Ehm, no, 7% or 26% is not "much greater" than a 1.3 (or even a 4.5) times improvement. Saying something is 1.3 times faster is the same as a 30% (or 33.333333% or something similar) improvement.![]()
Phoronix: Google Releases VP8 Bali Codec SDK
Google has released a new version of its open VP8 codec. This new version, which is codenamed "Bali", there's a focus on improving the performance and video quality...
http://www.phoronix.com/vr.php?view=OTE5MQ
Ehm, no, 7% or 26% is not "much greater" than a 1.3 (or even a 4.5) times improvement. Saying something is 1.3 times faster is the same as a 30% (or 33.333333% or something similar) improvement.![]()
I know that ffmpeg guy have done an other one which was really fast.
So what is the winner between ffvp8 and libvpx?
I'd be pretty cool if VP8 guys would stop caring about crappy metrics like PSNR for video quality.
Looks like he wrote it in haste.. A quick click on the link to the blog shows where he got his figures.
So they're talking about different settings.. Best, good and real-time encoding."Best" mode average encoding speed: On x86 processors, Bali runs 4.5x as fast than our initial release and 1.35x faster than Aylesbury.
"Good" mode average encoding speed: Bali is 2.7x faster than our initial release and 1.4x faster than Aylesbury.
On ARM platforms with Neon extensions, real-time encoding of video telephony content is 7% faster than Aylesbury on single core ARM Cortex A9, 15% on dual-core and 26% on quad core.
On the NVidia Tegra2 platform, real time encoding is 21-36% faster than Aylesbury, depending on encoding parameters.
"Best" mode average quality improved 6.3% over Aylesbury using the PSNR metric.
"Best" mode average quality improved 6.1% over Aylesbury using the SSIM metric.