AMD vs NVIDIA drivers, that big difference?

[mirv]

Easy:
nVidia knows how to write decent drivers, while ATI fails miserably at driver implementations. ATI's linux/X drivers are particularly bad, while their Windows drivers are mostly acceptable.

fglrx has been one long trip to bugginess for me with linux, which means this is my first ATI GPU in almost 10y and the experience has been less than pleasing.

I think you'll find both companies to be as bad or decent as each other.

[Panix]

I think you'll find both companies to be as bad or decent as each other.

The guy tested and compared three cards, two Nvidia and one ATI card.

After the glxgears test, how do you figure the differences are negligible?

Both companies may be bad or decent but the tests show a considerable difference due to drivers. In Windows, the HD 4870 would smoke the Nvidia 8600 card. It just shows a lack of attention by ATI to Linux drivers. This is shown again and again but no change. Just a lot of talk that things are being 'worked on.'

[BlackStar]

After the glxgears test, how do you figure the differences are negligible?

The difference between 40000fps and 100000fps is negligible. It's an improvement of 0.000015 seconds per frame or a reduction of 0.1% in frame time. Do you understand now why glxgears is not a benchmark? Those huge differences (60000fps OMGOMG!) translate into absolutely nothing.

Why? Because fps are non-linear (15 vs 30fps is much more significant than 150 vs 300fps). You need to invert them before you can compare them directly.

[pingufunkybeat]

After the glxgears test, how do you figure the differences are negligible?

Because GLXGEARS IS NOT A TEST. It is meaningless.

Repeat this 100000 times before you post again. This is obvious trolling now.

[Panix]

The difference between 40000fps and 100000fps is negligible. It's an improvement of 0.000015 seconds per frame or a reduction of 0.1% in frame time. Do you understand now why glxgears is not a benchmark? Those huge differences (60000fps OMGOMG!) translate into absolutely nothing.

Why? Because fps are non-linear (15 vs 30fps is much more significant than 150 vs 300fps). You need to invert them before you can compare them directly.

What test is useful then?

I just thought it was noteworthy that the result is basically the same as an older inferior Nvidia card.

Also, even if Wine is tailored to Nvidia cards, is it still insignificant that the ATI card performs no better than the 8600GT card?

Maybe the poster should test with Nexuiz and Unigine? Maybe try Doom 3 and use the Test Suite???

[BlackStar]

What test is useful then?

I just thought it was noteworthy that the result is basically the same as an older inferior Nvidia card.

It is not. Glxgears tests a very specific part of the hardware that doesn't reflect actual performance.

Also, even if Wine is tailored to Nvidia cards, is it still insignificant that the ATI card performs no better than the 8600GT card?

Maybe it is, maybe it isn't. If both the Ati card and the 8600GT are getting >100fps then it is insignificant. If they are getting <60fps then it is significant.

Maybe the poster should test with Nexuiz and Unigine? Maybe try Doom 3 and use the Test Suite???

This would be more useful.