-
I should know this… YPbPr vs. RGB
Ok… I should know this – but I don’t. It all started with the two editors across the hall. One is outputting a quicktime for the other one to use; the video levels are fine for the output editor, but too black for the recieving editor. Well, obviously this doesn’t make any sense and has led to all of us sitting around one workstation with pensive looks on our faces trying to figure out why the levels would be percieved differently by the two different machines (levels observed from external waveform monitors supposedly calibrated within a hair of each other). Well, everyone starts rooting around and pretty soon we all notice that our video playback settings in FCP are all different from each other – maybe this has something to do with the problem? We all use the AJA Io. Editor A (the output editor in this case) has his playback setting at Uncompressed YPbPr SMPTE N10 – 8-bit. Editor B (recieving editor) is set at Uncompressed YPbPr Betacam 525 – 8-bit; and I seem to be set at Uncompressed RGB 8-bit.
#1 – Does any of this have to do with the percieved difference in video levels of the same file??
#2 – What’s the difference between all this YPbPr and RGB stuff? (I hate to word that so blatantly ignorantly.) Can anyone point me to some reading? I’ve tried to do some research but don’t seem to be hitting upon any clear explanations…
Nick
-
10 Replies
-
Is it black level or gamma? What does it look like on the FCP scopes? I think there’s luma shift issues with 10bit codecs – this keeps getting fixed and broken with FCP and quicktime updates.
Graeme
– http://www.nattress.com – Film Effects and Standards Conversion for FCP
-
It’s black level. The same spot on the timeline seems to measure the same with the internal FCP scopes…
Nick
-
It’s the same on internal scopes? Different on external? I’m not getting what’s happening here.
Graeme
– http://www.nattress.com – Film Effects and Standards Conversion for FCP
-
Apparently what’s happening is we’re trying to figure this thing out with part fact and part assumption – not a good idea. I’m not sure anymore what the external scopes are saying – I thought they were reading differently but apparently nobody’s actually checked them yet?? I’ll get back to you… (This may all boil down to someone’s subjective eye – or a difference in the actual monitors – I don’t know.)
Nick
-
Check the “setup” settings in the IO control panel while you’re on.
Graeme
– http://www.nattress.com – Film Effects and Standards Conversion for FCP
-
[Nick Ryan] “Editor A (the output editor in this case) has his playback setting at Uncompressed YPbPr SMPTE N10 – 8-bit. Editor B (recieving editor) is set at Uncompressed YPbPr Betacam 525 – 8-bit; and I seem to be set at Uncompressed RGB 8-bit.”
I’m thinking that’s at least part of your problem. There is certainly a difference between RGB an any YPbPr color space. I think I heard someplace (AJA tech support, perhaps?) that there should be a noticeable gamma shift between the SMPTE and Betacam versions of YPbPr, as well.
I know you said the external scopes were calibrated. Were the video monitors?
Arnie
https://www.arniepix.com -
As much as I hate to quote such a crabby guy, Bob Zelin over on the AJA IO forum has this to say about SMPTE vs Beta:
>>>>Never ever ever use SMPTE N10. I do not know why AJA included N10 as a choice, or why anyone gives a crap about what the SMPTE committee says. Sony and Panasonic make the rules. Not SMPTE. And we have seen historically that Sony does not follow SMPTE’s standards, in everything from time code, to video levels.
The only time you will EVER use SMPTE N10 is if you have a SDI waveform monitor like a Leader LV5100D or a Tektronix WFM601. These products DECODE a SDI signal into analog component, using the SMPTE N10 format. The output appears on the 3 BNC connectors on the back of these scopes, labeled PIX MON OUT. When you hook up your Sony PVM series monitors to the PIX MON OUT of the scopes, you set your Sony monitor menu’s to SMPTE N10, or the colors will not be correct. I have NEVER seen another application of SMPTE N10 in all these years.<<<< Do a search (SMPTE) and include the AJA forum, there are a ton of hits. Ed -
[Arniepix] “There is certainly a difference between RGB an any YPbPr color space. I think I heard someplace (AJA tech support, perhaps?) that there should be a noticeable gamma shift between the SMPTE and Betacam versions of YPbPr, as well.”
Ok… we’re still working on this (slowly – in the midst of working on everything else that has to go on as normal). Quick question: does the playback setting (beta vs. SMPTE) affect the way a quicktime is rendered out? It shouldn’t right? I should only affect viewed playback of said file. The further into this thing we go the more confused we get…
Nick
-
Quicktime files are rendered with the settings you chose when you generate the file. The playback settings don’t enter into the QT movie settings when the movie is made.
Doug Olin
-
Ah, I thought so. Thanks!
Nick
Reply to this Discussion! Login or Sign Up