Intel Wants YOUR Linux Questions, Feedback

[allquixotic]

Firefox is huge memory hog. That doesn't mean anyone is going stop using it. This thread is about Intel anyways.


You will always be at somebody's mercy. Live it. Accept it.



Being tied to a specific version of the kernel is only a disadvantage if the newer kernel is exceedingly better than the previous one. If you can't prove it there's an advantage, with real numbers..
How is that any less short-sighted? Are you just looking at version numbers? What about the articles on this website?

I'm looking at things like new hardware support; support for new compression algorithms in cleancache; bugfixes for my USB Audio that probably won't get backported to older kernels.

The mainline kernel is where it's at; things are moving at an incredible pace. 2011 was a bit of a bum year in that not a whole lot of monumental changes happened, but there were still some that were worthwhile enough that I was really motivated to upgrade to each new kernel release.

Only someone who is either extremely lucky or has very old hardware would not want to upgrade to new kernels (very old hardware would make it less beneficial to upgrade because most new kernel pushes add or improve support for newer hardware, or add features that are only applicable to newer hardware). Or just someone who isn't discerning enough to know better.

[Ansla]

@ansla

is it really? Plase excuse my indolence

PS. maybe then it is argument for Intel moving toward Gallium then?

Intel already has a fully working solution for vaapi, so moving to Galimu3D would not benefit them, what would really benefit Intel is that it would be possible to switch between on-board Intel graphics and discreet AMD/Nvidia graphics without restating the X server when all the free graphics will use the xorg state tracker. I don't mean that it would work right away, but at least it would theoretically allow it, the way fglrx switches now between on-board AMD and discreet AMD since both cards are handled by the same X driver. The same way the xorg state tracker would be able to switch between a Gallium Intel driver and radeon/nouveau without the X server even knowing this.

[jltyper]

Only someone who is either extremely lucky or has very old hardware would not want to upgrade to new kernels
Or just someone who isn't discerning enough to know better.

Extremely lucky? Loosing video performance and proper sleep/suspend for USB audio support isn't what I would call lucky.
Only someone with old hardware? You might as well say, anyone who doesn't own a new computer. Stop there.

I've seen the changelogs at kernel newbies, I've seen the articles on this website. Only "someone who isn't discerning enough to know better", upgrades just because of a version number. That's just blind faith.

What is exactly is this huge advantage? Some incomplete filesystem? Broken ACPI? Lower FPS average? Is that why people should upgrade? Even Linus himself, has doubts about the newer kernels. Notice how everything is on a support/fix/break cycle?

I have a theory and feel free to prove me wrong. Desktop users, don't need any kernel newer than 2.6.33.
Give me something practical. Really practical. Ext4 scaling to 48 cpus is nice, for example. Personally I'd rather have better/faster graphics. But USB audio support? pffft. That's weak. Weaker than weak. Whatever.

[123_qwe]

Hello all.
Is the libva encode implementation for Sandy (Ivy) Bridge using Quick Sync? Can we expect the same level of performance seen on Windows?

[przemoli]

Yes both AMD and Nvidia support odler hardware BUT it's due to shared code between Linux and Windows! They do not support linux clients, they support Windows clients (and those should be more by 2 orders of magnitude!). And it mean that their drivers are CLOSED SOURCE.....

I have question, too!

Why there is asimetrical support for DX and OGL in Win? DX11 should mean in theory that OGL 4 is possible O_o.
Does it mean troubles for Linux drivers (that you will not reach DX feature parity in form of OGL 4?).

[RussianNeuroMancer]

Yes both AMD and Nvidia support odler hardware BUT it's due to shared code between Linux and Windows! They do not support linux clients, they support Windows clients (and those should be more by 2 orders of magnitude!). And it mean that their drivers are CLOSED SOURCE.....

R300g share code with... what?!

[AlbertP]

There is one important difference in support for old cards between AMD and nVidia: AMD does not update the old drivers (pre R600) for newer OS'es like Windows 7 or modern Linux distros. nVidia releases those updates for many legacy cards.

[RussianNeuroMancer]

AMD does not update the old drivers (pre R600) for newer OS'es like Windows 7 or modern Linux distros.

Because they already working on R300g.

[AlbertP]

No. AMD has just dropped all support for those chips, on both Windows and Linux. This is not related to r300g - you don't even have r300g on Windows. The r300g is maintained by the Mesa people, not by the Catalyst crew.

[RussianNeuroMancer]

The r300g is maintained by the Mesa people

Only be the Mesa people, no one from AMD? AFAIK you are wrong here.