[David Margolis]

Just to say after many many hours and sleepless nights, I’ve found the only possible solution I can come up with in the time I have.

That is to output my ProRes 444 timeline as 422 HDCAM SR. All the problems just vanish. Shame to lose all that info but fortunately the broadcast spec allows for 422.

No response from AJA yet but I’ll be curious to know what they suggest.

[David Margolis]

Gary, thanks for this. The deck is SRW 5500.

And I do have the latest version of Snow Leopard with all the updates.

So I got my tape back from the QC facility. Now I’m really confused. I view the colour bars on the tape looking at external scopes. The facility are correct. The bars (100% and generated from FCP) have blacks at -10. Bad. However, I capture the tape and PING everything looks normal in FCP. Apart from a slight gamma shift which was caused by the YUV incorrect setting.

So, I’ve changed everything to RGB. I’ve output to tape again. Settings on the deck are 24PsF 1080 4:4:4 SQ HDCAM SR. I’m going dual link into the deck from a Kona card which is set up correctly. Black and burst is set up correctly. I’ve been through all my settings in FCP and everything is correct. However, I output my bars again. They look fine on the scopes. Then I play back on the tape and everything is still illegal. Black are still at -10.

This must be something the deck is doing. I’m not all that familiar with the SRW-5500. Does anybody know what might be causing this?

David

[David Margolis]

Hey Garry. Thanks for responding. The pipeline has been as follows…

REC709 TIFFs from DI facility

TIFFs then went to DCDM for a DCP encode and also to ProRes444 (all from Shake)

I can absolutely confirm that the ProRes 444 output from Shake was identical to to the original TIFFs. I did an A/B check on them within Shake. No difference.

Prores settings were – 24p, 1920×1080. They went into a matching sequence in FCP.

The only thing I changed in the sequence setting was to render using YUV which I think is what caused the problem. When I switch YUV on the gamma is up too high. When I turn it onto RGB everything looks correct. And I’m talking here about FCPs output, not the internal viewer. Everything looks normal in the viewer no matter what video processing setting I use.

Of course, until I get the tape back I can’t fully confirm this. But it seems the most likely explanation to me. The digibeta that passed QC (rendered from the same ProRes source) was set to RGB video processing.

My understanding of dual SDI setup is that so long as your deck is set to receive 444SQ, you can only really go wrong by having the A and B SDIs the wrong way round. Or having one not plugged in. And when either of those happen, it’s pretty obvious it’s not working correctly.

So that’s why I’m thinking gamma. Then again, the QC guys reported sub blacks. Can incorrect gamma cause this?

[David Margolis]

The QC facility decided not to go ahead when they realised that even the colour bars were illegal. They were kind enough to give us another chance to submit something. So as soon as I get the tape back from them I can check.

As far as I can tell, this is a YUV / RGB thing. The moment I put the timeline into YUV everything goes bright.

From what I’ve read, this is a known issue with Prores 444.

[David Margolis]

Hi. Unfortunately we don’t have external scopes (bad I know) But I think you hit the nail on the head. It is a gamma issue. Before we sent the tape off we captured a small clip from it. Comparing it to the FCP timeline it’s brighter. Not by a lot. But perhaps enough to mess up all our broadcast safe levels.

I changed video processing to Always Render in RGB and that seems to have fixed it.

That’s really annoying! Why is FCP shifting the gamma in this way?