Canon colors on latest VP 11

[Jon Shohet]

Hi guys and gals,

I'm running the latest build (595) of VP11 on win7x64.
I'm noticing Vegas does not render colors from my 60D .mov files in a way consistent with other software.

Below is a comparison of snapshots from Canon's ZoomBrowser and the Vegas timeline.

I've also taken the snapshot from the Vegas timeline, and applied a rec709->rec601 conversion to it in Avisynth.
It seems to me that Vegas does not correctly recognize the Canon footage as having a rec601 matrix, rather than a QuickTime/Zoombrowser mistake.

Can anyone confirm/reproduce/shed some light on this behavior?
Many thanks.

[Mike Kujbida]

My guess is that Vegas is limiting the images to a 16-235 range.
Drop the Secondary Color Corrector with the "Studio to Computer" preset on one and see if that fixes things.

[Jon Shohet]

Thanks Mike, but it's not (as far as I can tell) a Studio<->Computer RGB issue, but a Rec.601<->Rec.709 issue. It's especially noticeable on the red bus.

[Jon Shohet]

I ran the original mov through NeoScene HDLink, and imported the CF file into Vegas.

Colors now match ZoomBrowser (Although Vegas does indeed seem to treat the CF file as a "studio RGB" source, while it treats the original mov as a "computer RGB" source).

So it really seems to me Vegas is bungling up the color matrix conversion of the original mov. I'd really appreciate your comments before submitting a support ticket to Sony.

[Jon Shohet]

I'm really surprised by the lack of responses, as I'm sure there are many here with the Vegas Pro 11 / Canon EOS combination.

Can someone confirm that he's using Canon mov natively on the Vegas timeline and NOT seeing any color shift issues?

I'd really appreciate it if someone who also has CineForm could put a native Canon mov and a CF version of the same file side by side on a Vegas timeline and tell me if he's seeing a color mismatch.

Thanks again.

[Mike Kujbida]

Jon, I know you're using Pro 11 but Glenn Chan's article titled Color spaces and levels in Sony Vegas 9 and 10 may be of some help.

[Jon Shohet]

Thanks for chiming in again Mike :)

I've read most of the articles on Glenn Chan's site, as well as anything else I could google my hands on.
In fact, I used the calculations made by Jon Fairhurst on this old thread http://www.dvinfo.net/forum/what-hap...601-clips.html to create 601<->709 channel blend presets.
So I can indeed match the colors of the native .mov and the CF file on the Vegas timeline.

However - who's getting it wrong? Vegas, or Quicktime/Cineform/Zoombrowser?

I was under the impression that Vegas, Cineform and Zoombrowser do not rely on QT or any global system codecs when decoding the Canon mov, all them using their own internal decoders.
If so, it seems logical that if they do not all decode the same colors one of the internal decoders is getting it wrong.... and if it's Vegas getting it wrong, that a really big blunder!

Of course maybe I'm completely missing something very obvious here, that's why I really hope someone else can report how his system behaves and if it's consistent with mine.

Cheers.

[Jon Shohet]

I at last got some help from David Newman over at the CineForm Forum.

http://www.dvinfo.net/forum/cineform...-neoscene.html

It definitely appears as though Vegas is incorrectly treating the 60D footage as 709 instead of 601.
Now the question remains if it's a quirk on my system or a problem with Sony.

I'm going to contact Sony support. I still hope that any Vegas 11 Pro + Canon DSLR user out there can check to see if this color shift issue is present on their system as well.

[Craig Longman]

I've just recently moved from an XL-1 to a T2i and will do some tests this week to see what I'm seeing. Mind you, I'm on VMS v11, not Pro, but would still be interesting to see. Glad you were able to get some info on all this; CineForm in general, and David in specific have always seemed to be very helpful.

CraigL

[Jon Shohet]

I submitted a question to SCS customer support regarding this issue. This is the first time I have tried contacting Sony support.
It's been 3 weeks now and still no response. Is this typical?

[Roy Feldman]

In another three weeks you'll get a message thanking you and not answering your question.

[Gene Gajewski]

Here's something that'll hurt your your brain...


Canon DSLR video uses BT.601 SD matrix instead of BT.709 HD ?! Jorgen Escher

[Jon Shohet]

Gene, thanks for the link.

The fact remains that other software, like CineForm and Adobe, for example, correctly read the footage as rec601 and automatically handle the 601<->709 transformation, while Vegas ignores the 601 flag and incorrectly assumes it is 709, introducing color shifts.

As long as no else can confirm this on their system, I cannot be 100% sure this is not an error on my specific machine. However, since Vegas AFAIK is using it's own built in AVC decoder and does not rely on any externally installed codec, it seems to me highly unlikely that this is an issue only on my machine.

If the standard customer support route is inefficient, does anyone have any other suggestion as to how to bring this issue to Sony's attention?

[Gene Gajewski]

If we go by that link I posted, it would seem that the vendors had to make very specific exceptions in their codec for Canon footage, since it's flagged as 601, but uses the 709 chromaticities. A straightforward decoder would normally follow the 601 flag unquestioned.

My guess is SCS simply dropped the ball on hacking up their codec to support Canon (due to its popularity)- not that Canon doesn't deserve a brickbat in the face for doing so.

Here's a quote from wikipedia explaining that 601-709 conversion is more than just switching matrices.

"The definitions of the R', G', and B' signals also differ between BT.709 and BT.601, and differ within BT.601 depending on the type of TV system in use (625-line as in PAL and SECAM or 525-line as in NTSC), and differ further in other specifications. In different designs there are differences in the definitions of the R, G, and B chromaticity coordinates, the reference white point, the supported gamut range, the exact gamma pre-compensation functions for deriving R', G' and B' from R, G, and B, and in the scaling and offsets to be applied during conversion from R'G'B' to Y′CbCr. So proper conversion of Y′CbCr from one form to the other is not just a matter of inverting one matrix and applying the other. In fact, when Y′CbCr is designed ideally, the values of KB and KR are derived from the precise specification of the RGB color primary signals, so that the luma (Y′) signal corresponds as closely as possible to a gamma-adjusted measurement of luminance (typically based on the CIE 1931 measurements of the response of the human visual system to color stimuli)"

[Mike Calla]

Don't worry guys. You are not crazy.

This is known issues dating back to the very first weeks of 5MKII footage. You can easily see it when viewing a blocky histogram of footage.

The problem is there but its not a huge issue unless its some mission critical footage.

In the end make sure your levels are fine with scopes and that everything is looking good on your preview monitor and you'll be fine.

...been cutting with AVID side by side with Vegas for two years, and its never been a problem.