Intel Sandy Bridge Linux Graphics? It's A Challenge

[Prescience500]

Sandybridge is out, but no Gallium support or support expected in later drivers...makes me wonder. Does Intel actually plan to some day develop and optimize their open source drivers to the point of making them faster than what Gallium is capable of? What other reason would they have for not wanting to begin transitioning to Gallium?

[Shining Arcanine]

So in the end, we will have to wait for Q3/Q4 of 2011 in order to see a "proper" support of Sandy Bridge packed in our distribution.
Thankfully they have started the Linux support a year ago...

Gentoo Linux users probably do not have to wait that long. The source code is available now, so they just need to switch to instruct the package manager to install it, which might be somewhat involved, but it will become easier as more Gentoo Linux users obtain Sandy Bridge processors and write ebuilds for the newer software, assuming that they have not written them already.

[popper]

I don't know if deanjo does, but I certainly match my TV refresh to the video refresh. 99% of the time this is 24fps and it really does look better.

which is Not surprising given film's are filmed in 24fps.

it's not a problem AFAIK as x264/ffmpeg has the options to re-code to a given frame rate etc , and i assume the SB ASIC x264 code will also expose the right API to use these options to you (if not now then sooner rather than later)

[popper]

Arrgh, so close intel but "no 23.976 fps playback" "The limitation is entirely in hardware".

http://www.anandtech.com/show/4083/t...-2100-tested/7

thats OK, its about time the world just started using the far more sane generic Uk PAL (24) 25,50,100, and sooner than later the 200 FPS/freq's instead of those crazy USA dot whatever frame rates that cause lots of coding/timing problems and wasted time in the digital video age LOL.

No..i think its just a lack of time before they fix that programming error

[popper]

Initial testing shows much better quality than CUDA or Stream based accelerators. At least for Windows (Media Espresso 6). However, since this is a dedicated ASIC, the Linux implementation will have the same quality.

You Got access then , good.

[popper]

Please leave the "obligatory booth babes" crud until the end of all other CES related articles. I'm getting sick and tired of clicking on all these show links only to have them featured over and over and over on tech sites. Hardware first, adolescent eye candy last.

True, Michael would be Far better served if he were to get in contact with Francois (he popped up again in his doom10 thread today then disappeared) and see if he's also going to the show, as He May (or not) be able to sort out access to these Sandy bridge 4 core i7 etc for the phoronix tests.

If You Do get access Michael, please Be sure to get a copy of that ASIC x264 code from Francois and Also run that in your tests along side the normal CPU x264 run's.

in fact pop over to the #x264dev IRC channel and ask them, DS etc about running real informative x264 Encode benchmark tests with the very latest code base etc, most PR web reviews to date do it wrong ( or as Dark_Shikari likes to say/put it "pants on head retarded" ) and do not provide the required data to really know where the real CPU improvements are.


a simple

make checkasm;./checkasm

checkasm --bench

results on each CPU you run tests on would probably be a good start to 2011 benches.

http://akuvian.org/src/x264/freenode-x264dev.log.bz2

[popper]

Touch? Michael. Entirely correct, and a rightful shot at Intel.

Then it's a good sign that Linux is getting more publicity, maybe even resulting in the press kits having built drivers


Re the Sandy encoding support - anyone know if it will merely accelerate x264 (keeping its excellent quality), or will it be a more complete block like the GPU encoders (whose quality sucks, incidentally)?

according to Francois (Piednoel Senior Performance analyst at Intel Corp Santa Clara ... ) the Intel guy writing the initial Sandy bridge ASIC x264 code.

he said that it has pretty much all of x264's equivalent routines inside the HW ,so it's just a matter of what the API he wanted help defining gives you access to, id imagine most Video routines as found in FFmpeg can be made to use it in some form given that x264 dev's preferred Decoder etc.

but that's to be determined right now, and OC reliant on Francois if you believe his work behind the scenes at Intel to open up this ASIC for Linux and perhaps other OSS OS code access too (AROS etc) etc.

[deanjo]

General graphics performance: http://techpowerup.com/reviews/Intel...00K_GPU/8.html

OK, just wondering what you were referring to.

[efikkan]

2.5 times better in what?

General graphics performance: http://techpowerup.com/reviews/Intel...00K_GPU/8.html

[deanjo]

I don't know if deanjo does, but I certainly match my TV refresh to the video refresh. 99% of the time this is 24fps and it really does look better.

You bet I do.