Nathan Veer
Forum Replies Created
-
It definitely depends on your market. I’m dealing with a daily 28:30min show. While I would hate to think of YouTube being any kinds of standard for online video quality, again, most encoders will get by if you keep your bandwidths that high. Then I have to ask myself, what is compression? It’s the art of maximizing quality at the smallest acceptable file size.
back to the original discussion, multpass makes all the difference in encoding by the very nature of what it is. Checking the video and it’s content before it actually encodes. How can a computer know what data to throw out unless it has seen the footage at least once. Matrox doesn’t do this neither does Elgato and that is why you wont see it used in serious web encoding applications.
-
As you said every source video is different and it depends what kind of bandwidth you’re talking about. I’m talking about encoding for Web. Anything looks decent at high bandwidth. Once you get down to 700 or 500 that is where you can tell the good compressors from the great ones. You can take it up with Matrox them selves. Ask for Jean Pierre.
-
Not even close. I actually did my testing with the Matrox Engineers in Canada. In the end I sent them My custom settings and some uncompressed footage and asked them to tear it to pieces and show me what I was doing wrong. After several weeks of trying to match COmpressor’s settings the engineers themselves responded by saying they can’t come close and that they are working on including multipass encoding. They were great to work with and very helpful but facts are facts. To obtain good quality while achieving small file sizes you have to have multi pass.
Single pass encoding is like asking your encoder to fly blind. It has no idea what’s coming up. With multipass it at least knows where the scene changes are and fast motion so it can allocate more data to those areas.
-
I have yet to find an affordable hardware encoder that does multipass encoding. Both Elgato and Matrox’s solutions only do single pass. Form my testing these devices the multipass option makes all the difference in quality. Compressor is slow, especially on a G5 (even if it’s dual) but it looks great. However, when I’m in a pinch and I don’t have the time I can achieve 90% of the quality with mpegstreamclip. Search it on google. It’s a wonderful, free encoder. I love it.1
-
Elias: The folder on XSANSATA/DVDAuthoring/1nd844-test
My findings are exactly the same. I get this error every time I chain a job and submit it to a cluster. Virtual cluster and real, multi computer cluster. Yes I could turn qmaster off etc etc…but that defeats the point. Shouldn’t I be able to chain these jobs and take advantage of the speed boost I get when using a cluster?Has anyone tried this in the new FCS?