Follow Slashdot blog updates by subscribing to our blog RSS feed

 



Forgot your password?
typodupeerror

Slashdot videos: Now with more Slashdot!

  • View

  • Discuss

  • Share

We've improved Slashdot's video section; now you can view our video interviews, product close-ups and site visits with all the usual Slashdot options to comment, share, etc. No more walled garden! It's a work in progress -- we hope you'll check it out (Learn more about the recent updates).

×
This discussion was created for logged-in users only, but now has been archived. No new comments can be posted.

First Look: H.264 and VP8 Compared - StreamingMedi

Comments Filter:
  • He uses jpeg files to show the end results! Also, read http://x264dev.multimedia.cx/?p=377 [multimedia.cx]
    • It's fine to use JPEG for video codec comparisons as long as the noise introduced by JPEG coding is far lower than the noise introduced by the video codecs. In some cases, the quantizer setting that GIMP calls "95" is enough.
    • The H.264 video in this demonstration was encoded using Baseline Profile, which is designed for low-power mobile appliances, so this test is terrible at doing a proper comparison.

      Using Main or High Profiles, they would be able to gain an average of 10-20% improvement using the CABAC coder, 20-30% improvement using B-frames, and 5-6% using adaptive 8x8 DCT. So, really, this is the worst case H.264 performance versus the best VP8 can provide—and H.264 still manages to look better most of the time.

  • Anyone else notice that different frames were used for to compare? Sure, it may have been in the same or next second. But, that can dramatically change the blur in the frame. Quite frankly, given the apples to oranges comparison, I'd say that the only conclusion that can be made is that they produce similar quality output, but it is still unclear which is superior. Especially, when frame choice can drastically change the outcome.

Computers can figure out all kinds of problems, except the things in the world that just don't add up.

Working...