What is the best CODECS compressor for apple quicktime. animation is too big an jittery and cinpack sucks (too lossy) so what do you guys think is the bewst compressor.
render to something
then convert to a codec you like
divx, xvid, quicktime
…
whatever, you get the best encoding if you do it more than one pass at a time, which essentially rules out entirlely trying to compress any animation created by blender by simply choosing a different codec.
It’s not just the codec that affects your quality. Your bitrate and other settings will affect the image and sound. Generally speaking, the smaller you make the file the lower the quality will be.
Each type of video may need a different type of codec. The animation codec is pretty old, based on the days when most animation was still flat colors with sharp edges. 3D rendered work is more like video, and you may find the video codec works better than the animation codec.
MPEG-4 works well for web situations assuming you set things correctly, but because it is highly compressed you will see artifacts. That’s just what happens when you compress things. The more compressed it gets the more artifacts appear. This applies to still and audio compression as well.
Quicktime’s Pixlet codec was developed by Pixar for transferring HD quality clips digitally. Because of that it is generally good quality, but it doesn’t create files as small as MPEG-4. If you want lots of people to view your work, choosing a proprietary codec like Pixlet will limit the number of people that can view your files.
Divx is not directly supported by Apple and Quicktime, but there are Divx Quicktime codecs you can find. Divx uses a lot of the same technology as MPEG-4 but is not as standards compliant, meaning some people will be able to use it fine but others won’t.
It all depends on what you want to do with the video…
Yeah, and the best thing to do is NOT compress as you render. Save in a non-lossy format with the quality setting on 100%. Then you compress later and use the codec and technology that is best for the media you are targeting.
As an example:
DVDs = MPEG-2
VCDs = MPEG-1
You don’t get a choice with those. If you save your renders in a compressed format, you have to re-compress them into another format for DVD and VCD. The results get worse every time you do it. Make sure your original is non-compressed.
Sorenson 3 is the codec Apple distributes movie trailers in.
Look at extremetech’s review of video codecs. Thay have comparisons of quality and size and encoding times at http://www.extremetech.com/article2/0,1558,1544898,00.asp
That is an interesting link. None of their conclusions match anything I have in several books, three years of professional video magazines, nine years of professional video and web work, or any other source I’ve seen about the subject.
WMV9 gets a 9/10? That is completely contrary to every source - except, or course, Microsoft. If WMV is good, why doesn’t anyone in the pro world use it? Why doesn’t Hollywood or even the local TV station?
Ahh… it’s from Ziff Davis. That explains it. Follow the money trail.
I would seriously suggest visiting sites like http://www.dv.com/ (DV magazine) or http://www.videomaker.com/ (Videomaker Magazine) or other pro video publications for more unbiased information. Because these are not directly related to computer technology but to how computer technology is applied to AV, they are more likely to call the shots where they lie. The AV industry is tied into sponsors, but their sponsors are more oriented to the cameras and other AV hardware than they are to computers and platforms. The magazines I’ve listed criticize Apple, Adobe, Avid, Microsoft, and others more equally than computer or “tech” sources. Their comparisons are also based on tangible, standards compliant methods and they often provide the actual video and results from using the codecs. Reviews and articles from AV sources will also explain what codec and software is appropriate for the method of delivery, since none of the existing software is the best solution for all delivery methods - what works best on the web really stinks when making a DVD and so on.
Looking for unbiased information from any computer or general “tech” publication has become impossible. You wouldn’t read PC Magazine to find info about Macs and you wouldn’t read MacWorld to learn more about PCs. While sources like CNet and Ziff Davis proclaim they are unbiased and independent, look carefully at the relationship between the products they actually review and rate well and the larger advertisers in their magazines and websites. Their is a noticable correlation between who advertises with the publication and what products are reviewed and how they are rated.
Don’t go to a Ford dealer and ask if you should buy a Ford or a Honda.
And also, make sure that you do it different to most other Blender users: do NOT encode to AVI Raw. PLEASE encode to SEPERATE IMAGES. You will love that you have done so in various situations.
… and for a reason:
avi files (well, files on most 32 bit sytems) max out at 2GB
for avi raw, that is reached pretty quick
… and for a reason:
avi files (well, files on most 32 bit sytems) max out at 2GB
for avi raw, that is reached pretty quick[/quote]
Also, AVI Raw are totally uncompressed, so you’ll get a lot more image date per HD space if you use a lossless format like Targa or PNG, both of which also supports an alpha channel.
Martin
And for other reasons:
- You cannot edit individual frames easily
- If blender crashes after 1 week, you loose one week
- You cannot preview individual frames while rendering
- You cannot process an AVI with ImageMagick
etc. 
That’s not true. You could just as easily restart rendering where you left and stick the two AVI files together afterward.
Martin
theeth: Many filesystems (most notably win32 ones) will trash files if the process writing them crashes.
Yah, with Win95 and Win98…
Don’t get me wrong, I’m still saying rendering to individual image is better, I’m just saying that the “loose your work when crash” argument is BS.
Martin