To test AVC: Looking for raw video sample...

[Bluedan]

...or better, one compressed with a lossless codec as Huffyuv is for exemple.

My aim is to compare picture quality of MPEG-4 Part10 aka AVC/h.264 with MPEG-2, not in terms of transcoding like DVD contents to AVC but in regard to the fact that it will be one of the video codec standards used in HD-DVD.

Frankly, I'm an XVid disciple and my interest in AVC started when I saw Doom9's codec shootout with impressing results concerning Nero Digital/AVC.

For my purpose I need an uncompressed or losslessy compressed video sample which does not derive from previously compressed material. The resolution should at least be 720 x 540 px.
I was already searching this forum and via mighty google.
The only clip I found was unsuitable because it was a take of someone clicking through a programmes menue structure!

The scene I need should be "movielike": people and movement.
It doesn't have to be too long, I'll be be glad to get a grip on a sample anyway.

So, does anyone here have material or know of a link???

thx in advance

P.S: I sure may be able to post pictures if somebody is interested.

[thegeby]

Some of these (http://www.avsforum.com/avs-vb/showthread.php?s=&threadid=439927) are probably raw.

[Bluedan]

First look: a lot of links to check...
thank you

[babayaga]

Originally posted by Bluedan
For my purpose I need an uncompressed or losslessy compressed video sample which does not derive from previously compressed material. The resolution should at least be 720 x 540 px.
I was already searching this forum and via mighty google.
The only clip I found was unsuitable because it was a take of someone clicking through a programmes menue structure!

The scene I need should be "movielike": people and movement.
It doesn't have to be too long, I'll be be glad to get a grip on a sample anyway.

You will find some standard raw sequences in ftp://ftp.crc.ca/crc/vqeg.
Those are some of the sequences that are used to qualify encoders (for instance football or mobile). Despite their "academic" aspect, they are very good for evaluation purposes.

Be carefull when evaluating, most (if not all) of the full D1 sequences are interlaced so you need an encoder with interlaced support. And since they are raw 4.2.2 sequences, you have to convert them to 4.2.0 to feed your test encoders (if they don't support natively 4.2.2).