Cory Caplan
Forum Replies Created
-
Yes, the results are the same on my Aja LHE. As a matter of fact, it plays perfectly fine on a monitor unrendered, and then, once rendered looks on the monitor like it does on the screen.
Now I’m getting nervous…
-
It’s arial narrow, but that’s not so much a concern, it’s a rasterized “graphic” …does the font matter so much? I’m having a problem with a layer above it too, an animated qt.
I just don’t understand how it’s processing the alpha differently in RT vs render SO DRASTICALLY. Might be a new QT bug or something, don’t know if anybody else is experiencing something similar.
https://www.spacecadet.com/video/gkphr/renderproblem.jpg
Tracks are like this:
>V4 – Disclaimer PNG
>V3 – QT Animation Codec w/ Alpha
>V2 – Transparent PNG blue letterbox
>V1 – Video/Solid PNGTime to panic.
-
Lisa, if you do not wish to re-rip, then you can run them through compressor using ProRES 422 as the output format, and you’ll be able to use those clips naitively in FCP. Although, as pointed out earlier, this is not the “best way” to do it.
In the past, I got okay results doing exactly that, ripping with handbrake (h.264 100% quality > compressor > FCP) but Streamclip will do it in one pass.. I’m not sure if it will work with copy-protected material, as I’ve never tried…
-
Paul, you may want to read Chris Poisonn’s informative post, so you might learn how not to answer a question like a total Dickin.
-
Exporting a quicktime reference is always the best practice, as mentioned, it doesn’t tie up FCP, but also, you can use Compressor clusters– which is way, way, way faster on the encode, it does it in the background of whatever apps you’re using etc… The only bummer is when you have either a very quick encode (it would be nice to just send to compressor) or a large one that’s unrendered, as you must first render it, and then send to compressor– but you will more than make up the time in the speed of the encode, plus you’ll have a QT file to make encodes later, should you need it.
-
You just about nailed my list, just add “Fix motion rendering speed, text tool, and especially templates.”
To me, I have no problem using encore for Blu ray should I need it– I can buy another tool for that– I can’t buy something to help me with the other core issues mentioned above.
-
Have you played around with color profiles? Calibrated it? used the right one for the monitor?
FCP automatically adjusts the gamma realtime, assuming you’re in 1.8, so if you’re in 2.2, it will “double crush”
Obviously, you’re seeing major hue issues, and that sounds like a wrong display profile.. Maybe there’s an updated one? Or you could do a manual calibration w/in OS x…
(I’m on a pair of benq 241ws, and I think they do a really good job of color reproduction when calibrated…)
Cory
-
Not too difficult. The AE guy gets it, assuming he is on CS3 or has nucleo on a previous version. Use qmaster to make it a rendering node for compressor as well, and you’ll be making it 100% efficient. Don’t forget to turn on distributed processing on AE CS3.
https://barefeats.com/imacal4.html
The quad might actually be faster in FCP depending on what speed octo it is. Quad 3 @ 3 ghz beats octo 2.8 in a surprising number of benchmarks. (non cpu-intensive) Bus speed is king for most video operations today. You will get a boost from running system level I/O on 8 cores, and that usually makes up the gain, but there’s no significant benefit to FCP on an octo today vs. a 4 except for compressor.
FCP is NOT multithreaded well. Renders of all sorts use half of my Quad’s CPU or less, quite often.
https://barefeats.com/octopro5.html
As you can see from the above, during render FCP is using 125% (1.25 processors of 8) and motion is using 57% — less than 1.) I quite often render FCP in the background and use photoshop and work (but not render) in AE in the foreground without noticing any performance hit.
As appreciative as I am of the benchmarks at barefeats, I can’t fathom a workflow that would allow you to render in 4 different apps at once very often.. Talk about an organizational genius.. 🙂 Unfortunately, he didn’t do FCP render vs FCP render, which would have been helpful. (Maybe it’s there and I didn’t find it…)
The caveat, of course, is workflow. If you spend a lot of your day rendering AND doing other processor intensive stuff (AE, etc.) WHILE you render, then certainly, the 8 core might be an advantage, but based on the information given, without knowing HOW MUCH AE your AE guy does, it certainly would be “MOST EFFICIENT” to the AE guy, and as a render node for compressor.
Cory
-
[Cliff Vaughn] “o y’all think it’s normal for a SD Hollywood studio DVD to look a lot better (though by no means perfect) on a HD display than what I might author using DVDSP?”
What’s your source? Mpeg-2 compression is very much optimized for exactly that– Hollywood film, meaning shallow depth of field shot to grainy celluloid. It hides the majority of its compression in the “soft focus” areas and grain. Whereas most video that you will shoot, even HD has a very deep DoF, meaning lots of image detail that really exceeds the amount of data available– so lots more noise.
One thing I’ve notices is that LCDs show compression artifacts in much higher contrast than CRTs, probably to do with the enhanced contrast ratios (they boost these for “sharpness”)
Keep in mind that LCDs have a native display resolution, and unless you have a 720×480 LCD (you don’t) You’re scaling the video– and this is going to make things look not as good as they do at native resolution (or native upscaled resolution)
CRTs show things in “native resolution” of course.. 🙂Furthermore, as others have said, make sure to go through all the settings on compressor. Even on “best” there is still one or two settings that can be cranked higher– the bitrate, resize filter, and rate conversion (if applicable) can all be set “higher” than the default DVD “best quality”
Finally, you might even try different encoders. Encore’s encoder is really nice for some stuff. I’ve never tried to quantify exactly *what* type of source does better, but I have noticed occasional better results.
I still use procoder on occasion (especially from my Avid system) Slow on best settings, but superb compression result, with tons of tweakable options.
-
Yeah, I remember that actually. Many modern videogames have black level calibration, too.. (Zelda on the wii pops to mind) However, even if you put it on the disc, my feeling is that those who would take the time to do it have already run an Avia (or whatever) calibration disc through the system.. (Or the standard THX calibration)
Dealing with “investors” and “producers” is tricky– I don’t know how well they’d take to a “ya gotta calibrate your TV to watch your movie” thing– even though WE know you really should, anyway….