.639 --> 00:10:42.475
in understanding the real distance

239
00:10:42.683 --> 00:10:45.686
that things relate to the real world,

240
00:10:45.770 --> 00:10:47.355
from the digital to the real world,

241
00:10:47.438 --> 00:10:50.900
which is why we tend to LIDAR scan our sets.

242
00:10:51.150 --> 00:10:54.070
So we will go ahead and do a full LIDAR scan

243
00:10:54.236 --> 00:10:55.321
of the set environment,

244
00:10:55.404 --> 00:10:56.739
whether it's just in a studio,

245
00:10:57.406 --> 00:10:59.283
where the panels in position,

246
00:10:59.867 --> 00:11:01.369
where they are in position,

247
00:11:01.452 --> 00:11:03.037
where our camera is expecting,

248
00:11:03.287 --> 00:11:04.663
we create something called a home base

249
00:11:04.747 --> 00:11:07.124
so we know where the camera wants to always be

250
00:11:08.084 --> 00:11:09.210
at the beginning of a shoot.

251
00:11:09.919 --> 00:11:12.171
And so we then do a LIDAR scan

252
00:11:13.005 --> 00:11:15.633
and from that point, we filter

253
00:11:15.716 --> 00:11:16.842
that back into Unreal Engine.

254
00:11:17.093 --> 00:11:19.387
So we actually have real world scale positioning

255
00:11:19.470 --> 00:11:20.221
of everything.

256
00:11:20.805 --> 00:11:21.680
And once we've done that,

257
00:11:21.764 --> 00:11:24.725
we'll then align our nDisplay node in Unreal

258
00:11:25.059 --> 00:11:26.894
to where the panels are, and

259
00:11:26.977 --> 00:11:28.312
then where that set exists.

260
00:11:29.313 --> 00:11:31.357
And then we can work out distances

261
00:11:32.233 --> 00:11:35.528
from practical positioning.

262
00:11:36.028 --> 00:11:38.280
Like obviously, for instance, if we had Artabar

263
00:11:38.364 --> 00:11:41.492
and bump in some foreground elements or whatever,

264
00:11:42.451 --> 00:11:43.828
those elements need to exist

265
00:11:43.911 --> 00:11:45.830
in the space digitally as well.

266
00:11:46.664 --> 00:11:47.540
It's what'll end up happening

267
00:11:47.623 --> 00:11:50.292
as things will overlap on the screen.

268
00:11:50.376 --> 00:11:51.752
And then something on the background

269
00:11:51.836 --> 00:11:53.963
is supposed to be in front, et cetera, et cetera.

270
00:11:54.046 --> 00:11:55.256
And so you need to be very

271
00:11:55.339 --> 00:11:58.843
clear about where the opening is

272
00:11:59.009 --> 00:12:01.721
and how you align that.

273
00:12:01.804 --> 00:12:02.304
And it's all about

274
00:12:02.388 --> 00:12:04.849
calculation, real world calculation.

275
00:12:05.433 --> 00:12:07.184
And then that needs to be translated

276
00:12:07.351 --> 00:12:09.353
into the digital space and vice versa.

277
00:12:09.770 --> 00:12:11.522
- Is that why it's called the volume?

278
00:12:11.856 --> 00:12:13.732
Because it's all about the space,

279
00:12:14.316 --> 00:12:17.528
the actual depth and distance between things?

280
00:12:17.611 --> 00:12:19.196
Is that something that gets

281
00:12:19.280 --> 00:12:21.115
its name or is that some other?

282
00:12:21.323 --> 00:12:23.033
- For me, I think volume

283
00:12:23.117 --> 00:12:24.535
because it's a volume of light.

284
00:12:25.202 --> 00:12:27.246
So when you have multiple panels

285
00:12:27.455 --> 00:12:28.456
and generally it sort of

286
00:12:28.539 --> 00:12:32.168
encompasses you as the environment

287
00:12:32.376 --> 00:12:33.127
and you're creating

288
00:12:33.210 --> 00:12:35.254
essentially an exterior environment

289
00:12:35.337 --> 00:12:38.132
inside an interior environment,

290
00:12:38.215 --> 00:12:39.383
that becomes a volume of

291
00:12:39.467 --> 00:12:40.801
light, its own environment.

292
00:12:41.093 --> 00:12:42.219
And so that's how I...

293
00:12:42.887 --> 00:12:44.096
That's where I'm getting it from.

294
00:12:44.180 --> 00:12:45.890
I still like what you're saying though.

295
00:12:45.973 --> 00:12:46.640
I don't mind that.

296
00:12:46.891 --> 00:12:47.725
I can think of it that way.

297
00:12:48.017 --> 00:12:49.935
- I've been trying to just,

298
00:12:50.019 --> 00:12:52.188
as I've thought about speaking with you about this,

299
00:12:52.271 --> 00:12:55.107
trying to imagine where the term came from.

300
00:12:55.191 --> 00:12:56.775
So I'm looking for it everywhere.

301
00:12:57.860 --> 00:13:01.697
It sounds so futuristic, but that is fascinating.

302
00:13:02.031 --> 00:13:05.993
So you've kind of got these projected elements

303
00:13:06.285 --> 00:13:07.912
as well as physical elements.

304
00:13:07.995 --> 00:13:09.038
And that was certainly the case

305
00:13:09.121 --> 00:13:10.915
in what we worked on together.

306
00:13:10.998 --> 00:13:13.334
There was a foreground that was physically bumped

307
00:13:13.417 --> 00:13:14.335
into the space.

308
00:13:15.127 --> 00:13:17.671
And then on top of those two

309
00:13:17.755 --> 00:13:19.381
things just existing together,

310
00:13:19.465 --> 00:13:21.425
you've also done a LiDAR scan

311
00:13:21.509 --> 00:13:23.969
of those elements in the space

312
00:13:24.136 --> 00:13:27.890
and you fed them back into the digital simulation

313
00:13:28.432 --> 00:13:29.725
of the environment.

314
00:13:30.059 --> 00:13:31.810
- Yes, that's great.

315
00:13:32.061 --> 00:13:33.187
- That must be the trickiest

316
00:13:33.270 --> 00:13:34.855
thing is getting that scene

317
00:13:34.980 --> 00:13:37.274
between the digital and the

318
00:13:37.358 --> 00:13:40.528
real kind of working seamlessly.

319
00:13:41.612 --> 00:13:43.948
- Yeah, that's definitely the trick.

320
00:13:45.199 --> 00:13:47.451
And we tend to build a lot

321
00:13:47.535 --> 00:13:50.287
of, in our digital world,

322
00:13:50.829 --> 00:13:53.582
and depending on what sort of practical set

323
00:13:53.707 --> 00:13:55.167
that we have available to us,

324
00:13:55.376 --> 00:13:57.419
we tend to build into our

325
00:13:57.503 --> 00:13:58.587
digital set little tricks.

326
00:13:59.171 --> 00:14:01.924
Like, for instance, in that Sydney scene,

327
00:14:02.174 --> 00:14:03.425
I know it's hard to talk

328
00:14:03.509 --> 00:14:04.510
about it when you can't see it,

329
00:14:04.593 --> 00:14:06.178
but we built in a

330
00:14:06.262 --> 00:14:07.763
pavement, an edge, a pavement edge

331
00:14:07.846 --> 00:14:08.889
where we're expecting the

332
00:14:08.973 --> 00:14:10.516
edge of the panels to exist.

333
00:14:10.766 --> 00:14:12.810
And so we will build in where

334
00:14:12.893 --> 00:14:14.603
that edge of the panel wants,

335
00:14:15.062 --> 00:14:16.647
sorry, the edge of the pavement,

336
00:14:17.606 --> 00:14:19.567
which happened to be in the digital set.

337
00:14:20.150 --> 00:14:21.235
And then we decided, okay,

338
00:14:21.318 --> 00:14:22.528
we'll replace that pavement

339
00:14:22.861 --> 00:14:24.029
with the real pavement

340
00:14:24.113 --> 00:14:26.699
edge, so that becomes practical.

341
00:14:26.782 --> 00:14:29.326
And then we align that up to our digital pavement

342
00:14:29.577 --> 00:14:30.828
and we delete our digital pavement,

343
00:14:30.911 --> 00:14:33.664
and then that becomes the new pavement position.

344
00:14:34.582 --> 00:14:35.291
And so when the camera

345
00:14:35.374 --> 00:14:38.252
moves, it should stick in a line.

346
00:14:38.752 --> 00:14:41.213
I mean, I think there's just,

347
00:14:42.256 --> 00:14:44.550
in those instances, you're then locked into

348
00:14:44.633 --> 00:14:49.346
a distance to the panels.

349
00:14:49.638 --> 00:14:54.560
And so I think one of the main advantages,

350
00:14:55.060 --> 00:14:58.397
I think, in volume and in sort of quick shooting,

351
00:14:58.480 --> 00:14:59.398
if you wanna do that,

352
00:15:00.399 --> 00:15:04.153
is to be able to cheat the

353
00:15:04.236 --> 00:15:06.113
distance backwards and forwards

354
00:15:07.239 --> 00:15:09.491
and move things around digitally quite quickly,

355
00:15:09.992 --> 00:15:11.660
and to be able to, so in other words,

356
00:15:11.744 --> 00:15:13.495
if you build something into your set and say,

357
00:15:13.579 --> 00:15:16.415
now that is now physically connected to the set,

358
00:15:17.416 --> 00:15:18.834
and it's very heavy, for instance,

359
00:15:19.793 --> 00:15:22.254
when you start spinning the environment around,

360
00:15:22.671 --> 00:15:23.922
and I guess we haven't

361
00:15:24.006 --> 00:15:25.299
really kind of gone into this,

362
00:15:25.382 --> 00:15:28.260
but for instance, if we wanted to do a reverse,

363
00:15:28.802 --> 00:15:30.137
there's no panels behind us,

364
00:15:30.220 --> 00:15:33.140
so we spin the environment around digitally,

365
00:15:33.724 --> 00:15:35.934
and then suddenly that's the reverse of the shot

366
00:15:36.018 --> 00:15:38.020
and keep the camera in the same position,

367
00:15:38.729 --> 00:15:41.357
so that you're kind of cheating it in that sense.

368
00:15:41.940 --> 00:15:45.235
But if you've now built a practical floor

369
00:15:45.819 --> 00:15:48.238
which wants to connect to your panels,

370
00:15:48.822 --> 00:15:50.199
that's incorrect, right?

371
00:15:50.282 --> 00:15:51.659
So then that would need to move too,

372
00:15:51.742 --> 00:15:53.035
so you'd have to flip that around,

373
00:15:53.118 --> 00:15:56.121
or you don't always wanna be doing wide shots

374
00:15:57.706 --> 00:15:59.333
that need to obviously

375
00:15:59.416 --> 00:16:01.377
align with each other in the cut

376
00:16:01.710 --> 00:16:03.796
and things like that, so you just need to be quite,

377
00:16:05.005 --> 00:16:06.507
you don't wanna build too much infrastructure

378
00:16:06.715 --> 00:16:07.591
into your foreground,

379
00:16:08.133 --> 00:16:09.968
which doesn't have the ability to move,

380
00:16:10.052 --> 00:16:11.053
I guess is what I'm saying.

381
00:16:11.804 --> 00:16:13.597
- Yeah, unless you have it on a turntable.

382
00:16:14.014 --> 00:16:14.890
- That's right, yeah.

383
00:16:14.973 --> 00:16:17.309
And a lot of that is the case,

384
00:16:17.601 --> 00:16:18.936
and people do tend to do that

385
00:16:19.019 --> 00:16:20.354
when it comes to much

386
00:16:20.437 --> 00:16:23.023
larger virtual production shoots,

387
00:16:23.816 --> 00:16:25.359
which is a massive advantage,

388
00:16:25.651 --> 00:16:29.697
and if there's that sort of budget thrown at it,

389
00:16:29.780 --> 00:16:31.949
then I will say yes and put my hand up

390
00:16:32.032 --> 00:16:33.701
and say let's do it every single time,

391
00:16:34.076 --> 00:16:35.369
because it just helps out,

392
00:16:35.452 --> 00:16:37.246
and helps the production designer out too,

393
00:16:37.663 --> 00:16:40.207
and of course, we also then,

394
00:16:40.457 --> 00:16:42.292
every time there's new props or anything like that,

395
00:16:42.376 --> 00:16:46.255
we try to scan those as quickly as we can,

396
00:16:46.338 --> 00:16:47.047
but those are just sort of

397
00:16:47.131 --> 00:16:48.632
photogrammetry style scans,

398
00:16:48.716 --> 00:16:49.925
not full LiDAR scans,

399
00:16:50.175 --> 00:16:51.301
and then we'll try and,

400
00:16:52.177 --> 00:16:53.470
because we would, like I said,

401
00:16:53.554 --> 00:16:55.097
when we spin the environment around,

402
00:16:55.180 --> 00:16:57.808
there could be elements that were created

403
00:16:58.642 --> 00:17:00.728
as the bump in practical set,

404
00:17:00.811 --> 00:17:04.815
those then now suddenly exist in the digital world,

405
00:17:04.898 --> 00:17:06.984
you'll actually see them through the screens,

406
00:17:07.735 --> 00:17:09.945
and if you haven't created those exact props,

407
00:17:10.237 --> 00:17:12.531
then they won't be able to be extended out

408
00:17:12.614 --> 00:17:13.699
into the digital world,

409
00:17:13.782 --> 00:17:17.161
so that's why we also scan and recreate all those,

410
00:17:17.828 --> 00:17:18.912
and match them if we can.

411
00:17:19.121 --> 00:17:20.622
- That is absolutely full on,

412
00:17:20.706 --> 00:17:22.833
like the level of precision that you would need.

413
00:17:22.916 --> 00:17:24.501
I know that in the color grading world,

414
00:17:24.668 --> 00:17:25.878
we're typically not working

415
00:17:25.961 --> 00:17:27.463
with that level of precision

416
00:17:27.671 --> 00:17:29.381
in terms of getting

417
00:17:29.465 --> 00:17:31.967
everything measured and accurate,

418
00:17:32.217 --> 00:17:33.343
it's sort of the approach is

419
00:17:33.427 --> 00:17:36.013
more about getting a feeling,

420
00:17:36.638 --> 00:17:39.475
and instead of doing like really tight shapes,

421
00:17:39.683 --> 00:17:41.769
we're typically doing softer shapes,

422
00:17:42.186 --> 00:17:43.312
and sort of being a little

423
00:17:43.395 --> 00:17:45.856
bit more gentle in our approach

424
00:17:45.939 --> 00:17:47.066
to manipulating the image

425
00:17:47.149 --> 00:17:48.442
and trying to do a bit less,

426
00:17:48.525 --> 00:17:49.651
but this is sort of the opposite

427
00:17:49.735 --> 00:17:51.153
where you're doing everything,

428
00:17:51.487 --> 00:17:52.905
you're generating everything.

429
00:17:54.031 --> 00:17:55.741
I imagine you get into the uncanny valley

430
00:17:56.074 --> 00:17:57.993
pretty easily doing that sort of thing.

431
00:17:58.827 --> 00:17:59.661
- Yeah, I mean, and look,

432
00:17:59.953 --> 00:18:02.164
and having sort of broad power windows,

433
00:18:02.247 --> 00:18:03.916
I mean, that stuff is great,

434
00:18:03.999 --> 00:18:06.710
and it certainly lends itself to what it is

435
00:18:06.794 --> 00:18:07.544
that you guys are doing,

436
00:18:07.628 --> 00:18:09.213
when all you have is a 2D image

437
00:18:09.546 --> 00:18:11.089
to manipulate,

438
00:18:11.298 --> 00:18:12.674
but I think for us, it's just,

439
00:18:12.966 --> 00:18:16.178
it can very easily go wrong,

440
00:18:16.804 --> 00:18:17.805
and so we're just always

441
00:18:17.888 --> 00:18:20.057
trying to stay physically accurate

442
00:18:20.557 --> 00:18:23.060
as much as we can before the cheating,

443
00:18:23.143 --> 00:18:24.770
and even though it is a full cheat anyway,

444
00:18:25.145 --> 00:18:28.982
the only way for that to be a reality

445
00:18:29.233 --> 00:18:32.569
is for us to be very strict and constrained

446
00:18:32.778 --> 00:18:35.572
to the physical aspects of what's going on,

447
00:18:35.656 --> 00:18:37.491
and so that needs to sort

448
00:18:37.574 --> 00:18:38.867
of loop back the whole time.

449
00:18:39.952 --> 00:18:42.496
And yeah, I mean, once you create a pipeline

450
00:18:42.579 --> 00:18:44.289
for what it is that you're doing,

451
00:18:44.373 --> 00:18:47.209
these sorts of things don't

452
00:18:47.292 --> 00:18:48.794
have to always be thought about

453
00:18:48.877 --> 00:18:50.504
as intricately.

454
00:18:50.754 --> 00:18:53.215
I mean, ultimately, I will be the one

455
00:18:53.298 --> 00:18:55.217
that needs to keep an eye on that sort of stuff,

456
00:18:55.300 --> 00:18:57.970
and if it goes awry,

457
00:18:58.512 --> 00:19:00.597
because everybody within our brain bar,

458
00:19:00.681 --> 00:19:02.099
which I haven't really sort of discussed,

459
00:19:02.182 --> 00:19:04.393
but they have individual jobs.

460
00:19:04.852 --> 00:19:08.230
- Yeah, but it has now become a universal term

461
00:19:08.313 --> 00:19:09.231
for some reason, so

462
00:19:09.314 --> 00:19:11.275
everybody uses it as the brain bar.

463
00:19:11.358 --> 00:19:12.150
They call it the brain bar.

464
00:19:12.442 --> 00:19:13.360
- That's cool.

465
00:19:13.694 --> 00:19:15.445
So I suppose I'll just do

466
00:19:15.529 --> 00:19:16.321
a little bit of background

467
00:19:16.488 --> 00:19:17.990
about the actual studios themselves.

468
00:19:18.240 --> 00:19:23.161
So Steelbridge are quite new facility in Brisbane,

469
00:19:23.871 --> 00:19:27.416
and it's been really popular,

470
00:19:27.875 --> 00:19:30.627
and Australia's actually been quite a large adopter

471
00:19:30.878 --> 00:19:32.838
of virtual production worldwide.

472
00:19:34.464 --> 00:19:35.257
So we've got some

473
00:19:35.340 --> 00:19:37.593
facilities in Melbourne as well at,

474
00:19:38.844 --> 00:19:40.095
jeez, I can't remember the name, but--

475
00:19:40.178 --> 00:19:40.846
- Nance Studios.

476
00:19:41.388 --> 00:19:42.347
- That's the one. - Nance Studios

477
00:19:42.681 --> 00:19:44.433
in the Docklands, yeah.

478
00:19:44.516 --> 00:19:45.601
- That's right, yes.

479
00:19:46.143 --> 00:19:47.936
And VCA have put in some

480
00:19:48.020 --> 00:19:50.939
virtual production facilities there

481
00:19:51.315 --> 00:19:54.902
in Melbourne as well, but Brisbane has Steelbridge,

482
00:19:55.319 --> 00:19:57.779
and then there's some volumes available

483
00:19:58.113 --> 00:20:00.157
on the Gold Coast in the studios there as well.

484
00:20:01.491 --> 00:20:02.910
But Steelbridge are probably

485
00:20:02.993 --> 00:20:04.828
the most active in the region

486
00:20:04.953 --> 00:20:07.831
in terms of doing a lot of work.

487
00:20:08.832 --> 00:20:10.959
They've got their pipeline pretty well down.

488
00:20:11.335 --> 00:20:13.211
They've done quite a lot of commercials.

489
00:20:13.629 --> 00:20:15.130
I'm not so sure about Longform,

490
00:20:15.881 --> 00:20:17.382
but I'll have to ask about that,

491
00:20:17.466 --> 00:20:18.592
what you guys have been involved in,

492
00:20:18.675 --> 00:20:19.885
if anything, in that space.

493
00:20:20.177 --> 00:20:21.887
But definitely in terms of commercials,

494
00:20:21.970 --> 00:20:22.763
they would be the ones that

495
00:20:22.846 --> 00:20:25.223
you would go to in our area.

496
00:20:25.891 --> 00:20:27.309
And some of the advantages of

497
00:20:27.392 --> 00:20:28.810
it in the work that I've seen

498
00:20:29.353 --> 00:20:31.563
is we worked on a car commercial together,

499
00:20:32.147 --> 00:20:34.274
and the environments were quite varied

500
00:20:34.441 --> 00:20:36.944
because our landscape in Australia is quite varied.

501
00:20:37.027 --> 00:20:40.489
So they were able to do sort of red desert dirt

502
00:20:40.906 --> 00:20:44.368
and city driving and coastal driving,

503
00:20:45.035 --> 00:20:47.371
all of these environments in one day.

504
00:20:48.121 --> 00:20:52.084
So obviously the practicality of that is great.

505
00:20:52.167 --> 00:20:55.379
You only have to take the cars into one space.

506
00:20:55.462 --> 00:20:56.964
You don't have to freight them around,

507
00:20:58.382 --> 00:21:00.884
and you can go to a lot of places quite quickly.

508
00:21:01.385 --> 00:21:03.512
So, although you're doing

509
00:21:03.595 --> 00:21:05.055
quite a lot more in pre-production,

510
00:21:05.305 --> 00:21:07.057
I think in terms of production,

511
00:21:07.224 --> 00:21:08.850
everything's a bit more sewn together.

512
00:21:10.394 --> 00:21:11.812
But yeah, so in Brisbane,

513
00:21:11.895 --> 00:21:13.438
they've got this great facility

514
00:21:13.814 --> 00:21:16.984
and the pipeline's pretty well sorted out now

515
00:21:17.109 --> 00:21:19.444
and some pretty seasoned staff on board,

516
00:21:20.153 --> 00:21:20.988
including the friend who

517
00:21:21.071 --> 00:21:22.364
we're talking to right now.

518
00:21:22.447 --> 00:21:22.614
- Yes.

519
00:21:23.657 --> 00:21:25.283
Yeah, look, I mean, Steelbridge has been

520
00:21:26.326 --> 00:21:28.537
certainly pushing really hard,

521
00:21:29.121 --> 00:21:29.997
and I think it's been,

522
00:21:31.957 --> 00:21:33.041
I've been very fortunate to

523
00:21:33.125 --> 00:21:34.918
be able to work with them.

524
00:21:35.168 --> 00:21:38.588
I mean, initially, it was a bit of a brainchild

525
00:21:39.339 --> 00:21:41.299
between myself and Colin,

526
00:21:41.383 --> 00:21:44.428
who's the founder of AltVFX and owner.

527
00:21:45.595 --> 00:21:46.722
And, you know--

528
00:21:46.805 --> 00:21:47.180
- That's Colin Renshaw.

529
00:21:47.431 --> 00:21:48.515
- That's Colin Renshaw, yeah.

530
00:21:49.850 --> 00:21:51.143
And essentially,

531
00:21:54.146 --> 00:21:55.605
obviously the new tech was coming in.

532
00:21:55.689 --> 00:21:57.024
This was quite a few years ago now,

533
00:21:57.107 --> 00:22:01.028
but it was just sort of something that we knew

534
00:22:01.111 --> 00:22:02.487
it was gonna become a thing.

535
00:22:04.072 --> 00:22:08.577
And at the same time for Alt Studios

536
00:22:09.494 --> 00:22:13.707
AltVFX, sorry, they were pushing their animation,

537
00:22:13.790 --> 00:22:14.791
which is obviously quite

538
00:22:14.875 --> 00:22:17.085
integral to the work that they do,

539
00:22:17.169 --> 00:22:18.587
a lot of creature work that they do.

540
00:22:18.670 --> 00:22:22.299
And so motion capture was certainly a thing

541
00:22:22.841 --> 00:22:23.633
that they used,

542
00:22:23.925 --> 00:22:26.470
and they used a lot of XN suits.

543
00:22:26.553 --> 00:22:27.054
- They were very good at motion.

544
00:22:27.804 --> 00:22:28.722
Very, very good at that.

545
00:22:28.805 --> 00:22:29.931
I remember doing some grades,

546
00:22:30.098 --> 00:22:32.726
and occasionally I'd see people walking past

547
00:22:32.809 --> 00:22:33.518
in those suits.

548
00:22:33.852 --> 00:22:33.977
- Yeah.

549
00:22:34.061 --> 00:22:35.937
- So those suits are sort of like a,

550
00:22:36.354 --> 00:22:38.857
almost like a wetsuit with dots all over them.

551
00:22:38.940 --> 00:22:39.733
- That's right, yeah.

552
00:22:39.816 --> 00:22:41.485
- And then as the character moves around,

553
00:22:41.568 --> 00:22:43.028
those points are captured,

554
00:22:43.195 --> 00:22:46.031
and then they can be transposed onto a 3D model.

555
00:22:46.365 --> 00:22:46.990
Like you could see a

556
00:22:47.074 --> 00:22:48.575
dancing polar bear or something,

557
00:22:48.658 --> 00:22:49.159
but it's just,

558
00:22:49.618 --> 00:22:51.328
I love seeing that behind the scenes thing

559
00:22:51.411 --> 00:22:52.913
of the person doing the silly dance,

560
00:22:53.121 --> 00:22:54.623
and then the polar bear does it behind them.

561
00:22:54.706 --> 00:22:56.375
I think that's always a cool party trick.

562
00:22:56.750 --> 00:22:58.001
- Exactly, yeah.

563
00:22:58.251 --> 00:23:01.421
And so I think that Alt just wanted to push

564
00:23:01.880 --> 00:23:04.841
their motion capture facility,

565
00:23:05.217 --> 00:23:10.263
and actually buy and look into the Vicon cameras,

566
00:23:10.347 --> 00:23:12.140
which are the very high end

567
00:23:12.224 --> 00:23:14.476
motion capture cameras for,

568
00:23:15.060 --> 00:23:15.811
to again, push their

569
00:23:15.894 --> 00:23:17.437
pipeline, their animation pipeline,

570
00:23:17.521 --> 00:23:18.563
but at the same time,

571
00:23:19.856 --> 00:23:24.152
find a permanent place for that as well.

572
00:23:24.236 --> 00:23:24.986
And I think that's where

573
00:23:25.070 --> 00:23:27.030
this sort of started growing

574
00:23:27.197 --> 00:23:28.073
from there, and of course,

575
00:23:28.156 --> 00:23:30.700
with having the motion capture facility,

576
00:23:32.494 --> 00:23:33.286
where you don't,

577
00:23:33.370 --> 00:23:34.496
it's not an X-Sense one,

578
00:23:34.579 --> 00:23:36.832
but it's more a camera driven one.

579
00:23:37.374 --> 00:23:39.459
You can have multiple people wearing suits

580
00:23:39.543 --> 00:23:40.669
at the same time interacting.

581
00:23:41.002 --> 00:23:43.171
So, you know, there's certainly a lot of fun things

582
00:23:43.255 --> 00:23:44.548
going on here, there,

583
00:23:44.631 --> 00:23:46.466
and we do a lot of that here at the moment as well.

584
00:23:46.550 --> 00:23:48.176
So it's fantastic to be

585
00:23:48.260 --> 00:23:51.680
able to push both parts of it,

586
00:23:51.763 --> 00:23:53.098
and it's quite integral in the

587
00:23:53.181 --> 00:23:54.349
visual effects side of things.

588
00:23:55.392 --> 00:23:57.435
Of course, I'm a visual effects supervisor too,

589
00:23:57.519 --> 00:23:59.688
so it's a very important

590
00:23:59.771 --> 00:24:02.899
aspect of visual effects too.

591
00:24:02.983 --> 00:24:07.154
So that in itself drove the volume,

592
00:24:07.487 --> 00:24:08.488
motion capture volume,

593
00:24:08.572 --> 00:24:10.157
and in that sense, you know,

594
00:24:10.240 --> 00:24:12.367
we're tracking a camera at the same time.

595
00:24:12.909 --> 00:24:14.828
And so we decided that, you know,

596
00:24:15.078 --> 00:24:18.915
with that technology working out and, you know,

597
00:24:18.999 --> 00:24:21.001
moving into the real time

598
00:24:21.084 --> 00:24:22.627
sort of software like Unreal,

599
00:24:23.837 --> 00:24:26.715
those obviously go hand in hand.

600
00:24:27.340 --> 00:24:29.926
And so we just decided to push

601
00:24:30.218 --> 00:24:32.721
onto the virtual production side of things as well.

602
00:24:33.430 --> 00:24:35.390
And while both those streams

603
00:24:35.473 --> 00:24:37.017
continue the motion capture,

604
00:24:37.100 --> 00:24:38.435
and then also the virtual production,

605
00:24:38.518 --> 00:24:40.020
and then it's developed into

606
00:24:40.103 --> 00:24:41.479
where Steelbridge, you know,

607
00:24:41.563 --> 00:24:44.399
creates a lot of virtual production commercials.

608
00:24:45.108 --> 00:24:48.195
And yes, there has been some long form inquiries,

609
00:24:48.278 --> 00:24:50.071
and we talk about a lot of

610
00:24:50.155 --> 00:24:54.117
long form projects with producers.

611
00:24:54.326 --> 00:24:56.328
And I think, and again, look,

612
00:24:56.411 --> 00:24:59.706
it's always gonna be something that producers

613
00:24:59.998 --> 00:25:02.500
and or production people, you know,

614
00:25:02.584 --> 00:25:04.461
it's a learning process for everybody.

615
00:25:04.544 --> 00:25:06.213
And I think it's hard

616
00:25:06.296 --> 00:25:08.256
sometimes for people to commit

617
00:25:09.466 --> 00:25:12.427
to a project, a virtual production project,

618
00:25:13.428 --> 00:25:14.930
especially if it's a long form project,

619
00:25:15.013 --> 00:25:16.264
and it's just a really big

620
00:25:16.348 --> 00:25:17.515
part of what they're doing.

621
00:25:17.849 --> 00:25:21.895
And there's so many things that can go wrong

622
00:25:21.978 --> 00:25:23.521
that they're not in control of anymore.

623
00:25:23.772 --> 00:25:25.482
And I think it's certainly

624
00:25:25.565 --> 00:25:28.276
hard to get people on board,

625
00:25:28.360 --> 00:25:30.528
but I don't think that it's far off.

626
00:25:30.737 --> 00:25:33.281
And we're talking to a lot

627
00:25:33.365 --> 00:25:35.659
of people is all I can say.

628
00:25:35.742 --> 00:25:38.245
- Do you think that's why, you know,

629
00:25:38.328 --> 00:25:40.080
the kinds of things that have been shot

630
00:25:40.163 --> 00:25:42.040
in the long form space in volumes,

631
00:25:42.666 --> 00:25:44.793
like Mandalorian and the

632
00:25:44.876 --> 00:25:46.127
Batman and things like that,

633
00:25:46.211 --> 00:25:47.671
that have used them quite a lot,

634
00:25:48.171 --> 00:25:49.881
they've had really big

635
00:25:49.965 --> 00:25:52.717
visual effects studios behind them

636
00:25:53.051 --> 00:25:55.762
that are actually responsible for the production.

637
00:25:56.054 --> 00:25:57.389
So in the case of the Mandalorian,

638
00:25:57.472 --> 00:25:58.974
it would be people who

639
00:25:59.057 --> 00:26:00.809
intimately understand the technology,

640
00:26:01.184 --> 00:26:02.852
who are kind of funding

641
00:26:02.936 --> 00:26:05.105
and controlling the decision making process,

642
00:26:06.147 --> 00:26:08.984
as opposed to more traditional productions,

643
00:26:10.777 --> 00:26:12.487
going out on that limb and

644
00:26:12.570 --> 00:26:14.489
having to have that leap of faith

645
00:26:14.656 --> 00:26:17.409
and trust in someone else's,

646
00:26:17.492 --> 00:26:19.703
like obviously amazing abilities,

647
00:26:19.786 --> 00:26:21.079
but just, it can be

648
00:26:21.162 --> 00:26:22.706
difficult when you don't understand

649
00:26:23.707 --> 00:26:25.458
what's involved, right?

650
00:26:25.917 --> 00:26:26.835
- That's very true.

651
00:26:27.294 --> 00:26:29.212
And also, and this is certainly something

652
00:26:29.296 --> 00:26:30.463
that Colin mentions a lot,

653
00:26:30.547 --> 00:26:31.798
and it's very true is that

654
00:26:31.881 --> 00:26:34.217
if you have a post-production

655
00:26:35.218 --> 00:26:36.886
facility that's behind a

656
00:26:36.970 --> 00:26:38.388
virtual production facility,

657
00:26:39.222 --> 00:26:41.391
you have the ability to fall

658
00:26:41.474 --> 00:26:43.476
back on sort of visual effects

659
00:26:43.810 --> 00:26:45.645
to fix sort of things.

660
00:26:45.854 --> 00:26:47.564
And that sort of happens a

661
00:26:47.647 --> 00:26:49.316
lot in virtual production.

662
00:26:49.399 --> 00:26:51.067
I mean, as much as we're all

663
00:26:51.151 --> 00:26:54.237
trying to get 100% in camera

664
00:26:54.571 --> 00:26:56.614
out of the virtual production techniques,

665
00:26:57.949 --> 00:27:00.952
I mean, it's most probably more like, you know,

666
00:27:01.369 --> 00:27:02.746
it could be up to 60 to 70.

667
00:27:02.996 --> 00:27:04.372
If it's a long form project,

668
00:27:04.914 --> 00:27:07.709
the problem is that so many changes can happen

669
00:27:08.001 --> 00:27:09.169
further down the pipeline.

670
00:27:09.377 --> 00:27:11.588
You know, it's very hard to make all the decisions

671
00:27:11.755 --> 00:27:13.381
at the beginning of the project,

672
00:27:13.465 --> 00:27:15.425
and that is essentially what is happening

673
00:27:15.508 --> 00:27:16.301
in virtual production.

674
00:27:16.593 --> 00:27:19.387
So I think you kind of then look at it as,

675
00:27:20.263 --> 00:27:21.222
what percentage are you

676
00:27:21.306 --> 00:27:22.390
gonna actually get in camera

677
00:27:22.474 --> 00:27:24.059
that you're happy with,

678
00:27:24.142 --> 00:27:25.226
and it's not gonna have changed

679
00:27:25.310 --> 00:27:26.311
when you've suddenly decided

680
00:27:26.394 --> 00:27:27.437
you're changing your character

681
00:27:27.520 --> 00:27:29.397
or you're changing your monster, whatever it is,

682
00:27:29.731 --> 00:27:30.732
later on in the process,

683
00:27:30.815 --> 00:27:32.233
then of course everything that you shot

684
00:27:32.400 --> 00:27:33.234
gets thrown out the window.

685
00:27:34.527 --> 00:27:36.112
- But it's the same in

686
00:27:36.196 --> 00:27:37.655
regular production, isn't it?

687
00:27:37.739 --> 00:27:40.116
Like it's exactly the same, you know,

688
00:27:40.367 --> 00:27:42.535
possibly more visual effects required

689
00:27:42.786 --> 00:27:44.871
for things that are shot photographically

690
00:27:44.954 --> 00:27:47.040
in the real world, right?

691
00:27:47.457 --> 00:27:50.668
Like nearly every shot in a long form show

692
00:27:50.835 --> 00:27:52.962
can have a hidden visual effect in it

693
00:27:53.046 --> 00:27:54.464
that the audience would never know about.

694
00:27:54.547 --> 00:27:55.882
- That's true, but the problem is,

695
00:27:55.965 --> 00:27:58.051
is that when you don't wanna pay twice, you know?

696
00:27:58.134 --> 00:28:00.220
And so the issue in that sense

697
00:28:00.303 --> 00:28:01.179
is that if you're shooting it,

698
00:28:01.262 --> 00:28:02.222
that's fine, but then you

699
00:28:02.305 --> 00:28:04.432
might have a visual effects budget

700
00:28:04.682 --> 00:28:06.309
on top of that and that works out.

701
00:28:06.393 --> 00:28:07.936
But then if you're kind of, you know,

702
00:28:08.019 --> 00:28:10.605
already doing a virtual production budget,

703
00:28:10.814 --> 00:28:12.607
which in some cases can be higher

704
00:28:12.816 --> 00:28:14.818
than just going out and shooting on location,

705
00:28:16.528 --> 00:28:17.445
then you're sort of,

706
00:28:17.821 --> 00:28:18.905
and then you then need to

707
00:28:18.988 --> 00:28:20.198
do the visual effects process

708
00:28:20.323 --> 00:28:21.366
back on top of that,

709
00:28:21.449 --> 00:28:24.369
then that can become a second, like a double cost.

710
00:28:24.953 --> 00:28:26.996
I think that there certainly is some sort of,

711
00:28:28.206 --> 00:28:29.416
you know, you have to think about that

712
00:28:29.499 --> 00:28:31.668
and make sure that everybody's on board

713
00:28:32.043 --> 00:28:32.710
with what you're gonna get

714
00:28:32.794 --> 00:28:34.337
and the fact that you probably have to build

715
00:28:34.421 --> 00:28:36.297
a contingency plan into the budget

716
00:28:36.506 --> 00:28:38.800
that there will be shots that need to be fixed

717
00:28:38.883 --> 00:28:42.679
or shots that need to change or extended, you know,

718
00:28:42.762 --> 00:28:44.180
and that tends to always be the case.

719
00:28:45.098 --> 00:28:46.933
- So can you talk to me a little bit about

720
00:28:47.058 --> 00:28:49.769
how you do that pre-visualization

721
00:28:50.145 --> 00:28:53.022
and how you might digitally scout a location

722
00:28:53.356 --> 00:28:57.610
and how do you get that world built digitally

723
00:28:58.153 --> 00:29:01.406
before you go ahead and, you know, storyboard and,

724
00:29:02.031 --> 00:29:03.283
yeah, how do you take the director

725
00:29:03.408 --> 00:29:04.617
through these environments?

726
00:29:04.909 --> 00:29:07.328
Because you're not physically location scouting

727
00:29:07.412 --> 00:29:09.372
and looking for physical props here.

728
00:29:10.665 --> 00:29:12.041
- Well, I mean, you say that,

729
00:29:12.125 --> 00:29:13.710
but sometimes that is the case

730
00:29:13.793 --> 00:29:14.794
and certainly in the proof

731
00:29:14.878 --> 00:29:16.796
that we worked on together,

732
00:29:17.714 --> 00:29:19.090
all those locations were

733
00:29:19.174 --> 00:29:22.093
essentially sort of located

734
00:29:22.385 --> 00:29:24.888
in the real world locations.

735
00:29:25.221 --> 00:29:27.974
And then we went and, you know,

736
00:29:28.099 --> 00:29:30.268
sometimes went and LIDAR scan those sets

737
00:29:30.393 --> 00:29:33.229
and converted them to digital sort of sets.

738
00:29:34.314 --> 00:29:37.233
Or we just took heaps and heaps of reference photos

739
00:29:37.358 --> 00:29:38.318
from that and then just

740
00:29:38.401 --> 00:29:40.695
digitally replicated those places

741
00:29:40.862 --> 00:29:43.239
because those needed to be iconic locations

742
00:29:43.323 --> 00:29:44.741
that people needed to recognize.

743
00:29:45.783 --> 00:29:48.369
But again, in say, you know,

744
00:29:48.453 --> 00:29:51.080
we've done a lot of interiors and things like that,

745
00:29:52.332 --> 00:29:54.667
you know, in that sense,

746
00:29:54.751 --> 00:29:57.378
we will actually initially

747
00:29:57.462 --> 00:29:59.839
talk to the production designer

748
00:30:00.048 --> 00:30:01.132
and or the client,

749
00:30:01.382 --> 00:30:03.885
depending on the size of the project

750
00:30:04.594 --> 00:30:06.888
and, you know, come up with

751
00:30:06.971 --> 00:30:08.973
the locations with, you know,

752
00:30:09.057 --> 00:30:10.975
treatments, directors boards, all the rest of it

753
00:30:11.059 --> 00:30:12.644
that you generally come up with

754
00:30:12.936 --> 00:30:14.812
and send that to our VAD team

755
00:30:14.896 --> 00:30:17.106
who will start building it out

756
00:30:17.315 --> 00:30:21.277
and will do sort of large renders

757
00:30:22.403 --> 00:30:24.531
as we're moving along in the process

758
00:30:24.697 --> 00:30:25.740
and showing those to the

759
00:30:25.823 --> 00:30:28.076
client and director initially

760
00:30:28.952 --> 00:30:30.370
for them to sort of kind of

761
00:30:30.453 --> 00:30:31.663
enjoy what they're looking at.

762
00:30:32.455 --> 00:30:34.624
And then once we know we're on board

763
00:30:34.791 --> 00:30:37.502
with that environment we're building to the right,

764
00:30:38.336 --> 00:30:40.713
or at least the correct creative on that,

765
00:30:41.881 --> 00:30:43.258
then we will go in and we'll

766
00:30:43.341 --> 00:30:44.592
do sort of digital scouting,

767
00:30:45.176 --> 00:30:47.470
which is, you know, you can either do that

768
00:30:47.971 --> 00:30:49.055
with the virtual headset,

769
00:30:49.305 --> 00:30:50.723
like a VR headset at least,

770
00:30:51.140 --> 00:30:53.393
or, you know, we'll just,

771
00:30:53.476 --> 00:30:55.353
you know, get a director in,

772
00:30:55.562 --> 00:30:57.188
you know, sit over one of the artists shoulders

773
00:30:57.522 --> 00:30:58.940
or whoever it is, and then

774
00:30:59.023 --> 00:31:00.358
we'll actually lens it up

775
00:31:00.441 --> 00:31:01.901
for them, you know, with the

776
00:31:01.985 --> 00:31:04.195
proper camera and sensor sizes

777
00:31:04.487 --> 00:31:06.739
and then we can sort of look at finding

778
00:31:06.864 --> 00:31:08.491
those right frames that they like.

779
00:31:09.867 --> 00:31:11.911
And then those can become part of the storyboards

780
00:31:11.995 --> 00:31:13.162
and then we'll go ahead and

781
00:31:13.246 --> 00:31:15.081
either have digital storyboards

782
00:31:15.206 --> 00:31:17.083
or we'll kind of draw back over top of them

783
00:31:17.166 --> 00:31:18.960
and add people in, or we

784
00:31:19.043 --> 00:31:20.628
could do that in Unreal as well.

785
00:31:20.878 --> 00:31:21.713
I mean, that's also sort of

786
00:31:21.796 --> 00:31:23.923
quite a few methods in that sense.

787
00:31:24.924 --> 00:31:25.967
And then after we've sort

788
00:31:26.050 --> 00:31:27.552
of built these environments

789
00:31:28.678 --> 00:31:30.054
that I think the director's

790
00:31:30.138 --> 00:31:31.514
also sort of enjoying and liking

791
00:31:31.598 --> 00:31:33.141
and we know now what to build,

792
00:31:33.975 --> 00:31:35.810
we'll then do something called techfers

793
00:31:35.893 --> 00:31:40.148
where we create the panel layout,

794
00:31:40.231 --> 00:31:41.899
which we've already sort of worked out.

795
00:31:41.983 --> 00:31:43.568
Well, it's kind of like a

796
00:31:43.651 --> 00:31:44.819
mutual sort of decision there

797
00:31:44.944 --> 00:31:45.945
because here at Steelbridge,

798
00:31:46.029 --> 00:31:48.489
we tend to build a volume out

799
00:31:48.573 --> 00:31:52.493
based on the needs of the shoot.

800
00:31:53.036 --> 00:31:56.331
It's not always a general LED

801
00:31:56.414 --> 00:31:58.166
volume layout that you would have.

802
00:31:58.249 --> 00:32:00.251
You know, we wouldn't always have a floating wall

803
00:32:00.335 --> 00:32:02.712
that can be maneuvered if we don't have a car

804
00:32:02.795 --> 00:32:04.505
or sort of some sort of object

805
00:32:04.589 --> 00:32:07.050
that needs to be reflected on all the time.

806
00:32:08.426 --> 00:32:11.137
So there's different ways to build the volume.

807
00:32:11.220 --> 00:32:12.555
We won't always have a ceiling in.

808
00:32:13.264 --> 00:32:16.309
We probably won't always build a curve to our wall.

809
00:32:16.392 --> 00:32:17.393
It might be a flat wall.

810
00:32:17.852 --> 00:32:20.730
I mean, there's multiple sort of constraints

811
00:32:21.022 --> 00:32:22.982
and also reasons why we would

812
00:32:23.066 --> 00:32:24.817
build them in a certain way.

813
00:32:24.901 --> 00:32:28.529
So then in that sense, we would then work out

814
00:32:29.030 --> 00:32:32.033
what size volume we need based on those

815
00:32:32.116 --> 00:32:35.912
that tech scout essentially that we've done,

816
00:32:35.995 --> 00:32:38.206
or sorry, that virtual scout that we've done.

817
00:32:38.539 --> 00:32:39.749
And then we'll add in the

818
00:32:39.832 --> 00:32:41.584
panels that we've built digitally

819
00:32:42.001 --> 00:32:43.211
into those environments.

820
00:32:43.294 --> 00:32:44.379
And you can actually then work

821
00:32:44.462 --> 00:32:47.632
out the distances prior to actually,

822
00:32:48.883 --> 00:32:51.719
you know, your your art

823
00:32:51.803 --> 00:32:53.388
department comes in or anything like that

824
00:32:53.471 --> 00:32:55.348
when you're starting to build out the sets.

825
00:32:55.973 --> 00:32:57.058
And you can then work out

826
00:32:57.141 --> 00:32:59.143
the scale inside the studio

827
00:32:59.352 --> 00:33:02.063
and what your best focal length could be

828
00:33:02.146 --> 00:33:04.649
or what the constraints are on

829
00:33:04.732 --> 00:33:06.359
the the lensing that you need,

830
00:33:06.442 --> 00:33:07.235
you know, and you can sort

831
00:33:07.318 --> 00:33:09.278
of then inform the the DOP

832
00:33:09.570 --> 00:33:12.448
as to sort of what, you know, we

833
00:33:12.532 --> 00:33:14.826
then want to talk and engage the DOP

834
00:33:14.992 --> 00:33:16.703
obviously early on as well.

835
00:33:17.620 --> 00:33:18.538
And also with the director

836
00:33:18.621 --> 00:33:19.997
and what sort of shots he wants,

837
00:33:20.248 --> 00:33:22.834
but and what the constraints are within the studio

838
00:33:22.917 --> 00:33:24.210
or whether we need to go to another

839
00:33:24.293 --> 00:33:26.129
studio and build it out over there.

840
00:33:27.088 --> 00:33:28.172
And then sort of all those

841
00:33:28.256 --> 00:33:29.841
decisions are informed by this tech

842
00:33:29.924 --> 00:33:32.218
that you end up doing after

843
00:33:32.301 --> 00:33:33.970
sort of building the environments.

844
00:33:35.221 --> 00:33:35.888
Yeah, that's cool.

845
00:33:35.972 --> 00:33:38.015
And then the DP, like, do

846
00:33:38.099 --> 00:33:39.142
you find that there are certain

847
00:33:39.225 --> 00:33:41.185
DPs who are becoming specialised

848
00:33:41.269 --> 00:33:42.770
in shooting in these environments

849
00:33:43.020 --> 00:33:45.148
or can sort of anybody with an

850
00:33:45.231 --> 00:33:46.441
understanding of how to shoot

851
00:33:46.524 --> 00:33:47.608
in the physical world

852
00:33:47.692 --> 00:33:49.444
shoot in a digital environment?

853
00:33:50.111 --> 00:33:51.612
Or is it a bit of a specialised thing?

854
00:33:52.530 --> 00:33:54.699
No, I mean, I think that, you know,

855
00:33:54.782 --> 00:33:57.535
for us, we love getting new DPs in

856
00:33:58.244 --> 00:34:01.372
to certainly to train them up.

857
00:34:01.664 --> 00:34:03.666
And also when I say train them up,

858
00:34:03.916 --> 00:34:05.668
we don't obviously have to train them up on this

859
00:34:05.752 --> 00:34:07.879
craft because, you know,

860
00:34:08.755 --> 00:34:11.257
it's all very much translatable, like

861
00:34:11.340 --> 00:34:12.967
everything's very much translatable.

862
00:34:13.718 --> 00:34:18.181
It's more just about understanding the process

863
00:34:19.140 --> 00:34:22.727
where he can relate it to what he's used to.

864
00:34:22.810 --> 00:34:24.645
And then he just essentially

865
00:34:24.729 --> 00:34:27.106
needs to know who he needs to talk to,

866
00:34:27.190 --> 00:34:30.401
to be able to fix something that he needs to fix.

867
00:34:30.485 --> 00:34:32.361
And at the same time, so it's still his eye.

868
00:34:32.445 --> 00:34:33.821
It's still what he wants to do.

869
00:34:34.197 --> 00:34:35.156
There certainly are some

870
00:34:35.239 --> 00:34:38.034
constraints on lensing that that will be

871
00:34:38.743 --> 00:34:40.536
purely based on the framing

872
00:34:40.620 --> 00:34:42.789
and the boards and and then also,

873
00:34:43.122 --> 00:34:44.916
you know, the aperture like, you

874
00:34:44.999 --> 00:34:47.418
know, we really need to always be

875
00:34:47.919 --> 00:34:49.295
on that fine line of

876
00:34:50.797 --> 00:34:52.590
wide open if we can,

877
00:34:53.758 --> 00:34:55.134
purely because of the scale

878
00:34:55.218 --> 00:34:57.428
of our studio and also the

879
00:34:58.054 --> 00:35:01.808
the distance that the panels are from from our

880
00:35:02.892 --> 00:35:04.769
our main actors or whatever.

881
00:35:04.852 --> 00:35:07.522
So the idea is is that we can't

882
00:35:07.605 --> 00:35:09.524
allow those panels to come into focus

883
00:35:09.607 --> 00:35:10.441
because, of course, we have

884
00:35:10.525 --> 00:35:12.985
this moire issue that can happen.

885
00:35:13.820 --> 00:35:15.571
And so we always need to have

886
00:35:15.655 --> 00:35:18.616
quite a shallow depth of field.

887
00:35:18.699 --> 00:35:19.867
Like we don't want to have too

888
00:35:19.951 --> 00:35:21.661
much too much depth of field.

889
00:35:22.411 --> 00:35:24.205
We need to limit that a little.

890
00:35:24.372 --> 00:35:26.249
And so unfortunately, that's sort of something that

891
00:35:26.332 --> 00:35:29.168
not all DPs want to be working with.

892
00:35:29.710 --> 00:35:31.128
And it does give a certain look.

893
00:35:32.046 --> 00:35:34.799
But we we always try to work

894
00:35:34.882 --> 00:35:37.218
right on that edge of moire

895
00:35:37.468 --> 00:35:39.011
because more focus in the

896
00:35:39.095 --> 00:35:43.057
background, more realism creeps back into it.

897
00:35:43.140 --> 00:35:44.892
You know, if it's everything feels a bit sort of

898
00:35:44.976 --> 00:35:46.811
under watery, it's just not looking right.

899
00:35:46.894 --> 00:35:49.939
You know, so it's always it's a very that's why it

900
00:35:50.022 --> 00:35:51.607
becomes quite mathematical.

901
00:35:52.400 --> 00:35:55.945
And the lenses need to be very we have to profile

902
00:35:56.028 --> 00:35:57.572
the lenses and really test the lenses

903
00:35:58.281 --> 00:35:59.866
to make sure that we're not

904
00:35:59.949 --> 00:36:01.993
suddenly on the edge of moire.

905
00:36:02.869 --> 00:36:04.787
And in that project that you and I

906
00:36:04.871 --> 00:36:07.248
worked on, which was super, super crazy,

907
00:36:07.456 --> 00:36:10.334
the lenses had a weird pincushion

908
00:36:10.835 --> 00:36:13.963
scenario where the outer edges were

909
00:36:14.046 --> 00:36:15.590
more in focus than the inner edges.

910
00:36:15.965 --> 00:36:19.051
And so it was it was very strange.

911
00:36:19.760 --> 00:36:20.803
And, you know, these were sort

912
00:36:20.887 --> 00:36:24.140
of old school lenses.

913
00:36:24.223 --> 00:36:25.391
And we really didn't do

914
00:36:25.474 --> 00:36:26.434
our homework on those lenses.

915
00:36:26.726 --> 00:36:31.480
And so it kind of, you know, essentially meant that

916
00:36:31.564 --> 00:36:32.899
we had to push everything further

917
00:36:33.357 --> 00:36:35.151
away from the screens even more,

918
00:36:36.235 --> 00:36:37.486
which meant that we then were limited

919
00:36:37.653 --> 00:36:39.697
by the framing that we wanted to

920
00:36:39.780 --> 00:36:42.992
achieve. So, yeah, so it's not something

921
00:36:43.075 --> 00:36:44.869
that I typically think about is like

922
00:36:44.952 --> 00:36:48.748
the amount of focus across the lens.

923
00:36:48.998 --> 00:36:51.709
I you know, usually you're thinking about depth of

924
00:36:51.792 --> 00:36:53.920
field and, you know, your F stop

925
00:36:54.045 --> 00:36:55.338
and all of that and how much

926
00:36:55.421 --> 00:36:56.422
light you've got in the scene.

927
00:36:56.505 --> 00:36:58.925
But, yeah, I mean, I suppose when

928
00:36:59.008 --> 00:37:00.384
you're really on the edge there,

929
00:37:00.468 --> 00:37:02.303
if it's a bit sharper in one area of

930
00:37:02.386 --> 00:37:04.096
the lens and a bit softer in the other,

931
00:37:04.513 --> 00:37:06.349
you don't want that sharp part of

932
00:37:06.432 --> 00:37:09.352
the lens to start to see the actual LED

933
00:37:10.519 --> 00:37:11.896
points, right? Because that's

934
00:37:11.979 --> 00:37:13.564
what creates the moire pattern.

935
00:37:13.856 --> 00:37:17.443
Yes, that you've got these, you know, grid

936
00:37:17.526 --> 00:37:21.197
essentially of points that will strobe.

937
00:37:22.657 --> 00:37:24.325
Exactly. Yeah. And that's yeah.

938
00:37:24.408 --> 00:37:27.620
I mean, that's sort of the we're always working to

939
00:37:27.703 --> 00:37:29.413
make sure we don't see that.

940
00:37:30.206 --> 00:37:32.208
Yeah. Yeah. Wow. There's so

941
00:37:32.291 --> 00:37:33.876
many considerations there.

942
00:37:34.502 --> 00:37:36.587
And on top of that, you're actually

943
00:37:36.671 --> 00:37:40.091
creating either a 3D or a two and a half

944
00:37:40.174 --> 00:37:43.177
day environment, and I take it

945
00:37:43.260 --> 00:37:45.388
that there's actually quite a bit of

946
00:37:45.721 --> 00:37:49.600
like color pipeline that has to happen between the

947
00:37:49.684 --> 00:37:50.810
start and the end of that,

948
00:37:50.893 --> 00:37:54.355
because you're actually projecting an image that is

949
00:37:54.438 --> 00:37:57.024
quite final at that point.

950
00:37:57.400 --> 00:37:58.901
That's right. Yeah.

951
00:37:59.610 --> 00:38:02.530
How do you how do you design the

952
00:38:02.613 --> 00:38:04.407
pipe there to get the colors accurate?

953
00:38:05.574 --> 00:38:07.785
Well, I mean, for us, you know, the most important

954
00:38:07.868 --> 00:38:09.829
thing is to make sure that this

955
00:38:10.204 --> 00:38:11.872
and anything on the screen just

956
00:38:11.956 --> 00:38:15.001
looks realistic, you know, and natural.

957
00:38:16.043 --> 00:38:18.504
So if you're looking at the screen with your eye,

958
00:38:18.587 --> 00:38:20.548
it needs to sort of feel real

959
00:38:20.715 --> 00:38:24.760
as well. But and I know it sounds sort of, you

960
00:38:24.844 --> 00:38:27.221
know, you think that's what you would

961
00:38:27.304 --> 00:38:29.181
expect, you know, but the problem is,

962
00:38:29.265 --> 00:38:30.683
is that when you're creating digital

963
00:38:30.766 --> 00:38:33.269
environments, you can very easily get illegal

964
00:38:33.352 --> 00:38:34.812
colors, as we'd like to call them,

965
00:38:34.895 --> 00:38:39.608
which are not really real life kind of colors.

966
00:38:39.692 --> 00:38:41.402
And that all is dependent on the

967
00:38:41.485 --> 00:38:44.155
renderer if you're doing digital content.

968
00:38:45.489 --> 00:38:47.366
And that can be kind of quite tricky.

969
00:38:47.450 --> 00:38:51.370
So we do employ an ACES pipeline.

970
00:38:52.329 --> 00:38:56.000
And for us, we've got, you know, OCIO

971
00:38:56.834 --> 00:39:00.129
library, which we output

972
00:39:00.212 --> 00:39:02.548
from Unreal, an ACES color.

973
00:39:04.008 --> 00:39:05.676
And, you know, whether we then

974
00:39:05.760 --> 00:39:08.512
re-project that out onto the panels

975
00:39:09.472 --> 00:39:13.517
as an ACES profile, and then that gets converted

976
00:39:13.601 --> 00:39:15.394
with a look up on the processor

977
00:39:16.395 --> 00:39:17.897
before it hits the panel.

978
00:39:18.856 --> 00:39:22.109
That's sort of something that we sometimes shift

979
00:39:22.193 --> 00:39:23.611
between whether we actually do

980
00:39:23.694 --> 00:39:24.904
the conversion on in

981
00:39:24.987 --> 00:39:27.782
Unreal and then onto the screen.

982
00:39:27.865 --> 00:39:28.908
But the thing is, is that obviously

983
00:39:28.991 --> 00:39:31.702
the panels are all slightly different

984
00:39:32.161 --> 00:39:33.871
in some instances, like as in

985
00:39:33.954 --> 00:39:35.915
different brands have got like different color

986
00:39:37.500 --> 00:39:38.751
with different looks, you know.

987
00:39:38.959 --> 00:39:42.713
And then the problem with the LEDs is that they,

988
00:39:43.214 --> 00:39:44.423
you know, of course, it's an RGB

989
00:39:46.550 --> 00:39:47.927
light, essentially, each one

990
00:39:48.010 --> 00:39:50.012
of those little pixels on there.

991
00:39:50.387 --> 00:39:52.389
And you can get something called

992
00:39:52.473 --> 00:39:55.393
color shift very easily at certain angles

993
00:39:55.476 --> 00:39:58.562
that you look at a panel and, you know, you get

994
00:39:58.646 --> 00:40:00.314
these kind of weird magenta reds

995
00:40:00.397 --> 00:40:01.440
or you get the greens if

996
00:40:01.524 --> 00:40:02.942
you're looking at opposite angles.

997
00:40:03.275 --> 00:40:05.736
And so that that's kind of

998
00:40:05.820 --> 00:40:07.947
got multiple issues with that.

999
00:40:08.906 --> 00:40:13.410
Is that how it translates onto skin tones as well?

1000
00:40:13.494 --> 00:40:14.578
That can be a real problem.

1001
00:40:14.662 --> 00:40:17.998
And if you start sort of shifting color, overall

1002
00:40:18.082 --> 00:40:19.750
color, for instance, on a panel

1003
00:40:20.376 --> 00:40:25.172
and what that actually does to skin tone is it's

1004
00:40:25.256 --> 00:40:26.173
quite remarkable, actually,

1005
00:40:26.257 --> 00:40:27.383
in some instances, you just

1006
00:40:27.466 --> 00:40:29.260
sort of hope that it just works.

1007
00:40:29.718 --> 00:40:32.388
Everybody's skin tones are slightly different and

1008
00:40:32.471 --> 00:40:34.181
the LED panels can really

1009
00:40:36.100 --> 00:40:39.103
just just warm it up or just make things look red.

1010
00:40:39.186 --> 00:40:41.897
Some skins just goes red under panels.

1011
00:40:42.481 --> 00:40:44.316
And it's very, very tricky.

1012
00:40:44.567 --> 00:40:46.193
And some skin tones just do not.

1013
00:40:46.277 --> 00:40:47.361
So I'm just look fine, you

1014
00:40:47.444 --> 00:40:48.529
know, and so you have to.

1015
00:40:48.612 --> 00:40:50.072
It's quite a tricky sort of

1016
00:40:50.156 --> 00:40:52.741
science to get that right.

1017
00:40:52.992 --> 00:40:55.578
So our main aim is to make sure everything's sort

1018
00:40:55.661 --> 00:40:58.330
of linear as it comes out of unreal

1019
00:40:59.248 --> 00:41:00.499
and into the processor.

1020
00:41:00.708 --> 00:41:02.168
So we want to try and make sure

1021
00:41:02.251 --> 00:41:04.837
that there's not too much done.

1022
00:41:04.920 --> 00:41:06.881
Like we don't want to do much color management

1023
00:41:06.964 --> 00:41:08.591
prior to it going to the Brompton

1024
00:41:09.300 --> 00:41:12.595
processor, which is the processors that we use to

1025
00:41:12.678 --> 00:41:14.680
drive a signal to the panels.

1026
00:41:15.764 --> 00:41:18.934
And on the Brompton processor, there is quite a bit

1027
00:41:19.018 --> 00:41:20.311
of color management that we

1028
00:41:20.394 --> 00:41:26.400
we tend to adjust and and use to help merge,

1029
00:41:28.819 --> 00:41:31.363
you know, merge the foreground like the practical

1030
00:41:31.447 --> 00:41:32.656
and the background together.

1031
00:41:33.324 --> 00:41:34.867
And then, of course, you really

1032
00:41:34.950 --> 00:41:36.785
only want the camera to be driving

1033
00:41:37.536 --> 00:41:40.039
the color after that fact, you know, with its color

1034
00:41:40.122 --> 00:41:41.332
temperature and then the lights

1035
00:41:41.415 --> 00:41:43.042
that then change accordingly.

1036
00:41:44.001 --> 00:41:45.920
So, you know, just like all lights,

1037
00:41:46.003 --> 00:41:49.048
panels can be warmed up and cool down.

1038
00:41:49.757 --> 00:41:51.759
And most LEDs run quite cool.

1039
00:41:52.384 --> 00:41:54.470
They're very sort of like as a default, they're

1040
00:41:54.553 --> 00:41:56.138
quite a cool sort of temperature.

1041
00:41:56.472 --> 00:42:00.100
So you can warm them up to

1042
00:42:00.184 --> 00:42:01.602
to try and balance it out.

1043
00:42:01.685 --> 00:42:03.938
But in doing so, you know,

1044
00:42:04.939 --> 00:42:06.982
it's very easy to break things.

1045
00:42:07.316 --> 00:42:09.109
And so that's why, like I said, we just try and

1046
00:42:09.193 --> 00:42:10.152
make sure that everything coming

1047
00:42:10.236 --> 00:42:12.279
out of unreal is very linear.

1048
00:42:12.488 --> 00:42:14.281
And so like, again, we transcode

1049
00:42:14.365 --> 00:42:17.159
everything that as in textures and or

1050
00:42:18.118 --> 00:42:19.119
footage that we're using

1051
00:42:19.203 --> 00:42:21.205
into the ACES color workflow.

1052
00:42:22.331 --> 00:42:26.335
And then we will then convert it out onto the

1053
00:42:26.418 --> 00:42:27.836
panels through the Brompton.

1054
00:42:28.671 --> 00:42:32.466
So it's kind of and then we could possibly look at

1055
00:42:32.549 --> 00:42:34.134
it in any other look up after that.

1056
00:42:34.218 --> 00:42:36.720
But we just want to have as much color gamut as we

1057
00:42:36.804 --> 00:42:38.430
can have, you know, and make sure

1058
00:42:38.514 --> 00:42:40.099
that those panels are working correctly.

1059
00:42:41.016 --> 00:42:43.018
And so, oh, sorry, you go.

1060
00:42:43.811 --> 00:42:44.979
Oh, I was just going to ask -

1061
00:42:45.062 --> 00:42:48.232
So about that linear signal, I mean, my

1062
00:42:48.315 --> 00:42:51.694
understanding of the usefulness of linear

1063
00:42:52.111 --> 00:42:55.281
for visual effects is that you can do more

1064
00:42:55.364 --> 00:42:57.658
realistic things because that's the way

1065
00:42:57.741 --> 00:43:00.160
that light kind of functions in the in the real

1066
00:43:00.244 --> 00:43:03.622
world is that it does work like linear.

1067
00:43:04.456 --> 00:43:06.834
Not like logarithmic curves.

1068
00:43:07.585 --> 00:43:09.503
You don't you know, you don't end up with those

1069
00:43:09.586 --> 00:43:12.256
soft roll offs that we then put on once

1070
00:43:12.339 --> 00:43:15.342
we're starting to work in the log space afterwards.

1071
00:43:16.218 --> 00:43:20.180
Is that why the linear pipe works so well for this

1072
00:43:20.264 --> 00:43:22.474
or is it just because it's such a

1073
00:43:23.142 --> 00:43:25.894
robust pipeline for visual

1074
00:43:25.978 --> 00:43:28.063
effects working in ACES and OCIO?

1075
00:43:29.273 --> 00:43:32.192
Well, I mean, I think just obviously keeping it

1076
00:43:32.276 --> 00:43:35.195
linear just allows us not to add any sort

1077
00:43:35.279 --> 00:43:38.824
of look ups on top so that you can kind of keep the

1078
00:43:38.907 --> 00:43:40.784
color information there for as long

1079
00:43:40.993 --> 00:43:45.622
as possible and then try and only, you know,

1080
00:43:45.706 --> 00:43:47.416
convert it later down the pipe.

1081
00:43:48.584 --> 00:43:50.336
And again, like you said, if we have some sort of

1082
00:43:50.419 --> 00:43:52.338
roll off that's happening prior to

1083
00:43:52.421 --> 00:43:56.133
it hitting the panels, then it's very hard to

1084
00:43:56.216 --> 00:43:59.261
invert that, you know, the other way around.

1085
00:43:59.511 --> 00:44:01.930
So keeping the signal as linear as possible all the

1086
00:44:02.014 --> 00:44:04.391
way through in terms of what you were

1087
00:44:04.475 --> 00:44:08.937
saying in digitally, it makes total sense that we

1088
00:44:09.021 --> 00:44:12.066
like to linearize everything so that

1089
00:44:12.149 --> 00:44:15.986
when our final look up happens in the renderer, we

1090
00:44:16.070 --> 00:44:17.780
actually then convert it so we can look

1091
00:44:18.072 --> 00:44:21.992
at it. Essentially, we'll look at it in, you know,

1092
00:44:22.076 --> 00:44:23.869
Rec 7-  Rec 2020 or whatever it is

1093
00:44:23.952 --> 00:44:25.329
that we tend to look at it in.

1094
00:44:25.871 --> 00:44:28.082
But we will still be working

1095
00:44:28.165 --> 00:44:31.502
linearly the whole way through.

1096
00:44:32.002 --> 00:44:34.171
And it's also because of the renderers that we use.

1097
00:44:34.254 --> 00:44:37.633
So our renderer is the way that they actually

1098
00:44:37.716 --> 00:44:40.135
render, you know, whether you're using like

1099
00:44:40.219 --> 00:44:45.224
you know, it depends on if it's an unbiased

1100
00:44:45.307 --> 00:44:47.267
renderer, which means essentially it just shoots

1101
00:44:47.434 --> 00:44:50.604
out light rays just like you would have in reality,

1102
00:44:51.146 --> 00:44:52.523
you know, with photons, you know,

1103
00:44:53.148 --> 00:44:55.859
essentially, your materials are like balanced and

1104
00:44:55.943 --> 00:44:57.444
real world balanced as well.

1105
00:44:57.736 --> 00:45:01.031
So they just work far better in terms of a linear

1106
00:45:01.115 --> 00:45:03.409
workflow, because essentially computers

1107
00:45:03.492 --> 00:45:05.786
work linearly initially, and you don't want to bake

1108
00:45:05.869 --> 00:45:09.581
into the the any of the color or anything

1109
00:45:09.748 --> 00:45:12.209
like that within your digital, the digital

1110
00:45:12.292 --> 00:45:14.044
component of your renderer.

1111
00:45:14.336 --> 00:45:17.339
If you start introducing some sort of look up, it

1112
00:45:17.423 --> 00:45:19.299
just breaks and then you have to invert it

1113
00:45:19.383 --> 00:45:22.219
somehow down the line. So we stay linear as much as

1114
00:45:22.302 --> 00:45:23.971
possible all the way up until the end.

1115
00:45:25.472 --> 00:45:27.599
And then, of course, as you know, like, you know,

1116
00:45:27.683 --> 00:45:30.018
the camera itself will be adding to that

1117
00:45:30.102 --> 00:45:33.230
as well. And so, you know, you just have to

1118
00:45:33.313 --> 00:45:36.150
mitigate too many things that can change all

1119
00:45:36.233 --> 00:45:37.443
the way up until that point.

1120
00:45:38.068 --> 00:45:40.237
Yes, yeah, because the camera itself is going to

1121
00:45:40.320 --> 00:45:43.407
have, you know, its own ISO natively and it's

1122
00:45:43.490 --> 00:45:46.034
going to have its own range of stops that it can

1123
00:45:46.118 --> 00:45:49.788
render onto the actual image and capture.

1124
00:45:51.290 --> 00:45:54.334
Wow, so much can go wrong,

1125
00:45:54.418 --> 00:45:55.919
so much needs looking at.

1126
00:45:56.712 --> 00:45:58.464
So do you essentially color

1127
00:45:58.547 --> 00:46:02.468
the material in that process?

1128
00:46:02.885 --> 00:46:04.136
I know that you're saying that you

1129
00:46:04.219 --> 00:46:06.346
are trying not to do very much to it.

1130
00:46:06.430 --> 00:46:11.852
But like when I looked at the raw ungraded footage

1131
00:46:11.935 --> 00:46:14.062
for the car commercial that we looked at

1132
00:46:14.146 --> 00:46:18.233
together, there was nothing like there weren't any

1133
00:46:18.317 --> 00:46:20.611
parts of the image that were wildly different

1134
00:46:20.694 --> 00:46:23.864
from others. It did feel as though somebody with an

1135
00:46:23.947 --> 00:46:28.327
eye had integrated and composited a world

1136
00:46:28.494 --> 00:46:29.828
together that was kind of

1137
00:46:29.912 --> 00:46:33.123
harmonious and worked color wise.

1138
00:46:33.790 --> 00:46:36.126
Choices had been made, I

1139
00:46:36.210 --> 00:46:39.838
felt, on the way to the volume.

1140
00:46:40.839 --> 00:46:44.384
Is there somebody like a compositor or a colorist

1141
00:46:44.468 --> 00:46:46.845
or a visual effects artist who is kind of

1142
00:46:46.929 --> 00:46:49.806
responsible for that or is it

1143
00:46:49.890 --> 00:46:50.807
something that just happens?

1144
00:46:51.683 --> 00:46:54.811
No, I mean, so I mean, to answer your question,

1145
00:46:54.895 --> 00:46:56.688
that's probably me that's doing that.

1146
00:46:56.772 --> 00:47:01.109
But I think, you know, again, like if we've got a

1147
00:47:01.193 --> 00:47:03.987
realistic environment in the background, the aim

1148
00:47:04.071 --> 00:47:06.281
is to make sure that that looks like a realistic

1149
00:47:06.365 --> 00:47:09.785
environment with your eye looking at it and so that

1150
00:47:09.868 --> 00:47:14.915
it needs to actually, you know, it needs to react

1151
00:47:14.998 --> 00:47:18.043
correctly to light in digitally, digital

1152
00:47:18.126 --> 00:47:20.546
light, at least. And then that light needs to sort

1153
00:47:20.629 --> 00:47:23.507
of feel realistic, you know, as you're looking at

1154
00:47:23.590 --> 00:47:25.884
it. So that's the first part of it. And then

1155
00:47:25.968 --> 00:47:27.469
secondly, when you start

1156
00:47:27.553 --> 00:47:28.971
integrating your foreground

1157
00:47:29.096 --> 00:47:32.808
elements into that, you know, the there's only so

1158
00:47:32.891 --> 00:47:35.435
much light wrapped, for instance, and this depends

1159
00:47:35.519 --> 00:47:38.105
on the size of your volume that is encompassing

1160
00:47:38.188 --> 00:47:41.400
your person or objects that you've got that are

1161
00:47:41.483 --> 00:47:45.487
being lit as well. And so essentially, a lot of

1162
00:47:45.571 --> 00:47:49.074
that can be matched, you know, from the lighting

1163
00:47:49.199 --> 00:47:51.618
that's coming from the volume. But again, like, for

1164
00:47:51.702 --> 00:47:53.954
instance, if you consider that project, we have

1165
00:47:54.037 --> 00:47:56.331
the floor, you know, that the, you know, if we look

1166
00:47:56.415 --> 00:48:00.460
at the the outback scene where, you know, it was

1167
00:48:00.544 --> 00:48:02.504
like a reddish, the reddish color never sort of we

1168
00:48:02.588 --> 00:48:05.841
never quite got that one working quite right. So we

1169
00:48:05.924 --> 00:48:09.469
can essentially, we're trying to match to the

1170
00:48:09.553 --> 00:48:14.182
foreground environment, texture and color to our

1171
00:48:14.766 --> 00:48:17.060
digital environment needs to match to that. So

1172
00:48:17.144 --> 00:48:19.896
there is a level of that happening in real time.

1173
00:48:19.980 --> 00:48:22.524
And in our brain bar of people that we have,

1174
00:48:22.608 --> 00:48:25.861
they'll be making those adjustments, looking,

1175
00:48:26.653 --> 00:48:28.655
looking at the screen, looking through the split to

1176
00:48:28.739 --> 00:48:31.408
make sure that that sort of feels more in line,

1177
00:48:31.491 --> 00:48:34.161
especially if it's like a wide global, a global

1178
00:48:34.244 --> 00:48:36.997
adjustment to a particular material, for instance,

1179
00:48:37.080 --> 00:48:40.208
not a global color adjustment that you do to a 2D

1180
00:48:40.292 --> 00:48:42.878
image, but to a specific piece, you know, like

1181
00:48:42.961 --> 00:48:45.756
the sand will change the color of the sand to match

1182
00:48:45.839 --> 00:48:47.883
to the sand that we have in the foreground.

1183
00:48:48.759 --> 00:48:51.511
Those are things that we can sort of do on the day,

1184
00:48:51.595 --> 00:48:53.221
and we'll try to get those as close as

1185
00:48:53.305 --> 00:48:55.474
possible. And that generally would happen during

1186
00:48:55.557 --> 00:48:58.602
the pre-light day, which is you'd hope to be in

1187
00:48:58.685 --> 00:49:02.648
that position. And then afterwards, there's sort of

1188
00:49:02.731 --> 00:49:06.234
subtle adjustments in color, overall color of

1189
00:49:06.318 --> 00:49:10.572
the panels that want to then match with the, I

1190
00:49:10.656 --> 00:49:11.573
would be working directly

1191
00:49:11.657 --> 00:49:14.076
with the DOP to understand

1192
00:49:14.159 --> 00:49:16.953
when he exposes in a certain way what we want to do

1193
00:49:17.037 --> 00:49:19.581
to our highlights in our scene, for instance,

1194
00:49:19.665 --> 00:49:23.710
to make those work and sort of make them feel

1195
00:49:23.794 --> 00:49:27.130
correct as well. And then we have something else,

1196
00:49:27.214 --> 00:49:30.175
which is called, well, ICV effects, which is

1197
00:49:30.258 --> 00:49:31.259
in-camera visual effects,

1198
00:49:31.343 --> 00:49:33.804
but essentially it's a new

1199
00:49:33.887 --> 00:49:37.057
projection, kind of like a, you know, power windows

1200
00:49:37.140 --> 00:49:41.269
that you can add to the background panel,

1201
00:49:41.478 --> 00:49:44.189
and you can actually isolate areas and draw mats

1202
00:49:44.272 --> 00:49:47.275
around certain areas and grade those independently

1203
00:49:47.401 --> 00:49:50.987
to try and fix that scene where you might see the

1204
00:49:51.071 --> 00:49:54.032
integration of the foreground elements hitting the

1205
00:49:54.116 --> 00:49:56.535
panel at the background, and so that those can be

1206
00:49:56.618 --> 00:49:59.996
graded independently. But we try to stay away from

1207
00:50:00.080 --> 00:50:02.332
that for as long as possible until there's just

1208
00:50:02.416 --> 00:50:04.793
like a scene that we need to fix because those,

1209
00:50:06.628 --> 00:50:09.589
you know, because they actually exist there all the

1210
00:50:09.673 --> 00:50:11.717
time, like they're stuck to that piece of the

1211
00:50:11.800 --> 00:50:14.636
panel. And so when you start moving your camera

1212
00:50:14.720 --> 00:50:15.554
around, you can imagine

1213
00:50:15.637 --> 00:50:17.097
that that doesn't necessarily

1214
00:50:17.264 --> 00:50:22.853
relate in volume, per se. And so like, as in like

1215
00:50:22.936 --> 00:50:24.396
volume tracking, so it

1216
00:50:24.479 --> 00:50:25.814
might then stay in a position

1217
00:50:26.022 --> 00:50:28.150
and just not be in the right position after that.

1218
00:50:28.525 --> 00:50:32.404
So we try to avoid that if we can, but sometimes

1219
00:50:32.487 --> 00:50:35.240
you have to actually use that. And also, you know,

1220
00:50:35.323 --> 00:50:40.036
this ICV effects option also allows us to add like,

1221
00:50:40.120 --> 00:50:42.706
you know, hotspot lights that we want to use for

1222
00:50:42.789 --> 00:50:45.959
reflections and added reflections and animate

1223
00:50:46.042 --> 00:50:48.795
those that you have, kind of like what you would do

1224
00:50:48.879 --> 00:50:52.007
with sort of like a certain LED lighting that

1225
00:50:52.174 --> 00:50:54.718
the gaffers use to be able to actually have some

1226
00:50:54.801 --> 00:50:56.970
running lights over cars and things like that,

1227
00:50:57.679 --> 00:50:59.973
which is quite cool. That's really cool. Yeah,

1228
00:51:00.474 --> 00:51:02.142
that's fun. That's really cool. I mean,

1229
00:51:02.225 --> 00:51:05.020
it does seem like there's an opportunity there for

1230
00:51:05.103 --> 00:51:09.566
a colourist on the set to be part of that

1231
00:51:09.649 --> 00:51:12.444
process. And I know that visual effects artists and

1232
00:51:12.527 --> 00:51:14.529
compositors are extremely good colourists,

1233
00:51:14.613 --> 00:51:16.198
because that's sort of essentially what a

1234
00:51:16.281 --> 00:51:19.075
compositor does is matching elements and making

1235
00:51:19.201 --> 00:51:22.412
them integrate seamlessly. Can you tell me about

1236
00:51:22.496 --> 00:51:24.706
the brain bar and who's on that and what's

1237
00:51:24.790 --> 00:51:27.834
happening there? Yeah, I mean, before I do that, I

1238
00:51:27.918 --> 00:51:30.504
mean, you're 100% right. And you know, having like

1239
00:51:30.587 --> 00:51:33.340
a colourist eye is probably, you know, would be

1240
00:51:33.423 --> 00:51:35.842
amazing to have. And I think that, you know,

1241
00:51:36.718 --> 00:51:40.972
you know, the DITs, after it's coming out of the

1242
00:51:41.056 --> 00:51:43.850
camera, I think is a better place for that

1243
00:51:43.934 --> 00:51:45.602
currently the way I see it. And that's purely

1244
00:51:45.685 --> 00:51:48.021
because, you know, we want the foreground to

1245
00:51:48.104 --> 00:51:52.108
stay married to it. I think what ends up and can

1246
00:51:52.192 --> 00:51:53.610
end up happening is that when,

1247
00:51:54.945 --> 00:52:00.283
if the background panel is being colour adjusted to

1248
00:52:00.367 --> 00:52:03.495
match to what initially is being seen,

1249
00:52:04.412 --> 00:52:07.290
the problem with that is that the balancing of

1250
00:52:07.374 --> 00:52:10.377
reality can kind of really shift, you know,

1251
00:52:10.460 --> 00:52:12.879
in that sense. So when our background becomes sort

1252
00:52:12.963 --> 00:52:16.299
of separated, I mean, yeah, there could be very

1253
00:52:16.424 --> 00:52:19.010
easy adjustments that we can make. But when you

1254
00:52:19.094 --> 00:52:22.013
start making adjustments before it hits the panels,

1255
00:52:22.681 --> 00:52:26.309
you're then in this world of changing the

1256
00:52:26.393 --> 00:52:30.856
environment to match prior to doing it after

1257
00:52:30.939 --> 00:52:33.066
the fact when you have your integrated environment

1258
00:52:33.149 --> 00:52:39.614
in there. So it's just been hard to then work out

1259
00:52:39.698 --> 00:52:42.534
why certain things aren't working. You know, if we

1260
00:52:42.617 --> 00:52:44.870
have a grade that's happening prior to it,

1261
00:52:45.203 --> 00:52:47.330
then going, we've tried this and then it going into

1262
00:52:47.414 --> 00:52:48.790
the Brompton and then getting

1263
00:52:49.082 --> 00:52:53.253
projected up on. And so when something is not

1264
00:52:53.336 --> 00:52:56.298
working, it can't all be fixed in grade before

1265
00:52:56.381 --> 00:52:58.466
it goes through, because some things, like I said,

1266
00:52:58.550 --> 00:53:00.886
for instance, we need to adjust the ground colour,

1267
00:53:01.219 --> 00:53:04.848
for instance, specifically on a material, like a 3D

1268
00:53:04.931 --> 00:53:07.392
material for that to work, or we need to adjust

1269
00:53:07.475 --> 00:53:11.855
the position, the sun position and what that then

1270
00:53:11.938 --> 00:53:14.733
ends up changing in the environment by just

1271
00:53:14.816 --> 00:53:17.277
changing the lights on everything and how then the

1272
00:53:17.360 --> 00:53:20.572
grade fights against that opposite, you know,

1273
00:53:20.655 --> 00:53:24.284
because we need to be quite fluid, essentially. But

1274
00:53:24.367 --> 00:53:27.162
I think just maybe before that would be okay,

1275
00:53:27.245 --> 00:53:29.205
but like, there's so many shots that we're doing

1276
00:53:29.289 --> 00:53:32.042
that sort of have to happen. I mean, I think

1277
00:53:32.125 --> 00:53:34.419
there's certainly, you know, heel and toe and

1278
00:53:34.502 --> 00:53:36.421
levels that need to, you know, mids that need to

1279
00:53:36.504 --> 00:53:38.381
be moved around every sort of shot. And that would

1280
00:53:38.465 --> 00:53:42.427
be amazing, you know, almost like a one light feel

1281
00:53:42.844 --> 00:53:45.055
that would be great, you know, for that to sort of

1282
00:53:45.138 --> 00:53:46.932
happen. And that is sort of something that is

1283
00:53:47.015 --> 00:53:51.937
happening every time just before we roll. But a DIT

1284
00:53:52.020 --> 00:53:54.189
approach after the fact where we have the

1285
00:53:54.272 --> 00:53:56.816
foreground in would just be amazing, you know, and

1286
00:53:56.900 --> 00:53:59.945
if that was sort of set with the director and the

1287
00:54:00.695 --> 00:54:03.531
the DOP to have a look at what's more the DOP, I

1288
00:54:03.615 --> 00:54:06.910
suppose, at that point, I think that would be,

1289
00:54:07.410 --> 00:54:09.704
you know, amazing, you know, because what that does

1290
00:54:09.788 --> 00:54:11.957
is gives the client even more

1291
00:54:13.124 --> 00:54:15.418
you know, happiness when they see that. Yeah, it's

1292
00:54:15.502 --> 00:54:17.462
perfect. It's great. This is exactly what we want.

1293
00:54:17.545 --> 00:54:19.089
And then we sort of designing a look and

1294
00:54:19.172 --> 00:54:21.257
automatically on the day, and then suddenly,

1295
00:54:21.341 --> 00:54:23.468
you know, just like you'd normally have in a long

1296
00:54:23.551 --> 00:54:26.680
form project, it just becomes perfect, you know,

1297
00:54:26.846 --> 00:54:28.807
in that I say perfect, but you know what I mean?

1298
00:54:28.890 --> 00:54:31.643
And then the clients just even more into it,

1299
00:54:31.726 --> 00:54:33.728
you know, and then they feel like and they, you

1300
00:54:33.812 --> 00:54:35.897
know, you just they just feel like a way to lift

1301
00:54:35.981 --> 00:54:38.274
it off them, because sometimes it might not feel

1302
00:54:38.358 --> 00:54:40.735
exactly like or they're not used to that process.

1303
00:54:40.986 --> 00:54:42.904
And it just, you know, sets them at ease a little

1304
00:54:42.988 --> 00:54:45.991
bit more. That's really interesting. So in terms

1305
00:54:46.074 --> 00:54:48.994
of the skills that you think that that colorist who

1306
00:54:49.077 --> 00:54:52.038
who could be part of the virtual production

1307
00:54:52.122 --> 00:54:55.834
pipeline, they would be more of a DIT to you than

1308
00:54:55.917 --> 00:54:59.254
somebody who is skilled in Unreal.

1309
00:55:00.922 --> 00:55:05.760
Yeah, I mean, I think so again, skilled in Unreal,

1310
00:55:07.178 --> 00:55:08.930
if a colorist is skilled in Unreal,

1311
00:55:09.014 --> 00:55:11.599
I think that that's an amazing thing to have. I

1312
00:55:11.683 --> 00:55:15.478
think that that would be and integrating those

1313
00:55:16.187 --> 00:55:19.566
that person would be would be sort of a priority.

1314
00:55:19.649 --> 00:55:22.152
I'd love to have somebody like that in the team

1315
00:55:22.235 --> 00:55:24.654
to be able to do that, because I think and I know

1316
00:55:24.738 --> 00:55:26.698
there's a lot of tools that are being developed

1317
00:55:26.781 --> 00:55:30.827
now where there is more control in grade that

1318
00:55:30.910 --> 00:55:34.789
happens in Unreal and as a, you know, as a wrapper

1319
00:55:34.873 --> 00:55:39.502
that goes to the next stage. And I think that that

1320
00:55:39.586 --> 00:55:41.838
that's sort of quite important. I think one

1321
00:55:41.921 --> 00:55:45.175
thing to keep in mind, especially in Unreal Engine,

1322
00:55:45.258 --> 00:55:47.635
if we're talking about digital sort of creation

1323
00:55:47.719 --> 00:55:51.890
here, the renderer is not a real, it's not a real

1324
00:55:51.973 --> 00:55:55.018
world balanced renderer like you would get in say

1325
00:55:55.101 --> 00:55:59.439
some new visual effects style renderers or not,

1326
00:55:59.522 --> 00:56:02.901
sorry, not new old renderers that have been around

1327
00:56:02.984 --> 00:56:05.987
for a long time. So there's a lot of cheats that

1328
00:56:06.071 --> 00:56:09.157
happen in Unreal Engine to be able to get that

1329
00:56:09.240 --> 00:56:12.619
renderer to, you know, work correctly for

1330
00:56:12.702 --> 00:56:15.330
optimization for shadows for bounces and all these

1331
00:56:15.413 --> 00:56:18.708
sorts of things. And the problem that ends up

1332
00:56:18.792 --> 00:56:19.959
happening is is that you

1333
00:56:20.043 --> 00:56:22.003
need to actually understand

1334
00:56:22.087 --> 00:56:25.090
what the limitations are of the renderer in order

1335
00:56:25.173 --> 00:56:28.009
to look at an image as well and go, this is what's

1336
00:56:28.093 --> 00:56:30.512
missing. Because in reality, when you're looking at

1337
00:56:30.595 --> 00:56:32.472
something, you just expect things to look real

1338
00:56:32.555 --> 00:56:34.099
and when you see it on the wall, you're expecting

1339
00:56:34.182 --> 00:56:38.937
things to look real. But the reality is is that

1340
00:56:39.020 --> 00:56:41.481
there's a lot of cheats that are happening. And you

1341
00:56:41.564 --> 00:56:44.025
need to be aware of those. And if you make

1342
00:56:44.109 --> 00:56:46.653
color adjustments to fix things to make it feel

1343
00:56:46.736 --> 00:56:51.282
visually appealing, the problem is is that these

1344
00:56:51.366 --> 00:56:54.202
cheats that are actually happening are just

1345
00:56:54.285 --> 00:56:57.372
exacerbated, they keep getting worse, you know.

1346
00:56:57.455 --> 00:57:01.960
And so I think you're not starting, you have to

1347
00:57:02.043 --> 00:57:03.962
keep in mind that you're not starting from,

1348
00:57:04.295 --> 00:57:06.714
you know, something that looks real. You need to be

1349
00:57:06.798 --> 00:57:08.800
looking at something and understand

1350
00:57:09.425 --> 00:57:12.470
the fact that what you're cheating already to get

1351
00:57:12.554 --> 00:57:16.808
it to look real is happening, you know. And

1352
00:57:17.267 --> 00:57:19.811
I think that that's a broader thing to think about.

1353
00:57:20.478 --> 00:57:21.771
You know, it's like, oh, we've got one

1354
00:57:22.147 --> 00:57:24.566
light that's on and it's casting some beautiful

1355
00:57:24.649 --> 00:57:26.651
light on something, but we've turned off shadows

1356
00:57:26.818 --> 00:57:28.778
for some objects here and there because it's

1357
00:57:28.862 --> 00:57:31.447
costing too much money. Sorry, too much in

1358
00:57:31.531 --> 00:57:34.450
processor and it can't render correctly on the wall

1359
00:57:34.534 --> 00:57:36.995
because it's starting to get stepy. And so

1360
00:57:37.078 --> 00:57:39.164
then we go, well, there's a shadow missing over

1361
00:57:39.247 --> 00:57:41.583
there. We've done that on purpose so that we can

1362
00:57:41.666 --> 00:57:44.878
actually render it out in real time, you know, and

1363
00:57:44.961 --> 00:57:47.005
that needs to be hidden and translated a certain

1364
00:57:47.088 --> 00:57:49.424
way. And we're doing a lot of these sorts of things

1365
00:57:49.507 --> 00:57:52.510
all the time. And when you integrate sort of

1366
00:57:52.844 --> 00:57:59.100
somebody who, well, people need to be aware of all

1367
00:57:59.184 --> 00:58:01.311
of those sorts of aspects of what's actually

1368
00:58:01.394 --> 00:58:04.898
happening. And then on top of that, add two rather

1369
00:58:04.981 --> 00:58:07.859
than just change in a 2D because of course,

1370
00:58:07.942 --> 00:58:10.486
as a grader, you most of the time, everything

1371
00:58:10.570 --> 00:58:13.656
that you're grading is realistic already is a real

1372
00:58:13.740 --> 00:58:16.451
environment is real, you know, and so you just need

1373
00:58:16.534 --> 00:58:18.203
to keep that in mind. I think.

1374
00:58:18.703 --> 00:58:21.581
Oh, look, that's so true. I mean, in terms of my

1375
00:58:21.664 --> 00:58:24.876
experience grading, I mentioned at the start that

1376
00:58:24.959 --> 00:58:29.797
I didn't realise that it was a 3D background. I

1377
00:58:29.881 --> 00:58:31.758
didn't realise that it was virtual production

1378
00:58:32.300 --> 00:58:36.137
until halfway through the grade because you're 100%

1379
00:58:36.387 --> 00:58:39.849
right. You expect to see reality and you're

1380
00:58:39.933 --> 00:58:43.311
very forgiving of what you're seeing because why

1381
00:58:43.394 --> 00:58:46.356
wouldn't it be real? But then once you start

1382
00:58:46.439 --> 00:58:49.192
looking, you start to see things. And I'm not

1383
00:58:49.275 --> 00:58:51.236
saying that there was anything glaring or major,

1384
00:58:51.527 --> 00:58:53.988
but I just started, you know, OK, so this is

1385
00:58:54.072 --> 00:58:56.074
constructed. I would like to see some evidence

1386
00:58:56.157 --> 00:58:58.701
of that. So I start looking around and then I find

1387
00:58:58.785 --> 00:59:01.246
little things. And then, you know, you start to

1388
00:59:01.621 --> 00:59:04.290
get in there surgically and shape and reduce

1389
00:59:04.374 --> 00:59:06.668
anything that might not look real. And the goal

1390
00:59:06.751 --> 00:59:10.964
is always realism as opposed to some kind of

1391
00:59:11.047 --> 00:59:13.800
stylised treatment or just like you say,

1392
00:59:13.883 --> 00:59:16.761
just changing sometimes as a colourist, you go,

1393
00:59:17.178 --> 00:59:18.930
well, we don't want it to look like the offline

1394
00:59:19.013 --> 00:59:23.268
did, so we better do something. But, you know, that

1395
00:59:23.351 --> 00:59:25.103
might not be the right approach to this.

1396
00:59:25.186 --> 00:59:27.105
Yeah, because I could break it. I could just break

1397
00:59:27.188 --> 00:59:28.648
it like immediately. And you're like,

1398
00:59:28.731 --> 00:59:32.694
oh, actually, now I can see why it's digital now.

1399
00:59:33.361 --> 00:59:36.281
Yeah. So like a light touch and not doing something

1400
00:59:36.364 --> 00:59:39.575
unless you have to would be probably the best way

1401
00:59:39.659 --> 00:59:41.828
forward as a colourist in that scenario

1402
00:59:42.578 --> 00:59:47.458
is not to do anything unless you have to. But

1403
00:59:47.542 --> 00:59:49.711
again, because like I do, you guys add

1404
00:59:50.169 --> 00:59:53.506
this massive add to so I'm not saying sure, but,

1405
00:59:53.589 --> 00:59:55.341
you know, we we couldn't we can always

1406
00:59:55.425 --> 00:59:57.844
do the grade afterwards. I'm saying what what do

1407
00:59:57.927 --> 01:00:01.806
you do during the production? You know,

1408
01:00:01.889 --> 01:00:04.434
how do you add to that process? Because you can

1409
01:00:04.517 --> 01:00:06.561
always grade it afterwards. And I think there's

1410
01:00:07.020 --> 01:00:08.688
like, you know, every time I've ever graded

1411
01:00:08.771 --> 01:00:11.149
something that has been virtually produced,

1412
01:00:11.441 --> 01:00:15.069
whether it's a virtual production or

1413
01:00:15.153 --> 01:00:18.156
a 3D animation or something, you know,

1414
01:00:18.906 --> 01:00:21.701
it always adds to it because it gives it something

1415
01:00:21.784 --> 01:00:24.203
cinematic and it gives it something filmic that

1416
01:00:24.746 --> 01:00:27.457
it just needs to give it coherence and to make it

1417
01:00:27.540 --> 01:00:31.210
feel correct to to be part of our screen language.

1418
01:00:31.878 --> 01:00:35.298
But it's slightly different to the job of creating

1419
01:00:35.381 --> 01:00:37.842
the thing to begin with. It's 2D. It's about

1420
01:00:38.760 --> 01:00:41.179
it's just a different approach, right? Vignettes

1421
01:00:41.262 --> 01:00:45.975
and exposure and, you know, tones in different

1422
01:00:46.059 --> 01:00:49.228
parts of the curve and, you know, generating that

1423
01:00:49.312 --> 01:00:51.439
graceful curve. You're just thinking about a whole

1424
01:00:51.522 --> 01:00:56.361
different bunch of considerations as a colourist.

1425
01:00:56.444 --> 01:01:00.031
Exactly. Yeah. So what happens in the brain bar?

1426
01:01:01.032 --> 01:01:04.535
Well, we've got we've got so we've got a operations

1427
01:01:04.619 --> 01:01:09.040
controller, I guess, is what we call him head of

1428
01:01:09.123 --> 01:01:14.295
operations, Johnny. And he sort of so I guess to

1429
01:01:14.379 --> 01:01:17.173
start from the beginning, I think Unreal has a

1430
01:01:17.840 --> 01:01:22.887
multi-user approach where you have one scene that

1431
01:01:22.970 --> 01:01:25.515
is the environment, for instance, and then you

1432
01:01:25.848 --> 01:01:27.850
have a server that's sort of hosting that

1433
01:01:27.934 --> 01:01:29.769
environment essentially. And then you have these

1434
01:01:29.852 --> 01:01:33.815
multiple machines that are all connected to that

1435
01:01:33.898 --> 01:01:38.194
environment or that scene. And those multi-users

1436
01:01:38.277 --> 01:01:42.073
then can be identified as separate machines that

1437
01:01:42.156 --> 01:01:46.452
exist working on the same scene at the same time.

1438
01:01:46.536 --> 01:01:49.288
So it's kind of slightly different to how we would

1439
01:01:49.372 --> 01:01:51.499
do visual effects projects in that sense,

1440
01:01:51.582 --> 01:01:53.835
because things, you know, when something changes on

1441
01:01:53.918 --> 01:01:56.671
a scene that changes and then, you know, you

1442
01:01:56.754 --> 01:01:58.923
have to publish that change down the line. So in

1443
01:01:59.006 --> 01:02:01.342
this instance, you kind of have a bunch of

1444
01:02:01.426 --> 01:02:04.220
collaborators were working on one scene at the same

1445
01:02:04.303 --> 01:02:07.682
time. And that's part of Unreal Engine's

1446
01:02:07.765 --> 01:02:09.475
architecture. And it's something that's sort of a

1447
01:02:09.559 --> 01:02:12.311
game driven architecture, but it works really,

1448
01:02:12.395 --> 01:02:15.857
really well in this instance. So we have someone

1449
01:02:15.940 --> 01:02:19.694
who's head of operations, and he would be he or

1450
01:02:19.777 --> 01:02:27.410
she would be integral in hosting this scene. And as

1451
01:02:27.493 --> 01:02:30.455
the server, it would also be where all of our

1452
01:02:30.538 --> 01:02:36.127
other sort of peripherals attach to. So like, for

1453
01:02:36.210 --> 01:02:38.713
instance, the virtual, sorry, the motion capture

1454
01:02:38.963 --> 01:02:42.467
Vicon system that then is plugged into its own

1455
01:02:42.550 --> 01:02:45.303
suite, and that has another computer running that.

1456
01:02:45.386 --> 01:02:50.683
And then that is then streamed into Unreal. So that

1457
01:02:50.766 --> 01:02:52.768
gives us the real world position of the

1458
01:02:52.852 --> 01:02:56.606
camera. So then again, in the scene, the operations

1459
01:02:56.689 --> 01:03:00.151
managed would then have this virtual camera that

1460
01:03:00.234 --> 01:03:02.904
we sort of liken to the real world camera. And that

1461
01:03:02.987 --> 01:03:05.198
has these the crown that I don't know if you've

1462
01:03:05.281 --> 01:03:07.283
seen it, but obviously a little, you know, crown

1463
01:03:07.366 --> 01:03:09.952
that exists on the top of the camera, and has

1464
01:03:10.036 --> 01:03:13.581
little dots on top. And when that moves, the Vicon

1465
01:03:13.664 --> 01:03:15.958
system is sort of plugged into Unreal and

1466
01:03:16.042 --> 01:03:18.252
streamed into Unreal, and it tells it where that

1467
01:03:18.336 --> 01:03:20.755
position is in space. And then Unreal,

1468
01:03:21.464 --> 01:03:23.841
in that Unreal scene, we can move that camera

1469
01:03:23.925 --> 01:03:27.678
around in the scene to reposition where we want

1470
01:03:27.762 --> 01:03:30.973
it to be existing. So then those are sort of two

1471
01:03:31.057 --> 01:03:33.392
roles there at that point. So you've got the

1472
01:03:33.559 --> 01:03:37.021
the Vicon operator, and who's also like looking at

1473
01:03:37.104 --> 01:03:38.481
the cameras because they overheat,

1474
01:03:39.106 --> 01:03:41.901
over time, you know, we've got things in the scene

1475
01:03:41.984 --> 01:03:43.653
that are occluding that are in front

1476
01:03:44.070 --> 01:03:46.572
of those cameras that can cause the tracking to

1477
01:03:46.656 --> 01:03:49.116
mismatch, there's quite a few things to keep in

1478
01:03:49.200 --> 01:03:53.120
mind over there. And so the tracking of the cameras

1479
01:03:53.204 --> 01:03:55.498
is super, super important. And so that

1480
01:03:55.581 --> 01:03:59.627
Vicon system, we're constantly, you know, it's

1481
01:03:59.710 --> 01:04:01.963
harsh, because, you know, certainly for the gaffer

1482
01:04:02.046 --> 01:04:03.548
and the grips, you know, where they want to put up

1483
01:04:03.631 --> 01:04:07.093
flags, and they want to put up, you know, like

1484
01:04:07.760 --> 01:04:10.179
big bounce slides, or they want to put a bunch of

1485
01:04:10.263 --> 01:04:13.182
stuff up to also help with the scene. When those

1486
01:04:13.266 --> 01:04:16.269
get in the way of the Vicon cameras, as well, then

1487
01:04:16.352 --> 01:04:18.729
that so we have to work together to make sure that

1488
01:04:18.813 --> 01:04:21.607
that doesn't become an issue. So anyway, so we have

1489
01:04:21.691 --> 01:04:24.902
the Vicon controller as well, and they operate

1490
01:04:25.319 --> 01:04:27.863
independently. And then that plugs into the main

1491
01:04:27.947 --> 01:04:31.409
scene with the operations manager running that.

1492
01:04:31.951 --> 01:04:34.704
And then we have two other, generally, it depends

1493
01:04:34.787 --> 01:04:36.205
on the size of the project, but then we have two

1494
01:04:36.289 --> 01:04:40.334
other positions there where we have the head of the

1495
01:04:40.418 --> 01:04:43.796
virtual art department, so the VAD supervisor.

1496
01:04:45.131 --> 01:04:47.717
And then anytime there's an update to the scene,

1497
01:04:47.800 --> 01:04:51.721
which were at least two objects and color and

1498
01:04:51.804 --> 01:04:53.306
things like that, that we just talked about by

1499
01:04:53.389 --> 01:04:56.100
updating the sand or updating the plants and

1500
01:04:56.183 --> 01:04:58.477
repositioning plants here and there. And we want to

1501
01:04:58.561 --> 01:05:01.647
maybe say, let's move an object in the scene

1502
01:05:01.731 --> 01:05:05.693
quite quickly. That will be taken care of by the

1503
01:05:05.776 --> 01:05:09.363
VAD supervisor, who then repositions all those.

1504
01:05:10.990 --> 01:05:15.077
And then we'll have an extra person that would be

1505
01:05:15.161 --> 01:05:18.456
essentially working independently in another

1506
01:05:18.539 --> 01:05:21.584
environment, and then that might be publishing new

1507
01:05:21.667 --> 01:05:24.837
objects into the scene, like brand new assets.

1508
01:05:24.920 --> 01:05:31.844
And so that would then be published into the VAD

1509
01:05:31.927 --> 01:05:35.681
supervisor's scene, and then he would make a

1510
01:05:35.765 --> 01:05:38.476
decision of when that gets published before we

1511
01:05:38.559 --> 01:05:41.145
change the setup. But of course, you can imagine

1512
01:05:41.228 --> 01:05:44.231
that when you've started shooting something and

1513
01:05:44.315 --> 01:05:46.525
then you add something new to it, and it can affect

1514
01:05:46.609 --> 01:05:49.111
everything. In fact, the optimization, the speed,

1515
01:05:49.195 --> 01:05:51.155
the lighting, absolutely everything. So we have to

1516
01:05:51.238 --> 01:05:54.283
be very careful about that sort of stuff. But

1517
01:05:54.367 --> 01:05:58.537
generally, that's sort of the brain bar and those

1518
01:05:58.621 --> 01:06:03.000
guys. And then I would be in direct contact with

1519
01:06:03.084 --> 01:06:06.545
the operations manager in general and also probably

1520
01:06:06.629 --> 01:06:11.509
the VAD supervisor. And then especially if we do

1521
01:06:11.592 --> 01:06:14.887
lens changes or we make any sort of changes like

1522
01:06:14.970 --> 01:06:18.099
that, or if the motion capture volume is not

1523
01:06:18.182 --> 01:06:21.060
capturing correctly, all these sorts of things,

1524
01:06:21.143 --> 01:06:24.313
then I would then translate all that information

1525
01:06:24.397 --> 01:06:27.900
back to the first AD and/or the DOP and director

1526
01:06:27.983 --> 01:06:31.487
so that we can be sure that before we do a take,

1527
01:06:31.570 --> 01:06:35.408
we've got that take. That take is sort of

1528
01:06:35.741 --> 01:06:39.412
working. And so there's a few steps that we need to

1529
01:06:39.495 --> 01:06:41.872
make sure that we're locked and ready to go

1530
01:06:41.956 --> 01:06:44.750
before we can actually roll on something as well.

1531
01:06:45.876 --> 01:06:48.921
Oh, cool. Wow. That's a pretty big team without

1532
01:06:49.004 --> 01:06:52.800
even thinking about the productions team. Yeah.

1533
01:06:52.883 --> 01:06:55.052
Well, I mean, digitally, yes, you're right. It is a

1534
01:06:55.136 --> 01:06:58.973
biggish team. I mean, also we record every single

1535
01:06:59.056 --> 01:07:01.642
camera as well. So every single camera that we're

1536
01:07:01.726 --> 01:07:04.729
sort of filming, we do an internal recording of the

1537
01:07:04.812 --> 01:07:07.106
3D positioning of that in case we need to

1538
01:07:07.189 --> 01:07:09.275
re-render things later on down the line. And that

1539
01:07:09.358 --> 01:07:11.569
needs to be time coded. We need to have things

1540
01:07:11.652 --> 01:07:14.321
genlocked so that the frame rate is running

1541
01:07:14.405 --> 01:07:17.658
optimally with our motion capture cameras,

1542
01:07:18.284 --> 01:07:21.912
with our actual, sorry, production camera, and also

1543
01:07:21.996 --> 01:07:23.789
that the screens are all locked together

1544
01:07:24.540 --> 01:07:26.917
at the same time. So we have quite a few things

1545
01:07:27.001 --> 01:07:31.130
that can go wrong and people that are in place

1546
01:07:31.213 --> 01:07:36.343
to keep an eye on those particulars. Oh, wow. I've

1547
01:07:36.427 --> 01:07:37.845
got a couple more questions for you. I know

1548
01:07:37.928 --> 01:07:41.140
I've kind of taken up already an hour of your time,

1549
01:07:41.223 --> 01:07:43.517
which I'm very grateful for. 11.08. Okay.

1550
01:07:45.060 --> 01:07:47.563
I would love to know a little bit about your

1551
01:07:47.646 --> 01:07:52.902
background. Okay. What brought you to the

1552
01:07:52.985 --> 01:07:55.821
position that you're in to be somebody who can

1553
01:07:55.905 --> 01:07:59.450
understand how this emerging technology can be

1554
01:07:59.533 --> 01:08:04.288
implemented in this way? What's your history? My

1555
01:08:04.371 --> 01:08:10.336
history? Okay. Well, I was a painter, a traditional

1556
01:08:10.419 --> 01:08:14.423
animation artist. Then I went to digital sculpting

1557
01:08:14.507 --> 01:08:17.718
or traditional sculpting first, then digital

1558
01:08:17.802 --> 01:08:21.055
sculpting, became a character artist way back,

1559
01:08:21.138 --> 01:08:23.390
studied at the Vancouver Film School quite a long

1560
01:08:23.474 --> 01:08:28.187
time ago now, and then was in sort of pre-rendered

1561
01:08:28.270 --> 01:08:31.440
game cinematics back then, doing characters and

1562
01:08:31.565 --> 01:08:34.652
sort of lighting and texture work, and then moved

1563
01:08:34.735 --> 01:08:38.656
into the film side of things, and then became more

1564
01:08:38.739 --> 01:08:43.786
of a lighter, so a 3D lighter and lighting scenes

1565
01:08:43.869 --> 01:08:46.330
for visual effects integration into live action

1566
01:08:46.455 --> 01:08:50.960
plates, and then continued doing that for a little

1567
01:08:51.043 --> 01:08:54.088
while and became a CG supervisor. So I ran

1568
01:08:55.339 --> 01:09:01.053
the floor of the 3D artists and implemented new

1569
01:09:01.136 --> 01:09:05.057
techniques. So back in the day, I was into fur

1570
01:09:05.140 --> 01:09:08.310
and feathers and hair and all these sorts of

1571
01:09:08.394 --> 01:09:10.729
things, and there weren't pipelines in place in

1572
01:09:10.813 --> 01:09:14.233
those days to be able to roll those out that were

1573
01:09:14.316 --> 01:09:17.152
like plugins to make those work, so we had to use

1574
01:09:17.444 --> 01:09:20.197
multiple bits of software to make them work, and

1575
01:09:20.281 --> 01:09:23.242
then I transitioned from that role as a CG

1576
01:09:23.325 --> 01:09:27.204
supervisor, and then I just wanted to, I was just

1577
01:09:27.288 --> 01:09:33.252
more into practical sort of filming as well. So I

1578
01:09:33.335 --> 01:09:36.046
went to AFTRS and studied cinematography, and I was

1579
01:09:36.130 --> 01:09:38.424
already sort of a visual effects supervisor at

1580
01:09:38.507 --> 01:09:41.427
that stage as well. Then I went to AFTRS and

1581
01:09:41.510 --> 01:09:47.600
studied cinematography in Sydney, and I was working

1582
01:09:47.683 --> 01:09:50.144
on film, still work with film back then, which was

1583
01:09:50.227 --> 01:09:54.523
great, and then transitioned to digital cameras,

1584
01:09:55.274 --> 01:09:57.526
like the red one, and you know sort of early days

1585
01:09:57.610 --> 01:10:00.154
with sort of digital cameras back then,

1586
01:10:01.238 --> 01:10:03.949
and then sort of pushed cinematography for a little

1587
01:10:04.033 --> 01:10:05.743
bit, went to a few different studios, it's

1588
01:10:05.826 --> 01:10:10.664
like a combination cinematographer and heading up

1589
01:10:10.748 --> 01:10:14.627
their 3D departments as well, and got into

1590
01:10:14.710 --> 01:10:17.212
sort of virtual cinematography, said doing sort of

1591
01:10:17.296 --> 01:10:19.423
full 3D environments, but you know working with

1592
01:10:19.506 --> 01:10:21.800
the camera because I was quite skilled at that

1593
01:10:21.884 --> 01:10:27.222
point in practical cinematography, and then sort

1594
01:10:27.306 --> 01:10:29.683
of started directing a little bit more, and then

1595
01:10:29.767 --> 01:10:34.813
sort of started my own production sort of company

1596
01:10:34.897 --> 01:10:38.567
as a visual effects production supervisor, where I

1597
01:10:38.651 --> 01:10:41.070
sort of worked mainly on long-form projects,

1598
01:10:41.612 --> 01:10:46.533
doing second unit directing, and just pretty much

1599
01:10:46.617 --> 01:10:49.954
being on set all the time working in visual effects

1600
01:10:50.037 --> 01:10:53.832
in long-form, and pushing motion capture as well,

1601
01:10:53.916 --> 01:10:55.751
and then sort of got back

1602
01:10:55.834 --> 01:10:57.795
into working in Steelbridge

1603
01:10:57.878 --> 01:11:00.965
when we sort of pushed this tech, and I had a LiDAR

1604
01:11:01.048 --> 01:11:03.467
scanning company as well, because I've always been

1605
01:11:03.550 --> 01:11:07.346
into the LiDAR, photogrammetry, drone photography

1606
01:11:07.429 --> 01:11:09.431
as well, and all those sorts of things, so I'm

1607
01:11:09.515 --> 01:11:11.433
creating digital assets, just all sort of work

1608
01:11:11.517 --> 01:11:13.852
together, and it all just sort of put me in the

1609
01:11:13.936 --> 01:11:16.689
right space to be able to sort of do this kind of

1610
01:11:16.772 --> 01:11:19.274
job, which is pretty much all the things you need

1611
01:11:19.358 --> 01:11:23.696
to know sometimes. Yeah, so you've done a very very

1612
01:11:23.779 --> 01:11:27.032
good mix of kind of physical work and digital

1613
01:11:27.116 --> 01:11:29.535
equivalents along the way with the physical

1614
01:11:29.618 --> 01:11:32.496
painting and sculpting to the digital sculpting,

1615
01:11:32.579 --> 01:11:35.708
and physical lighting to digital lighting, physical

1616
01:11:35.791 --> 01:11:39.753
camera to digital camera, so also working in

1617
01:11:39.837 --> 01:11:43.465
sort of emergent technologies as you go. It does

1618
01:11:43.549 --> 01:11:45.926
seem like a pretty pretty special combination

1619
01:11:46.927 --> 01:11:50.764
that can allow you to understand so many things

1620
01:11:50.848 --> 01:11:52.516
that need to come together.

1621
01:11:54.768 --> 01:12:00.441
I mean when you were doing the onset work in visual

1622
01:12:00.524 --> 01:12:02.151
effects, when you're talking about second

1623
01:12:02.234 --> 01:12:08.032
unit VFX shooting, with that you know you've got a

1624
01:12:08.115 --> 01:12:10.743
shark movie and they've shot the principal

1625
01:12:10.909 --> 01:12:13.537
photography, and then they've got a really

1626
01:12:13.620 --> 01:12:16.582
important scene with a shark that needs to

1627
01:12:16.665 --> 01:12:18.667
come up underwater or something, and you're there

1628
01:12:18.751 --> 01:12:22.212
with the sort of gray shark head and capturing

1629
01:12:22.337 --> 01:12:24.882
everything you need to create the 3D shark. Is that

1630
01:12:24.965 --> 01:12:26.633
the sort of thing we're talking about? Yeah,

1631
01:12:27.009 --> 01:12:30.721
exactly, yeah, and I still do that, and that's sort

1632
01:12:30.804 --> 01:12:33.348
of kind of a big part of what I still do,

1633
01:12:33.432 --> 01:12:35.851
so but you know I'd have wranglers that I would get

1634
01:12:35.934 --> 01:12:38.145
to go out and you know do that sort of stuff.

1635
01:12:38.228 --> 01:12:40.606
I wouldn't be the one holding the shark head, but I

1636
01:12:40.689 --> 01:12:43.942
would certainly be the one that decides on how

1637
01:12:44.026 --> 01:12:46.278
the methodology to be able to accomplish the shot,

1638
01:12:46.695 --> 01:12:49.531
and that's why doing sort of long form,

1639
01:12:50.407 --> 01:12:52.201
you know I might sort of you know break down

1640
01:12:52.284 --> 01:12:54.536
scripts and we'll look at what needs to be

1641
01:12:55.162 --> 01:12:57.956
visual effects, traditional visual effects, you

1642
01:12:58.040 --> 01:12:59.833
know and integration in that sense, and then also

1643
01:12:59.917 --> 01:13:03.170
what may require virtual production, and so I'm

1644
01:13:03.253 --> 01:13:05.297
sort of in a good position to also sort of break

1645
01:13:05.380 --> 01:13:07.800
down scripts and look at them because there's you

1646
01:13:07.883 --> 01:13:10.928
know it's yeah there's no as virtual production

1647
01:13:11.011 --> 01:13:15.182
is kind of there's only certain circumstances where

1648
01:13:15.265 --> 01:13:18.102
it really is required, you know in certain

1649
01:13:19.436 --> 01:13:21.939
cases where it's just it just works, you know, but

1650
01:13:22.022 --> 01:13:24.817
in other senses you just need locations,

1651
01:13:25.109 --> 01:13:28.529
you need movement, you need much larger expenses to

1652
01:13:28.612 --> 01:13:29.863
be able to do that, and sometimes you know

1653
01:13:29.947 --> 01:13:33.492
shooting on green screen is the best approach for

1654
01:13:33.575 --> 01:13:35.786
something, and sometimes shooting in a location

1655
01:13:35.869 --> 01:13:39.456
is the best approach for something, and so you know

1656
01:13:39.540 --> 01:13:42.084
there's quite a few reasons why you would

1657
01:13:42.167 --> 01:13:44.461
and wouldn't want to do that, and so that's sort of

1658
01:13:44.545 --> 01:13:47.548
what I'm still doing, so I'd be you know I work

1659
01:13:47.631 --> 01:13:50.050
with big studios as well, and we'll talk about

1660
01:13:50.134 --> 01:13:55.430
whether VPP is an option, but you know inevitably

1661
01:13:56.598 --> 01:13:59.977
I think it's not as mainstream, it is sort of in

1662
01:14:00.060 --> 01:14:03.897
some studios overseas for some instances, but you

1663
01:14:03.981 --> 01:14:07.985
know there's certainly type cases for for why you

1664
01:14:08.068 --> 01:14:09.736
would want to use virtual production, and I think

1665
01:14:09.820 --> 01:14:11.947
for commercials it actually lends itself really

1666
01:14:12.030 --> 01:14:15.075
well because you can just you know bump out heaps

1667
01:14:15.159 --> 01:14:19.204
and heaps of environments within a day, and there's

1668
01:14:19.288 --> 01:14:21.957
just you know the cost works out far better, you

1669
01:14:22.082 --> 01:14:26.044
know, and so you know the quantity, the scale

1670
01:14:26.128 --> 01:14:28.839
economy works just sort of well in that sense too,

1671
01:14:28.922 --> 01:14:32.384
so in that respect it works, but I think you know

1672
01:14:32.467 --> 01:14:35.220
long form again like I said I'm a production

1673
01:14:35.429 --> 01:14:37.931
visual effects supervisor, so I look at those sorts

1674
01:14:38.015 --> 01:14:40.809
of options as well and make those decisions.

1675
01:14:42.269 --> 01:14:46.064
And so you would, if you're on set you would be

1676
01:14:46.148 --> 01:14:49.484
looking at okay you've shot these lenses,

1677
01:14:49.776 --> 01:14:52.696
we need to profile them, and you've shot under

1678
01:14:52.779 --> 01:14:55.532
these lights, so we need to get reflection passes

1679
01:14:55.782 --> 01:14:58.911
and have the balls go out and all of that, and yeah

1680
01:14:58.994 --> 01:15:02.581
do the LiDAR scanning, and I can really see

1681
01:15:02.664 --> 01:15:05.584
how that works right in with doing virtual

1682
01:15:05.667 --> 01:15:07.878
production, but also like it's really great

1683
01:15:07.961 --> 01:15:11.298
that you know that it's not right for everything as

1684
01:15:11.381 --> 01:15:13.550
well, I think that would be, I don't know if

1685
01:15:13.634 --> 01:15:15.802
you were the studio trying to sign off on something

1686
01:15:15.886 --> 01:15:17.971
you don't want the hard sell, you just want to know

1687
01:15:18.055 --> 01:15:20.057
what's going to work the best for the scene that

1688
01:15:20.140 --> 01:15:23.143
you're trying to shoot. Exactly yeah and no one

1689
01:15:23.227 --> 01:15:25.896
really wants to you know put their hand up to

1690
01:15:25.979 --> 01:15:29.399
decide to do virtual production and then it just

1691
01:15:29.650 --> 01:15:31.777
breaks and doesn't work you know either, so I think

1692
01:15:31.860 --> 01:15:36.448
just you know if you're trying to sell a

1693
01:15:36.531 --> 01:15:38.784
technology you know you better be in it in the

1694
01:15:38.867 --> 01:15:41.203
middle of it to really know whether it can be

1695
01:15:41.286 --> 01:15:46.124
done as well and what the obvious

1696
01:15:46.208 --> 01:15:48.126
problems are but also the things that

1697
01:15:48.210 --> 01:15:51.546
the gotchas that can happen and they can happen you

1698
01:15:51.630 --> 01:15:55.342
know very quickly and technology is you know

1699
01:15:56.176 --> 01:15:59.596
is reliable but it's not always reliable, that's

1700
01:15:59.680 --> 01:16:02.057
all I can say, so you have to have contingency

1701
01:16:02.182 --> 01:16:07.020
plans and everything yeah exactly. And so thank you

1702
01:16:07.104 --> 01:16:09.189
for that, the final thing that I wanted to

1703
01:16:09.273 --> 01:16:14.861
ask you about was as a colourist what do you see as

1704
01:16:14.945 --> 01:16:17.948
being the most useful things a colourist can

1705
01:16:18.031 --> 01:16:21.827
bring to a virtual production, whether that's if a

1706
01:16:21.910 --> 01:16:23.870
colourist wanted to get involved during the

1707
01:16:23.954 --> 01:16:28.292
production side or colourists who are working on

1708
01:16:28.375 --> 01:16:30.043
something that's already been shot and edited.

1709
01:16:30.127 --> 01:16:35.465
Oh I mean that's a, I know that's a tricky

1710
01:16:35.549 --> 01:16:38.844
question. Yeah I mean I think for me

1711
01:16:40.429 --> 01:16:44.224
I think like I was I kind of alluded to earlier on

1712
01:16:44.308 --> 01:16:48.770
I think just what LED panels do to skin tone

1713
01:16:49.021 --> 01:16:55.193
is just is really really tricky and is certainly

1714
01:16:55.277 --> 01:16:59.906
something that needs to be you know given its

1715
01:17:00.782 --> 01:17:04.202
you know just the time in fixing and looking at and

1716
01:17:04.286 --> 01:17:07.581
if you as a colourist are and this is now

1717
01:17:07.831 --> 01:17:12.210
posts now at the end of and you know you're sort of

1718
01:17:12.294 --> 01:17:15.422
grading a final spot or whatever

1719
01:17:16.173 --> 01:17:18.383
those are the things you need to sort of kind of

1720
01:17:18.467 --> 01:17:21.261
look at and understand I think so understanding

1721
01:17:21.845 --> 01:17:25.223
maybe the limitations or the things that can cause

1722
01:17:25.307 --> 01:17:27.768
issues within in a virtual production environment

1723
01:17:28.643 --> 01:17:31.730
are things that would be so if you have this sort

1724
01:17:31.813 --> 01:17:34.399
of knowledge and the things to look at in terms

1725
01:17:34.483 --> 01:17:37.027
of trying to fix those are really those would be

1726
01:17:37.110 --> 01:17:40.906
super advantageous for for for you as a colourist

1727
01:17:41.782 --> 01:17:45.577
so that would be you know in my opinion is you know

1728
01:17:45.661 --> 01:17:48.038
that there's colour shift that happens too

1729
01:17:48.121 --> 01:17:49.915
so if you look around certain areas and you'll

1730
01:17:49.998 --> 01:17:51.750
notice that some things are looking greener and

1731
01:17:51.833 --> 01:17:55.921
redder in some areas again those are things that

1732
01:17:56.004 --> 01:17:58.924
aren't real they shouldn't be there and those are

1733
01:17:59.007 --> 01:18:00.634
things you want to sort of grade out so I think

1734
01:18:00.717 --> 01:18:05.305
just understanding what can happen in a virtual

1735
01:18:06.932 --> 01:18:10.394
virtual sort of environment and then looking at the

1736
01:18:10.477 --> 01:18:15.607
the pictures and trying to be critical you know

1737
01:18:15.899 --> 01:18:18.610
not overly critical but I mean critical in that you

1738
01:18:18.693 --> 01:18:20.570
know those are things that if you just

1739
01:18:20.695 --> 01:18:23.615
automatically fix those and hit those sort of bits

1740
01:18:23.698 --> 01:18:26.284
you're in a really good place to then do your

1741
01:18:27.661 --> 01:18:31.123
your creative grading on top of that you know so

1742
01:18:31.206 --> 01:18:32.874
it's almost like the one line which is almost

1743
01:18:32.958 --> 01:18:36.670
like to fix to fix the led issues that can happen

1744
01:18:36.753 --> 01:18:41.425
on the day you know and and for me that would be

1745
01:18:41.508 --> 01:18:44.094
great you know like for a colourist to actually

1746
01:18:44.177 --> 01:18:47.889
understand what could be problematic you know

1747
01:18:48.181 --> 01:18:53.186
and then I think you know moving ahead and and sort

1748
01:18:53.270 --> 01:18:56.481
of being sort of an earlier part of the pipeline

1749
01:18:57.315 --> 01:18:59.818
I think and I've worked with a few colourists that

1750
01:18:59.901 --> 01:19:01.945
have been really interested and just want to be

1751
01:19:02.028 --> 01:19:05.031
there you know and go can we can I just be there

1752
01:19:05.115 --> 01:19:07.367
while you guys are doing some testing and sort of

1753
01:19:07.451 --> 01:19:11.121
maybe we can get some panels you know like some you

1754
01:19:11.204 --> 01:19:13.498
know balls out and see if we can like plug

1755
01:19:13.582 --> 01:19:16.084
into that and just see what we can do with the

1756
01:19:16.168 --> 01:19:19.796
image and and try and understand it I think all

1757
01:19:19.880 --> 01:19:24.509
of those are invaluable for us as well and and

1758
01:19:24.593 --> 01:19:26.887
would be the kind of experience that would really

1759
01:19:26.970 --> 01:19:29.848
help you even afterwards so like post you know in

1760
01:19:29.931 --> 01:19:32.601
the post process just being there on those

1761
01:19:32.684 --> 01:19:35.228
shoots and understanding what's happening is

1762
01:19:35.312 --> 01:19:38.273
actually a really good place to be and then help

1763
01:19:38.398 --> 01:19:42.486
you down the line fixing sort of stuff yeah yeah of

1764
01:19:42.569 --> 01:19:45.697
course I mean when it comes to the skin tones

1765
01:19:46.323 --> 01:19:49.618
we're looking at the light that's reflected from

1766
01:19:49.701 --> 01:19:52.704
the panels that then hits characters here in the

1767
01:19:52.787 --> 01:19:55.957
space right so instead of being hit with like the

1768
01:19:56.041 --> 01:19:59.044
full spectrum of colours in the light there would

1769
01:19:59.127 --> 01:20:02.088
be certain colours that would be enhanced and

1770
01:20:02.172 --> 01:20:05.467
others other parts of the tonal range that might

1771
01:20:05.550 --> 01:20:08.553
not even be there right so you might get like a a

1772
01:20:08.637 --> 01:20:11.431
flatter skin tone instead of like all of the

1773
01:20:11.515 --> 01:20:15.060
variation that you would get under natural light is

1774
01:20:15.143 --> 01:20:19.147
that the kind of issues yeah that and also

1775
01:20:19.231 --> 01:20:21.983
there's certainly a bit of colour shift that can

1776
01:20:22.067 --> 01:20:24.653
happen and there's sort of you know you mentioned

1777
01:20:24.736 --> 01:20:28.114
red skin tones yeah for some people and some will

1778
01:20:28.198 --> 01:20:31.159
just go greenish or not have any change and then

1779
01:20:31.243 --> 01:20:33.870
they've got two people standing side by side like

1780
01:20:33.954 --> 01:20:36.540
for instance in the one job that we were on like

1781
01:20:36.623 --> 01:20:40.627
you had those two people on the beach and the guy

1782
01:20:40.710 --> 01:20:43.463
he kept going too pink and then she had a different

1783
01:20:43.547 --> 01:20:46.049
tone like more melanin in her skin and hers sort of

1784
01:20:46.132 --> 01:20:47.884
almost went green and there was like a shift

1785
01:20:47.968 --> 01:20:49.928
that you needed almost do between the two of them

1786
01:20:50.011 --> 01:20:53.431
and it's when they walk through underneath the top

1787
01:20:53.515 --> 01:20:57.143
panel you if you just move your hand and go along

1788
01:20:57.227 --> 01:20:59.145
there and along there the colour is slightly

1789
01:20:59.354 --> 01:21:01.481
different over top of the person all the way

1790
01:21:01.565 --> 01:21:04.609
through which is problematic very problematic

1791
01:21:04.776 --> 01:21:06.987
you know and we're always chasing that sort of

1792
01:21:07.070 --> 01:21:09.573
stuff as well so that's certainly something to

1793
01:21:09.656 --> 01:21:14.327
so keep an eye on um i would say and and also as a

1794
01:21:14.411 --> 01:21:17.080
as a as a colourist realize that

1795
01:21:18.164 --> 01:21:21.501
you know what's happening is not just an a natural

1796
01:21:21.585 --> 01:21:24.588
artifact it is a mistake well it is something that

1797
01:21:24.671 --> 01:21:26.590
shouldn't be happening if you know what i mean and

1798
01:21:26.673 --> 01:21:28.425
not just accept things that are happening and

1799
01:21:28.508 --> 01:21:31.511
just try and fix them because ultimately you know

1800
01:21:31.595 --> 01:21:34.055
when i'm in the grade or if uh you know

1801
01:21:34.139 --> 01:21:37.684
Colin or the director is in the grade if it's Colin

1802
01:21:37.767 --> 01:21:39.394
it's different but if it's maybe a director

1803
01:21:39.477 --> 01:21:41.980
that's not used virtual production before they

1804
01:21:42.063 --> 01:21:44.065
might miss those sorts of things you know and

1805
01:21:44.149 --> 01:21:46.568
so then it needs to have say the virtual production

1806
01:21:46.651 --> 01:21:51.072
supervisor or somebody that already has an idea

1807
01:21:51.156 --> 01:21:53.533
of what they should be expecting in those instances

1808
01:21:53.617 --> 01:21:56.036
and try and correct for those you know i think

1809
01:21:56.286 --> 01:21:59.122
right so it's just being being aware that things

1810
01:21:59.205 --> 01:22:03.001
aren't natural to begin with i suppose you know

1811
01:22:03.084 --> 01:22:06.171
i'm thinking back to a um a story that a friend of

1812
01:22:06.254 --> 01:22:11.009
mine told about checking renders for quite

1813
01:22:11.092 --> 01:22:15.472
a large 3d film that was very popular and saying

1814
01:22:15.555 --> 01:22:17.515
that the hardest thing about checking those

1815
01:22:17.599 --> 01:22:20.894
renders was knowing that if a character had been

1816
01:22:20.977 --> 01:22:23.438
removed for a scene their matte might still be in

1817
01:22:23.521 --> 01:22:26.274
there so they might be like a little ghost of a

1818
01:22:26.358 --> 01:22:29.444
character like lingering and and anything could

1819
01:22:29.527 --> 01:22:33.281
happen anywhere um in the shot so i suppose you've

1820
01:22:33.365 --> 01:22:35.659
got to be in that mindset a little bit when you're

1821
01:22:35.742 --> 01:22:38.995
grading virtual production if somebody moves

1822
01:22:39.079 --> 01:22:42.957
through light normally as a as a colorist i

1823
01:22:43.041 --> 01:22:46.044
would say well they're moving from you know

1824
01:22:46.127 --> 01:22:49.255
daylight into tungsten we want to see some

1825
01:22:49.589 --> 01:22:52.509
we want to see those because obviously that's been

1826
01:22:52.592 --> 01:22:55.512
done on purpose if it's in front of the camera

1827
01:22:56.262 --> 01:22:58.723
it's meant to be there but you're saying no we need

1828
01:22:58.807 --> 01:23:01.601
to be a bit more critical here we need to say just

1829
01:23:01.685 --> 01:23:03.978
because it's there doesn't mean that it was meant

1830
01:23:04.062 --> 01:23:07.732
to be we need to really interrogate if somebody's

1831
01:23:07.816 --> 01:23:10.819
skin tone changes as they walk from one side of the

1832
01:23:10.902 --> 01:23:13.071
screen to the other that's right is that because

1833
01:23:13.154 --> 01:23:16.116
they're walking past a window or is it because the

1834
01:23:16.199 --> 01:23:19.369
panel was outputting a slightly different shade of

1835
01:23:19.619 --> 01:23:22.997
white there yes exactly and that happens on the

1836
01:23:23.081 --> 01:23:25.875
floor too you know if the floor is in the

1837
01:23:25.959 --> 01:23:28.878
practical floors in shot and then the edges might

1838
01:23:28.962 --> 01:23:33.216
have slightly varying color adjustments on them

1839
01:23:33.299 --> 01:23:36.428
and it's not something we can fix on the day um but

1840
01:23:36.511 --> 01:23:38.596
again look you know panel technology is getting a

1841
01:23:38.680 --> 01:23:41.683
lot better they're introducing white pixels into

1842
01:23:41.766 --> 01:23:44.853
the leds now which really really helps a lot

1843
01:23:45.603 --> 01:23:49.816
especially for overall exposure but also the the

1844
01:23:49.899 --> 01:23:53.987
the angle of the so the

1845
01:23:54.070 --> 01:23:55.905
the field of view essentially

1846
01:23:56.072 --> 01:23:59.659
of these lights is broader and that actually helps

1847
01:23:59.743 --> 01:24:03.121
with a lot of the the color shift as well

1848
01:24:03.204 --> 01:24:05.206
and the tighter pixel pitch actually helps with

1849
01:24:05.290 --> 01:24:08.710
color shift too but then you run into sort of

1850
01:24:09.043 --> 01:24:13.465
issues with brightness levels so i mean the the

1851
01:24:13.548 --> 01:24:15.717
panel technology is getting a lot better

1852
01:24:16.593 --> 01:24:19.763
and so and so that does help with that but i would

1853
01:24:19.846 --> 01:24:23.349
always be on the on the side of trying to keep an

1854
01:24:23.475 --> 01:24:25.435
an eye on that and also you know the panels

1855
01:24:25.518 --> 01:24:27.520
themselves i mean there's a lot of light pollution

1856
01:24:27.604 --> 01:24:30.315
that can that can exist and hit the panels you know

1857
01:24:30.398 --> 01:24:33.651
and so you have a you know you have this

1858
01:24:33.735 --> 01:24:37.238
coating on the the the lights on the leds and the

1859
01:24:37.322 --> 01:24:38.865
better the coating and they've really

1860
01:24:38.948 --> 01:24:41.701
improved their technology in the coating the better

1861
01:24:41.785 --> 01:24:43.620
the coating the less light pollution

1862
01:24:43.703 --> 01:24:46.664
from practical sources or even from other panels

1863
01:24:46.748 --> 01:24:50.001
that you've got in the scene are not lighting the

1864
01:24:50.210 --> 01:24:52.962
the the panels themselves that is then losing

1865
01:24:53.046 --> 01:24:55.173
contrast and when you start losing the contrast

1866
01:24:55.256 --> 01:24:57.509
in your panels the blacks are getting lifted up

1867
01:24:57.592 --> 01:24:59.928
even more and obviously the idea is to retain the

1868
01:25:00.011 --> 01:25:03.264
blacks as black as possible and then when you get

1869
01:25:03.348 --> 01:25:05.433
that light pollution on there we can get shadows

1870
01:25:05.517 --> 01:25:08.019
you know like there's we've added a practical light

1871
01:25:08.103 --> 01:25:11.064
in in the shot and it actually then starts

1872
01:25:11.147 --> 01:25:13.608
showing off the edges of the panelists you know

1873
01:25:13.691 --> 01:25:17.695
because of the the reflected quality of the

1874
01:25:17.779 --> 01:25:20.865
panels themselves the physical kind of panel so you

1875
01:25:20.949 --> 01:25:22.659
know those are the things that you can actually

1876
01:25:22.742 --> 01:25:26.162
end up seeing in the screen but not really realize

1877
01:25:26.246 --> 01:25:28.122
it so you just also need to be aware of those

1878
01:25:28.206 --> 01:25:30.291
sorts of things and look at them and go oh actually

1879
01:25:30.375 --> 01:25:31.626
the blacks are just looking a little

1880
01:25:31.709 --> 01:25:34.128
too milky here for some reason they weren't like

1881
01:25:34.212 --> 01:25:38.132
that in this particular shot just before

1882
01:25:38.675 --> 01:25:40.760
um and it could have been that we added a practical

1883
01:25:40.844 --> 01:25:42.178
light in that instance and that

1884
01:25:42.262 --> 01:25:45.682
is also something to keep a real keep an eye out

1885
01:25:45.765 --> 01:25:50.103
for yeah i suppose um the colorist then can ask

1886
01:25:50.186 --> 01:25:53.731
as well because you do have so many elements that

1887
01:25:53.815 --> 01:25:57.193
can be separated potentially you've got some mattes

1888
01:25:57.277 --> 01:25:59.863
for the foreground and the background and if the

1889
01:25:59.946 --> 01:26:01.364
backgrounds are starting to look a little

1890
01:26:01.447 --> 01:26:04.576
milkier than the black point in the foreground you

1891
01:26:04.659 --> 01:26:07.245
know it'd be much more simple to request a

1892
01:26:07.328 --> 01:26:10.456
matte and pop that in to adjust the background rather

1893
01:26:10.540 --> 01:26:13.585
than try to spend your whole session

1894
01:26:14.210 --> 01:26:18.089
you know doing some complicated animated shape that

1895
01:26:18.172 --> 01:26:20.550
is a nightmare for everyone to sit through

1896
01:26:20.633 --> 01:26:24.137
yeah well do you often have a lot of mattes well

1897
01:26:24.220 --> 01:26:26.014
actually and that's funny you say there's this

1898
01:26:26.097 --> 01:26:28.600
new tech and unfortunately it's only the red camera

1899
01:26:28.683 --> 01:26:30.768
that can actually do subframe and actually

1900
01:26:30.852 --> 01:26:34.022
maybe you know in maybe there's more cameras that

1901
01:26:34.105 --> 01:26:36.858
can do it but the subframe technology is something

1902
01:26:36.941 --> 01:26:39.569
that we're looking into quite heavily at the moment

1903
01:26:39.652 --> 01:26:41.738
which means that your camera can pick up

1904
01:26:41.821 --> 01:26:46.743
a subframe and the Brompton processor can add

1905
01:26:46.826 --> 01:26:51.205
inject a subframe into the panel so you wouldn't

1906
01:26:51.289 --> 01:26:53.666
you can't perceive it with your eyes but

1907
01:26:53.750 --> 01:26:55.293
essentially you're adding an extra frame

1908
01:26:55.376 --> 01:26:58.212
and they have green screen in the background and

1909
01:26:58.296 --> 01:27:00.548
your camera is picking up subframes so

1910
01:27:00.757 --> 01:27:03.551
so you say this for instance is at 25 frames per

1911
01:27:03.635 --> 01:27:04.886
second that you're rolling at

1912
01:27:05.720 --> 01:27:08.932
it's essentially picking out 50 frames a second but

1913
01:27:09.015 --> 01:27:13.186
still interpreting it as a one over 50th sort

1914
01:27:13.269 --> 01:27:17.106
of shutter angle for a 25 frames per second sort of

1915
01:27:17.190 --> 01:27:20.777
film so it's not ultra sharp no it still retains

1916
01:27:20.860 --> 01:27:24.864
its motion blur across the the frames but then you

1917
01:27:24.948 --> 01:27:28.117
can have separate outputs coming into your split

1918
01:27:28.201 --> 01:27:30.954
where you can see straight green screen on the

1919
01:27:31.037 --> 01:27:34.832
panels and that action is happening and then this

1920
01:27:34.999 --> 01:27:37.585
on the other output that's coming out is your

1921
01:27:37.669 --> 01:27:40.129
beauty which is what the panels are looking like

1922
01:27:40.213 --> 01:27:44.008
at the same time which is amazing you know and then

1923
01:27:44.092 --> 01:27:46.260
you can then use that to create your mattes

1924
01:27:46.344 --> 01:27:49.389
from and inject that into your beauty after so you

1925
01:27:49.472 --> 01:27:52.558
always then have a mat running at the same time

1926
01:27:53.017 --> 01:27:55.395
yeah it's amazing i can't see any reason not to do

1927
01:27:55.478 --> 01:27:57.981
that that just sounds so useful i know it does

1928
01:27:58.064 --> 01:28:02.694
it sounds good the problem is is signal and so you

1929
01:28:02.777 --> 01:28:04.946
know we wouldn't have to do it from unreal

1930
01:28:05.029 --> 01:28:07.865
so we're spitting out 25 frames per second which is

1931
01:28:07.949 --> 01:28:10.743
fine or we'd probably go to 50 and then we just

1932
01:28:10.827 --> 01:28:15.373
roll at 25 but then when you inject an extra frame

1933
01:28:15.456 --> 01:28:18.626
you're actually adding so we've got say a 4k

1934
01:28:19.127 --> 01:28:21.379
image that's being projected onto the screen you're

1935
01:28:21.462 --> 01:28:24.674
then running it at 50 hertz or 50 frames

1936
01:28:24.841 --> 01:28:26.884
a second on there so you need to have that much

1937
01:28:26.968 --> 01:28:30.513
more signal going through the panels and so you

1938
01:28:30.596 --> 01:28:33.891
have these bronton processes can only output a

1939
01:28:33.975 --> 01:28:36.936
signal of 4k across these panels and then it's

1940
01:28:37.020 --> 01:28:40.231
all about math to work out you know what the signal

1941
01:28:40.314 --> 01:28:43.484
ratio is through and you want backup and

1942
01:28:43.568 --> 01:28:46.529
you want you know like so it's the power ratio as

1943
01:28:46.612 --> 01:28:50.616
well so we have the ability to have 50 frames per

1944
01:28:50.700 --> 01:28:53.578
second on 4k which is good but if you were trying

1945
01:28:53.661 --> 01:28:55.788
to shoot at a different frame rate and then it goes

1946
01:28:55.872 --> 01:29:01.210
up over a certain frame rate then the panels aren't

1947
01:29:01.294 --> 01:29:03.087
able to run that way so you then need another

1948
01:29:03.755 --> 01:29:07.258
processor which is fine it just implicates the cost

1949
01:29:07.341 --> 01:29:09.093
goes up every time we have more of these

1950
01:29:09.177 --> 01:29:12.847
processes which are very expensive pieces of kit so

1951
01:29:12.930 --> 01:29:15.141
nothing comes for free in terms of options

1952
01:29:15.224 --> 01:29:19.187
no nothing comes for free exactly yeah yeah so yeah

1953
01:29:19.270 --> 01:29:21.939
that's super interesting well as as useful as

1954
01:29:22.023 --> 01:29:24.776
it would be perhaps it can still be somebody's job

1955
01:29:24.859 --> 01:29:28.571
to clip around those elements in roto yeah i mean

1956
01:29:28.654 --> 01:29:33.034
look i think it's actually super super helpful

1957
01:29:33.117 --> 01:29:36.329
and i'm sure there's going to be projects

1958
01:29:36.412 --> 01:29:38.790
where we just go we just need it and and that's

1959
01:29:38.873 --> 01:29:40.666
we'll just you know pay for that extra

1960
01:29:41.459 --> 01:29:45.129
processor and just you know make sure we've

1961
01:29:45.213 --> 01:29:48.049
got the power for it and i think we'll be good

1962
01:29:48.132 --> 01:29:53.721
you know don't take down brisbane's grid yeah well

1963
01:29:53.805 --> 01:29:55.890
we've got at steel bridge here we've just had

1964
01:29:55.973 --> 01:30:02.563
like another we've got like three 32 phase

1965
01:30:02.647 --> 01:30:07.235
outs or ins or whatever so we've

1966
01:30:07.318 --> 01:30:09.654
really like you know bumped it up a little bit more

1967
01:30:09.737 --> 01:30:12.073
so for a small studio so also a good place

1968
01:30:12.281 --> 01:30:15.993
yes also a good place to uh to go and charge a

1969
01:30:16.077 --> 01:30:18.621
tesla if you need to well yeah let's not tell

1970
01:30:18.704 --> 01:30:20.790
too many people none of us are driving tesla's out

1971
01:30:20.873 --> 01:30:22.542
here we should do that it's a great idea

1972
01:30:22.875 --> 01:30:27.130
we'll take that on be a very quick charge yeah look

1973
01:30:27.213 --> 01:30:29.632
thank you so much bryn for your time i

1974
01:30:29.715 --> 01:30:31.926
really appreciate it i think it's a very

1975
01:30:32.009 --> 01:30:34.971
complicated bit of subject matter and

1976
01:30:35.179 --> 01:30:38.057
i'm sure we've only just scratched the surface on

1977
01:30:38.141 --> 01:30:41.435
understanding how it all works but definitely

1978
01:30:41.519 --> 01:30:43.771
know a lot more now having spoken to you about it

1979
01:30:43.855 --> 01:30:45.815
than i did beforehand because it's a totally

1980
01:30:45.940 --> 01:30:49.193
foreign different universe to me so i can't

1981
01:30:49.277 --> 01:30:50.903
thank you enough for explaining some of these

1982
01:30:50.987 --> 01:30:54.031
concepts and i really look forward to seeing how

1983
01:30:54.115 --> 01:30:57.535
people can integrate the the world of color

1984
01:30:57.618 --> 01:30:59.829
grading a little bit more into virtual production

1985
01:30:59.912 --> 01:31:03.958
as as the technology begins to become more

1986
01:31:04.041 --> 01:31:08.754
mainstream and you know more widely adopted yeah

1987
01:31:08.838 --> 01:31:11.257
well it was a pleasure chatting to you Kali so

1988
01:31:11.465 --> 01:31:14.010
you know it was good and i'm glad i mean you know

1989
01:31:14.093 --> 01:31:16.470
like i said you know the more people that

1990
01:31:16.554 --> 01:31:20.558
are on board especially you know colorist as well i

1991
01:31:20.641 --> 01:31:22.351
think it'll just make the the final

1992
01:31:22.435 --> 01:31:24.353
results so much better and i think that's what we

1993
01:31:24.437 --> 01:31:26.898
should all be striving for so happy with that for

1994
01:31:26.981 --> 01:31:30.776
sure absolutely well thank you so much for Mixing

1995
01:31:30.860 --> 01:31:34.238
Light this is Kali Bateman see you next time yeah