Imac i7 playback settings

[Scott Sheriff]

Matt,
“I’ve rendered the sequence and then all works fine but then any small adjustment of the PSD and you get the same message when the playhead hits the file and again, you have to render.”

Well, since you didn’t post the particulars of your source material and sequence settings, I’ll have a guess:
A. The background and foreground layers are different frame dimensions and codecs.
B. The background and foreground layers don’t match the sequence settings.
C. The .psd file is larger (in frame size) than it needs to be.
D. All of the above.

Scott Sheriff
Director
SST Digital Media
https://www.sstdigitalmedia.com

[Bret Williams]

He shared plenty. It’s a psd (always needs rendering) layered over a motion file (probably project file which could include 99 layers of who knows what) over a pro res video clip. The sequence plays fine until it hits the psd, so I’m guessing the pro res matches the timeline just fine.

But the simple answer is yes, you always have to render psds. Always. For true full res non dropped frame playback. If you want to push FCP to playback as much as possible turn on unlimited rt (which you have) and turn off the dropped frame warnings. Those are fairly useless unless you’re outputting to tape anyway. They’re nothing more than annoying. So when you playback, if the psd/motion section isn’t rendered, it might be able to stutter through and show you a decent approximation. If the psd is just huge and/or the motion file is too hefty, it might just show you a still until it passes that section.

You can try other settings like dropping the playback frame rate to half and the resolution to half or dynamic, etc. I haven’t had much luck with the dynamic thing though.

[Matt Surf]

Thanks for the advice guys. The workaround I’ve found is to save the PSDs as JPEG’s out of PS which seems to be much better (not 100% sure what impact this will have on the final film).

For more info; the sequence settings are 1920×1080 HDTV 1080i (16:9)
on V1 I have the footage imported using XDCAM 1080i50
on V2 I have my Motion file at 1440×1080 using codec Apple Pro Res 422
On V3 I now have my JPEG.

I am also now entering text into this equation through FCP which again just throws up a load of issues regarding rendering and any tiny change to the file needing to be re-rendered to preview.

Apart from what you’ve mentioned above – is there anything else I can do to improve playback or are these the infamous FCP shortcomings I read so much about before buying the Mac?

Cheers

[Bret Williams]

Shortcomings???? You have the 3 hardest things for a nle app to handle on top of one another. A uncompressed photoshop file (which is probably 20 times the size of a tiff or jpeg of same), a Motion file that isn’t even the same dimensions as your sequence (which as I said could be anything – layers of particle generators for all I know. How well does Motion play it back in real time?), and XDCAM long GOP format video which is already a huge nemesis for FCP and doesn’t match your sequence settings either. Nothing in your sequence matches your sequence.

Fro the utmost performance I would render out my motion project as a pro resfile 1920×1080 with alpha if needed and change the jpeg to a png or tif. Jpegs just always look soft until rendered. Even if they say they they don’t need any rendering. The motion file may be your big culprit. What’s in it? How many effects and layers? I know you’re adjusting the jpeg/psd, but when you render it, you’re rendering ALL the layers above and below it. Non destructive compositing. It doesn’t render the motion file then ADD the jpeg to it when you add the jpeg later. It renders the jpeg WITH the motion file all at the same time for the highest quality possible. So if you change the jpeg position, you have to rerender everything. And who knows what’s in that motion file.

[Matt Surf]

The Motion file is off the Flora templates and is the Vine growing across the screen, nothing too drastic I don’t think (but I may be asking too much of the system??) and the photoshop file I converted to a JPEG – so should I reconvert it as you mention?

With regards to sequence settings, in your opinion, am I doing the right thing in making my sequence settings the same as the majority of my footage? I’m using four/five different sources (Canon EOS, Sony Z1, Sony Z5e, Motion) so think it’s impossible to have a uniform sequence that will encompass everything??

PS Consider wrist slapped for over-estimating the power of the Mac :0)

[Scott Sheriff]

You might try using png instead of jpg. FCP and Motion play well with these, you can have alpha channel if you need it, and the quality is better than jpg. If you need to have a large frame size for movement, it is less processor intensive than .tif
Not a magic bullet, but something to try.

Scott Sheriff
Director
SST Digital Media
https://www.sstdigitalmedia.com

[Matt Surf]

Thanks for your help Scott

Matt

[Bret Williams]

I think you’re over estimating the power of 32bit computing. Especially a 32bit program on an imac. I would work in a pro res sequence yes. I’m confused. In one post you said it was xdcam footage, and in the first post you said it was pro res. I’m not familiar with the sony cams, but I’d convert them to pro res like the eos footage if that’s possible on ingest.