Intel Linux Driver Now Nearly As Fast As Windows OpenGL Driver

An anonymous reader writes "Intel's open-source Linux graphics driver is now running neck-and-neck with the Windows 8.1 driver for OpenGL performance between the competing platforms when using the latest drivers for each platform. The NVIDIA driver has long been able to run at similar speeds between Windows and Linux given the common code-base, but the Intel Linux driver is completely separate from their Windows driver due to being open-source and complying with the Linux DRM and Mesa infrastructure. The Intel Linux driver is still trailing the Windows OpenGL driver in supporting OpenGL4."

Comments

woo

[nomadic]

This should convince anyone that open source linux software can compete with windows, given 22 years.

Re:woo

[epyT-R]

This has little to do with the architecture and mostly to do with vendor support. This has always been a problem for non-windows OSes. Even apple's opengl isn't exactly the best in terms of performance. Linux easily outperforms it when using nvidia's driver.

Re: woo

[tysonedwards]

Microsoft blew its right foot off with Windows 8.
They went to the doctor to get it reattached with Windows 8.1 only to wake up to find out that a second left foot was attached in place.

Re: woo

[TheGratefulNet]

maybe windows 8.11, 'windows for podgroups'. not sure if that's an official project name or not...

Re: woo

[mewsenews]

How does microsoft do so much user testing and have no idea these products are going to be colossal flops?

I can't imagine the reasoning that went on behind the push for Windows 8:

"Let's unify our mobile and desktop interfaces, because we have a stranglehold on the desktop, people will gravitate towards our mobile offerings"

The public responds "we hate this" and they choose to do it anyway? Don't they do focus groups? Didn't they anticipate that people are disgusted by a touch interface on their keyboard+mouse system?

I'm fascinated and horrified but I'm also pleased because I am not fond of Microsoft, but what the hell do they think they are doing?

Intel

[Frosty Piss]

Someone smells a game plaform....

Re:Intel

[beelsebob]

Bear in mind that today's 5000 series actually does compete pretty bloody well with nVidia and AMD. It's near impossible to get a faster GPU in a thin and light laptop. The GeForce 740m is the same speed as it, the 750m is getting into power brackets that can't be put in a thin and light, and is only about 10-20% faster than the 5200 pro.

For me, Intel is doing a pretty impressive job of catching up. We've gone from intel being no where in terms of GPU performance to being able to equal the best nVidia and AMD can do at least in the power constrained market.

Re:Intel

[mrbluze]

We're also hitting diminishing returns with game graphics. It used to be generational differences between games was huge but these days can you really tell the different between this years shooter and last years shooter?

Barely.

You mean they might actually start to give a toss about playability?

Re:Intel

[Anonymous Coward]

Punctuation is what triggers the bug!!!

Re:Intel

[Bengie]

Geometry calculations scale with the cube of the number of polygons. Non-ray-tracing engine's days are numbers and ray-tracing is soon(tm). Ray tracing is O(1 K), where K is large, current engines are O(n^3 K) where K is small. N is becoming an issue.

Thank you Gabe Newell

[Dega704]

Hence why even non-gamers were so excited about Valve's gambit. Even with the few games released so far, it has brought tons of much-needed development effort to the areas GNU/Linux was lacking in. Imagine how things will be if SteamOS & Co. succeed and it becomes a major gaming platform. Free software purist or not, everyone is going to benefit.

Re:Linux DRM

[Guy Harris]

This is the problem with using DRM and other 3-letter acronyms in the article body; they become quite ambiguous.

Yup. Direct Rendering Manager, not Digital Rights Management.

(Having worked on Server Message Block protocol implementations, seeing "SMB" stand for "Small and Medium Businesses" gives my brain heartburn. :-))

NOT neck and neck

[Sits]

The headline is bad/misleading - many of those benchmarks are showing a disparity of more than 10% between the drivers. Using the numbers from the Phoronix article, Linux results are the highest number from any Linux driver (there are many cases where the most recent driver was not the best) to try and prove headline:

Linux = [35.88, 140.90, 43.37, 23.5, 32.23, 19.17, 25.17, 16.68, 99.24, 63.94, 46.80, 29.46]
Windows = [41.47, 162.88, 36.57, 27.0, 31.46, 19.37, 24.47, 16.85, 104.04, 65.15, 55.05, 36.63]
for i in range(len(Linux)):
  diff = abs(round((1 - Linux[i]/Windows[i])*100, 1))
  "Windows win by %d.1%%" % (diff) if Linux[i] < Windows[i] else "Linux . win by %d.1%%" % (diff)

'Windows win by 13.1%'
'Windows win by 13.1%'
'Linux . win by 18.1%'
'Windows win by 13.1%'
'Linux . win by 2.1%'
'Windows win by 1.1%'
'Linux . win by 2.1%'
'Windows win by 1.1%'
'Windows win by 4.1%'
'Windows win by 1.1%'
'Windows win by 15.1%'
'Windows win by 19.1%'

So out of 12 results, 5 showed a 10%+ difference between Linux and Windows Intel drivers in favour of Windows and 1 showed a 10%+ difference in favour of Linux. The conclusion that the drivers are neck and neck does not follow from the premise for around 40% of the results and that's when being unfairly generous to Linux!