Hey Todd,yeah, thats how i even came to the forum, through that post. But unfortunately this is not related, we have no glue tools installed and it seems to be maybe a color or red footage problem, not sure, we are stil trouble shooting.The weird thing is, that when we export the DPX sequence from Color and import it straight back into color it is a one to one match - perfect. Then we open the same sequence in AE cs5 and it is way darker, like a .4 gamma drop. Now, we already knew about this whole problem when rendering to prores 444, but never had teh problem when going to dpx.I wonder if that could be red footage related, since we are working with native red files within color, maybe thats the caveat. We can render out to prores 422 and import into AE cs5 and it matches perfect. Maybe there is something wrong with teh way that color writes teh dpx sequence. I will try a cineon sequence now.Any other ideas? Whats the change to the dpx import that AE CS5 does? I read somewhere that somehow CS5 now corrects the old Apple 1.8 gamma value?Thanks,Martin
a suggestion from one of our engineers:
"Have them play with the Cineon/DPX intent settings (i.e. Full Range or Not). It seems like our default is different from RED and it confuses people."
thanks for the suggestion, but thats not it. We are aware of the import settings. We are using DPX workflows since years on numerous films and never had a problem. I do have the feeling though, that the problem might be more Color and Red related.
We tried a numerous roundtrips from DPX sequences originating from a flame and other compositing systems and they all work flawlessly. Just the roundtrip with color is off and it seems with dpx sequences that are generated from original r3d source files.
We are kind of under the gun in terms of timing, so we exported to uncompressed 10bit quicktimes, which seems to work (even though it seems whatever we export from color - it always looks different in AE - but at least when we render the comps out of after effects and reimport into color it matches back when we use the 10bit uncompressed codec).
All other codes (like Prores 422 or cineon and DPX) are giving completely different gamma results in After effects. How is that possible? Could that be a screw up from Color, rendering out different gamma values to every different codec, or is it something between AE CS5 and Color translating the different codec with different gamma values?
We would love to figure this one out, since it is important for our workflow to get the same in and outs in our roundtrip workflow.
Sorry for the delayed response. I haven't looked at Color in a while, so I'm afraid I don't have any new information for you.
I'll definitely look into this issue in the near future and post my response here.
AE QA Engineer
I just want to iterate, we're seeing the same problem. Exporting a DPX sequence out of Apple Color 1.5 from R3D footage results in a gamma shift when viewed in AE CS5 and AE CS4. I'm trying to get my hands on another DPX player to see what that produces....
it does seem that changing the "PRINTING DENSITY" within the COLOR project settings to "VIDEO (64 black - 940 White : Linear)" creates DPX files that match in COLOR and AE CS5.