-
Converting 8 bit to 10 bit
Cows,
We have shot some footage on an HPX-170.
It is 1080 24PA. 8 bit, 1280×1080 (1.5 pixels). DVCPROHD.
I have log and transferred it into FCP7. I have removed the advanced pull down.
We are adding the video footage to a film that was shot on 16mm and is now 1920×1080, 10 bit. ProRes
I am converting the video footage to 1920×1080 (square pixels) in Compressor. But I also want to convert it from 8 bit to 10 bit simply so that all the clips end up all the same.
I can’t seem to do it in Compressor (maybe it’s possible, I just don’t see and option to do it). Any ideas how to do it? Can I do it in FCP or in Sorrenson Squeeze maybe?
Thanks much
-
10 Replies
-
If you convert the DVCPRO HD footage to ProRes…it will then be 10-bit. No version of ProRes is 8-bit
Shane
Little Frog Post
Read my blog, Little Frog in High Def -
[Shane Ross] ” No version of ProRes is 8-bit”
I guess that’s why I could not find a setting for 10 bit 🙂
Thanks! I didn’t know that about ProRes!
-
-
[Rafael Amador] “Check “Frame Control ON” in Compressor”
If I leave it off will it stay at 8 bit?
-
[Chris Jones] “If I leave it off will it stay at 8 bit?”
If you leave it OFF you will get a 10b Prores file, but all the process is done in 8b.
Only with the Frame Control ON, Compressor process in 10b Floating point.
rafael -
Good to know! Thanks!
-
I have a follow-up question:
I just did a test converting to ProRes4444 (frame controls ON). I opened the clip in DaVinci Resolve and looked at the clip details.
It says the clip is 16bit.
Is ProRes4444 always 16 bit? Or do you think 16bit might be how Resolve says in shorthand 10bit floating point?
Thanks
-
Up to Apple, PR444 “supports'”12b picture and 16b Alpha channels.
When is 10b, when is 12b or when is 16b?
No idea.
As I stated many times in this forum, the PR444 interface is, at least, incomplete.
You can’t really chose the bit depth, neither you can chose the YUV/RGB mapping (Full range/video range).
rafael -
Confusion about what constitutes the bit-depth of a container?
Consider how computers process data. Binary, octal, hexadecimal. Base two, Base eight, Base sixteen.
If you do not understand binary code wording, then whether something is “8-bit” or “10-bit” or “12 bit” or “16 bit” or “32-bit” or “64-bit” would be mysterious. They are all “words” or “bytes” that contain a value. Whether a particular value is carried in large-enough byte container does not alter its meaning, so “converting” an 8-bit data packet to fit in a “10-bit” container, which obviously has to be 16-bit or higher, is meaningless. It also does not change its intrinsic resolution as it was originally encoded.
Its as silly as demanding 4:4:4 output from a sub-Nyquist codec source.
jPo
“I always pass on free advice — its never of any use to me” Oscar Wilde.
-
I’ve heard that in video and audio manipulation that “damage” or maybe “distortion” is a better word – occurs.
For example when you push the gamma up you get noise.
I heard – maybe incorrectly – that if you increase the bit depth of video from 8 bit to 10 bit, that while you are not increasing the true numbers of colors – you are giving more room for the “damage” done to the video to be less severe.
Along the same lines as why people record at 96KHZ. The true KHZ from a 96KHZ sample is half that so 48Khz, but the human ear can only detect up to 22khz. But after heavy manipulation in post the quality is reduced but you can still in the end reduce what you are left with to 48khz (or 24khz really).
I’m not sure how much, or if any of this, is true.
Reply to this Discussion! Login or Sign Up