I've seen vade's useful patches for optimising playback of qt movies (interesting that a lot of the posts are 3 yrs old and there's been hardly anything google chucks out more recently).
However, before that, and before I bought Max/Jitter, through sheer contrived I've-just-bought-a-new-2010-MBP naivety, I've ripped some of my source material using Handbrake (h264, yuv420p, 1920x1080, 25 tbr, 25 tbn, 50 tbc), and *then* started working through the Max tutorials. Naturally, my brand new MBP plays anything I can throw at it when using MPEG Stream Clip (which I used to prepare sections for VJ practice). But as expected I would like optimal framerate through jitter when scratching/overlaying multiple effects/mixtures of high speed frame cuts and multiple visible sources simultaneously - a different kettle of fish.
I'm aiming at multiprojector live playback so I thought irrespective, for efficiency I'd try and optimise some test clips from the above res to others (notably Motion JPEG A) but I am not seeing much reduction in file size. I'm presuming this is due to the efficiency of H.264, or am I wrong and there is something I can do to reduce the file size and throw out mucho bits which the eye won't miss?
I guess thus the questions are:
1) Would I be better off reripping the src material to motion jpg (conversion from h264 isnt doing it) - given reducing the size from 1920 down to 960 in conversion only shaves 10% and other setting's I've tried aren't bettering this much
2) Given vades post was 3 years ago, I tested his multichannel GPU UYVY patch with a 15mb clip and a 78mb clip (MBP 2.8 dual core i7/8gb RAM) xfade together and got ~14-20fps - not good enough given what else I'm going to do. So I'm suspecting that the settings of the videos i'm using are not optimised.
3) What are the minimum res settings people use to maintain quality when projecting on 6 foot high screens? I've only seen vades so far and I'm interested in other people's successes to date.