hardware accelerated x264 almost here

[deadrats]

it looks like the guys behind x264 have managed to accelerate portions of the popular, open source h264 encoder using Open CL:

http://techreport.com/articles.x/23324

What happened to those OpenCL-accelerated video encoders we were all daydreaming about? Well, they're still in the works, believe it or not. The folks behind the popular x264 software encoder have been quietly plugging away at an OpenCL-accelerated version of their lookahead pipeline. Lookahead only accounts for 10-25% of the total encoding time, according to x264 lead developer Jason Garrett-Glaser, but the process allows for nearly unlimited parallelism and is relatively easy to implement in OpenCL. Re-writing all of the x264 encoder in OpenCL, by contrast, would be "very hard." Garrett-Glaser says the accelerated lookahead can increase performance by up to 40% on AMD's new Trinity APUs and by a factor of two on the latest Radeon graphics cards.

it's a start...

edit: i ran across this article while googling for more info:

http://www.anandtech.com/show/5835/testing-opencl-accelerated-handbrakex264-with-amds-trinity-apu

i can't wait to test it out with my quadro 600

second edit: i found a test build:

http://astrataro.wordpress.com/2012/05/17/x264-test-build-with-opencl-lookahead/

[aedipuss]

did you bother to read the techreport article? they all suck and always have.

[epsilonalpha]

did you bother to read the techreport article? they all suck and always have.

And did you? Distinguish between hardware accelerated h.264 encoder with hardware accelerated x264 please.

[deadrats]

did you bother to read the techreport article? they all suck and always have.

did you bother to read the quote i posted from that article, this is a hardware accelerated version of x264 that we're talking about.

[aedipuss]

who gives a rat's ass about "sandy bridge" or amd apu encoding, where's the nvidia gpu encoding.

[deadrats]

who gives a rat's ass about "sandy bridge" or amd apu encoding, where's the nvidia gpu encoding.

question, why are you hell bent on being a complete dick? being half a dick isn't enough for you?

if you look on the 5th page of the tech report article you will see benchmark results for a GT 640 as well as screenshots to compare quality.

[jmac698]

This is big news, because normal hardware accelerated encoders suck, but they are fast.
http://compression.ru/video/codec_comparison/h264_2012/

ps the appendix for the hw encoders tests are coming out soon

[aedipuss]

who gives a rat's ass about "sandy bridge" or amd apu encoding, where's the nvidia gpu encoding.

question, why are you hell bent on being a complete dick? being half a dick isn't enough for you?

if you look on the 5th page of the tech report article you will see benchmark results for a GT 640 as well as screenshots to compare quality.

it's my job. quality still sucks even for the nvidia gpu encoding.

[mh2360]

It's generally accepted that "regular" GPU accelerated encoding sucks. At present, you can't accelerate the whole encoding process without impacting quality. I think what the x264 team are trying to do here is only accelerate the portions of the encoding process that would have little or no impact on quality.

Even a 10-20% increase in speed over the original x264 would be better than nothing.

[RabidDog]

Actually they are re-coding the parts that allow for nearly unlimited parallelism. EG the easier parrts to recode for OPENCL. I presume they will still expect the same quality. Its not hardwardware bad software good, it just happens that, so far, hardware encoding has been a poorly written, one off the wrist, piece of Programming's.
Good software's been a long time coming. I still rememeber when an early experimental badaboom destroyed my mx440 video card. it went BadaBoom
Maybe in the professional domain there is better software ?

[mh2360]

Good software's been a long time coming. I still rememeber when an early experimental badaboom destroyed my mx440 video card. it went BadaBoom

That's a shame, was it an antique?

Maybe in the professional domain there is better software ?

That's not been my experience so far. "hardware bad software good" pretty much sums up the state of affairs at the moment if you exclude the high-end professional stuff.

[jagabo]

Even a 10-20% increase in speed over the original x264 would be better than nothing.

Probably more like 5 to 10 percent (40 percent of 10 to 25 percent). Either way, nothing to get excited about.