Converted footage darker on canvas when playing (?)

I ripped some content off youtube and transcoded the clips via streamclip to quicktime for FCP, only i notice the footage is darker on canvas when playing, then when you pause the playhead the clip goes back to it's 'normal' initial grade.

The difference is akin to crushing the blacks slightly and/or lowering the mid-tones. Any ideas as to the cause of this or of any workarounds, amendments?

Cheers,
Bluey

You can try using QT Edit to edit the gamma flag in the clip. QT Edit is found in the ProMedia Tools suite.

[www.digitalrebellion.com]



www.strypesinpost.com

But why is the gamma different when running and when paused? Could this happen with the sequence set to an FCP-happy codec and the rendering full?
After full rendering, if the gamma looks wrong both when running and when paused, then I'd alter the gamma within FCP using the gamma filter and render again.

Dennis Couzin
Berlin, Germany

i have had this behaviour before, but have refrained from answering as i have no idea what the issue is!

well a little bit of an idea,
as both times it has a involved non-suported codec.

once, i was running ProRes on a pre ProRes FCP
i had the later version of FCP installed, but had kept a project going on the previous version of FCP.

now i have having that same issue with DNxHD. These clips are MXF files rewrapped into QT files.


so a small idea about the issue, but none about a solution.



nick

>as both times it has a involved non-suported codec.

I got it a couple of times with a Color rendered sequence. It's ProRes, and presumably it didnt write the gamma flags properly.

Yea. Don't ask why. There is another bunch of QT related gamma questions that I have.



www.strypesinpost.com

Bluey described the colors as abnormal during running. So, does the FCP canvas ignore the gamma atom when paused and use the gamma atom when running? What horrible NLE software would do that?

Is there a gamma atom in these QT files? Way back in 1999 an Apple Whitepaper deprecated the gamma atom like so:

Quote

The 'colr' extension supersedes the previously defined 'gama' ImageDescription extension. Writers of QuickTime files should never write both into an ImageDescription, and readers of QuickTime files should ignore 'gama' if 'colr' is present.

The latest 2012 QuickTime File Format Specification repeats those exact words, and the 'colr' atom is now always present. The 2012 Specification does describe the 'gama' atom, but only in conjunction with "color table atoms":

Quote

These atoms are placed after the color table, if one is present.

Quote

Color table atoms define a list of preferred colors for displaying the movie on devices that support only 256 colors.

256 colors is not our situation. So why are any applications writing any gamma atoms in our QuickTime files?
Jon Chappell can enlighten us on this.

Dennis Couzin
Berlin, Germany

Interesting discussion, i'll look into it further and get back,

Bluey,

I had transcoded the clips to a different aspect ratio other than the original format. Reconnecting the media to versions which have been re-exported to match the initial aspect ratio, appears to have has solved the issue.

Although it seems others have had the same issue for different reasons, if the issue was to crop up again in another circumstance it'd be good to know of an analysis.

Bluey,

just had this happen again, this time with what should be perfectly formatted ProResLT files.

in fact, they play fine most of the time, but i just switched my canvas bg to something other than black, and the clip went dark,
and pops back to normal when i play.

only happens if the clip has a motion effect on it.


nick

Does it improve if you set your renders to do 'full' render, then render it? I found sometimes this stopped cross dissolves popping, even when they were supposed to be rendered.

perhaps more importantly, does the tone discrepancy transfer to the export? is it noticeable in the final file?

Bluey,

I had this problem years ago. It would change the gamma of the clips when in play mode, and have a different
gamma when in pause. I cured it, in this situation anyways by applying a Broadcast safe to the clips.

But this was years ago and may not apply to your situation.




dcouzin Wrote:
-------------------------------------------------------
> But why is the gamma different when running and
> when paused? Could this happen with the sequence
> set to an FCP-happy codec and the rendering full?
> After full rendering, if the gamma looks wrong
> both when running and when paused, then I'd alter
> the gamma within FCP using the gamma filter and
> render again.
>
> Dennis Couzin
> Berlin, Germany

The final file will be fine.



www.strypesinpost.com