NVIDIA 180.51 Display Driver Released

[jonnycat26]

Ignoring the fact that AMD pays people to work on the open source drivers.. look here, and here. That's over a dozen commits in the last 24 hours. I think that counts as stepping up.

Yet my brand spanking new ATI card with either the ATI drivers or the FOSS drivers doesn't perform as well (in Linux) as my older tech NVidia card.

So, seriously, I hope NVidia keeps doing what they're doing. Maybe one day ATI will catch up, but for now, it doesn't seem likely.

[krazy]

Just look at how many non-paid contributors are working on the radeon or intel drivers. There are lots of problems with these drivers, the manpower to fix these simply isn't there, and almost nobody is stepping up to do it.

Ignoring the fact that AMD pays people to work on the open source drivers.. look here, and here. That's over a dozen commits in the last 24 hours. I think that counts as stepping up.

My point is that NVidia isn't "evil" like many make them out to be.

There's quite some evidence that says otherwise.

[greg]

Since when has familiarity with codebases stopped anyone in the community from "fixing" anything?

Just look at how many non-paid contributors are working on the radeon or intel drivers. There are lots of problems with these drivers, the manpower to fix these simply isn't there, and almost nobody is stepping up to do it.

What does that have to do with the price of tea in Russia? If you'd bothered responding to what was said and not what you think was said you'd realize that I was not talking about nvidia's drivers or any problems with them, and you'd look less pedantic.

My point is that NVidia isn't "evil" like many make them out to be.

[yogi_berra]

That's not true -- especially for non-trivial stuff like GPU programming. There are very few developers intimate with that. Of course, you always have the possibility that someone will pick it up, but often enough that doesn't mean anything.

Since when has familiarity with codebases stopped anyone in the community from "fixing" anything?

And while NVidia's drivers are closed-source, they have a very open development model and communication and feedback with/from the developers are great. Some parts of the driver, like nvidia-settings and VDPAU are even open source.

What does that have to do with the price of tea in Russia? If you'd bothered responding to what was said and not what you think was said you'd realize that I was not talking about nvidia's drivers or any problems with them, and you'd look less pedantic.

[greg]

That is one way to spin it, another way is that it just means that anyone can fix the problems when developers screw up.

That's not true -- especially for non-trivial stuff like GPU programming. There are very few developers intimate with that. Of course, you always have the possibility that someone will pick it up, but often enough that doesn't mean anything.

And while NVidia's drivers are closed-source, they have a very open development model and communication and feedback with/from the developers are great. Some parts of the driver, like nvidia-settings and VDPAU are even open source.

[yogi_berra]

The NVIDIA 185.19 Beta is still the latest in the 185.xx series, but NVIDIA has provided a pre-release of the 180.51 driver.

Thank you for taking the time to make the Beta/Pre-Release distinction.

Open sourcing doesn't guarantee manpower. It presents opportunities for additional manpower but it doesn't mean it will happen.

That is one way to spin it, another way is that it just means that anyone can fix the problems when developers screw up.

[Jimmy]

Frankly, it doesn't matter how often nvidia releases a new driver -- their stuff is closed, and thus inherently limited by the amount of manpower they can personally throw at it, and their marketing department, and ...

Open sourcing doesn't guarantee manpower. It presents opportunities for additional manpower but it doesn't mean it will happen. Similarly a just because a project is closed doesn't automatically mean that there isn't enough manpower to produce a quality product.

Even if Nvidia is suffering from manpower issues as you would have the world believe, they still are releasing a higher quality product.

And if ATI's closed drivers are so manpower starved (because they're closed) and all the real progress is by the open community then there's no point in supporting their market share since it won't impact the quality of drivers available.

[psycho_driver]

Hmm the 9600 gso 512mb support must be in name only. I've had one in my secondary desktop for about a month now and it's been working fine with an older driver.

[deanjo]

Yeah, so will development unhindered by testing procedures. The 180.50 release somehow didn't work on 7xxx cards any more, and nobody noticed before release

as a nvidia user I always dread upgrading the nvidia-drivers. Regressions are way too common

So, you mean they released a non final driver for review and use, catch a bug because of it, and then 2 days later release another to fix the bug? THOSE BASTARDS!!!!

Ooops, wait a minute, isn't that the whole purpose of releasing a pre-release?

[greg]

I don't really see how they're pushing it. At least they're not doing it with any real code work. va-api seems to be pretty much dead (no commits for a year now to libva) and XvBA remains nowhere to be seen.

Besides, have you ever looked at va-api? It's much more complicated to use than VDPAU without offering any real benefits, and it's horribly documented, which means not at all.