how to export 16 bit color

[Bj Ahlen]

From Apple’s FCP7 specifications:

Import OpenEXR files (float processing requires a graphics card with 128MB of VRAM; 256MB or more recommended)

Export as 16-bit or 32-bit floating-point OpenEXR format (float processing requires a graphics card with 128MB of VRAM; 256MB or more recommended)

[Rafael Amador]

[B.J. Ahlen] “mport OpenEXR files (float processing requires a graphics card with 128MB of VRAM; 256MB or more recommended)

Export as 16-bit or 32-bit floating-point OpenEXR format (float processing requires a graphics card with 128MB of VRAM; 256MB or more recommended) “

These are the Motion 4 Specs.
Scroll up to see the FC I/O supported formats.
8/10b YUV and 8b RGB. Nothing new.
rafael

http://www.nagavideo.com

[Dale Larsen]

You guys have taken this way over my head.

can anyone dumb this down for me?

I am exporting from FCP. what codec allows me to adjust color depth? I can’t find any.

The client seems to think other vendors using FCP have delivered masters that are correct. Question is how that is possible?

I think the client is misusing the terminology, claiming that a DV/DVC Pro ntsc master is 32 bit.
(not that I understand what is going on here)

Dale

[Rafael Amador]

[Dale Larsen] “I am exporting from FCP. what codec allows me to adjust color depth? I can’t find any. “
You do not adjust the bit depth on any codec. You chose a codec with a certain bit depth.
Video codec are 8 or 10b, YUV or RGB.
That makes 24 or 30 bits to represent a full pixel.
If you want you can add an Alpha channel (no much sense for delivery).
You add 8 or 10 more bits for the Alpha so you end up with 32 or 40b to represent a full pixel with Alpha.

SO the only way to export from FC something that you can call 32b, is to export in any of the available 8b codecs that may support Alpha.
In QT you have: Animation, JPEG-2000, NONE and Planar RGB.

[Dale Larsen] “I think the client is misusing the terminology, claiming that a DV/DVC Pro ntsc master is 32 bit. “
If you want something like DV or DV50 in 32b, the only option I know are the AVID codec which being 8b, supports Alpha.
Rafael

http://www.nagavideo.com

[Jeremy Garchow]

I would download the Avid codecs and give him am Avid DV file.

Rafael has it right. Each color (R, G, B) gets a bit depth of 8 bits in most video codecs. If you add transparency (alpha channel) that is another 8 bit channel, so adding those all up, 8×4=32 bits total. All Avid codecs support an alpha channel, does your program have any transparency?

[Gary Adcock]

[Uli Kunkel] “Did you by chance see my last post in the FCP forum about outputting 4:4:4 RGB from with the Kona 3? “

No
repost it in the Kona Forum- 444 stuff gets lost in the noise.

ga

gary adcock
Studio37
HD & Film Consultation
Post and Production Workflows for the Digitally Inclined
Chicago, IL

https://blogs.creativecow.net/24640

[Matthew Bradshaw]

If you have fcp7 you could get him to supply you with a short clip in the codec he prefers and then try using the “copy codec” function in compressor. I forget exactly how you do it, drag the clip onto a newly created custom setting I think.
Matt.

[Arnie Schlissel]

[Dale Larsen] “The client seems to think other vendors using FCP have delivered masters that are correct. Question is how that is possible? “

I think that you will need to ask your client.

What format masters did these other vendors deliver. Deliver that and be done with it.

If he insists that these were 16 bits per color & HD, but were delivered as DV/DVCPro, then your client is misinformed. That’s his problem, and either you can take the time to educate him, or you can simply give him what other vendors have given him in the past.

Arnie
Post production is not an afterthought!
https://www.arniepix.com/

[Dale Larsen]

Thanks everyone,
this gives me a lot to go on.

I will post a resolution when I finally solve it.

Dale

[Bj Ahlen]

[Rafael Amador] “Scroll up to see the FC I/O supported formats.
8/10b YUV and 8b RGB. Nothing new. “

Ah, good catch.

Now if someone could explain this from Apple’s ProRes White Paper:

Apple ProRes 4444 supports image sources up to 12 bits and preserves alpha sample
depths up to 16 bits.
[…]
Like Apple ProRes 4444, all Apple ProRes 422 codecs can in fact accept image samples even greater
than 10 bits…
[…]
The Apple ProRes 4444 codec losslessly encodes alpha channel values of any bit depth up
to and including 16 bits.

So in FCP, it’s up to 12 bits in and up to 10 bits out, except for the up to 16-bit alpha channel in ProRes 4444?