Intel Atom: NVIDIA ION vs. Radeon HD 4330 Graphics

[FireBurn]

I think the reason the comparison uses the OSS drivers for ATI is that in Ubuntu 10.04 the Catalyst driver is not supported at this time. But adding a plot for Ubuntu 9.10 with Catalyst would have helped.

In my machine (Radeon HD3450), Ubuntu 9.10 + Catalyst gives four times the fps you get with 10.04 and the open source stack (Padman @ 1920x1080). This would place the ATI results slightly higher than the ION overall.

Are there no other distros that could have been used?

[monraaf]

Isn't it already time for a new fglrx beta driver for U 10.4? About 1 year ago it hit U repository as first release for Xserver 1.6, why is it different this year?

http://www.phoronix.com/scan.php?pag...item&px=NzE0Nw

Should be out soon.

[deanjo]

In my machine (Radeon HD3450), Ubuntu 9.10 + Catalyst gives four times the fps you get with 10.04 and the open source stack (Padman @ 1920x1080). This would place the ATI results slightly higher than the ION overall.

Are you using a Atom 330? If not then your "performance comparison" is useless.

[Michael]

This was just intended to be a very quick comparison, take it for what you want. The proprietary Catalyst driver would have been used if a public X Server 1.7 supporting release was available.

[[Knuckles]]

As some people said, this is a useless comparison. I'm all for comparing closed vs open drivers for the same chip, but comparing open vs closed drivers when its pretty much known what the results would be is really pointless.

I was actually interested in seeing the performance difference between the chips. My conclusion after this review: err, none at all...

You might as well just used nouveau and concluded that using open-source ati is faster than using mesa software rendering.

I too am very disappointed with this comparison.

[karl]

Seems I'm the only one that thought this was ok.
Everybody says this was common knowledge. Well not for me.
I mean I knew that foss drivers are not as fast as closed source drivers. But in this case the foss driver had as advantage a better hw. Still this was not enough.
So for me this test was not useless.

On the other hand, I also think that it would be nice to also have nouveau and fglrx in the tests.

[Nille_kungen]

I would have been much more interested to have seen Nvidia blob vs fglrx blob vs Nouveau Gallium vs r600 mesa

I second that.
Nvidia blob against open source ATI, I can't say that it surprises me that Nvidia blob was faster and i knew that before i read this article.
Feels like it's angeled to make ati look bad.
Please, try to be impartial in tests and articles.

[Michael]

I second that.
Nvidia blob against open source ATI, I can't say that it surprises me that Nvidia blob was faster and i knew that before i read this article.
Feels like it's angeled to make ati look bad.
Please, try to be impartial in tests and articles.

Nothing is angeled to make anyone look bad, besides making it difficult to test fglrx with the lack of an X Server 1.7 supported release that should be out by now. This was simply running a few tests with the available hardware at hand before the systems were thrown into the farm for the Ubuntu Tracker.

[deanjo]

Feels like it's angeled to make ati look bad.

Hey the truth hurts sometimes. Consider this a hard dose of reality when it comes to the price one pays for the sake of "free" drivers.

[mirv]

I would to point out that it was made quite clear in the article that it was closed source vs open source, and that things were as expected.
With that said, I don't see why a supported version of Ubuntu couldn't have been used - AMD have a history of not supporting "beta" distros; so for them, xserver 1.7 support isn't a priority until their supported distros have it in stable.