NVIDIA 396.54 Linux Driver Offers Big Performance Boost For Frequent Gamers

[sleeksorrow]

Can you please not promote Nvidia like that? "Offers big performance boost"? One would think that Nvidia developers managed to make a big breakthrough and increased performance. Did that happen? NO

What happened instead, was that they fixed a regression. But since most people read only the heading, impressions are different.

I really don't know why one of those expressions should be wrong. Yes, it was a regression and they fixed it. And fixing the regression caused a performance boost. Like when your car has a stuck handbrake and you fix that, this will give you a speed boost. ¯\_(ツ)_/¯

[tildearrow]

Minor minor typo:

or even just playing games occassionally

This is very sad...

[fuzz]

My first thought when I saw the headline was "what quality did they sneakily reduce without telling anyone this time?" I guess it's less clever marketing than I expected.

[M@yeulC]

First I thought wow! But then I started to compare what could be compared [...] performance diff doesnt seem to be too extreme from numbers presented us in earlier benchmarks.

Well, I noticed how the error bars were quite large on nvidia hardware lately; that's probably what they noticed as well, so it definitely has an impact. Sometimes...

[perpetually high]

Can you please not promote Nvidia like that? "Offers big performance boost"? One would think that Nvidia developers managed to make a big breakthrough and increased performance. Did that happen? NO

What happened instead, was that they fixed a regression. But since most people read only the heading, impressions are different.

I'm confused what you're asking Michael here. Are you expecting him to be an AMD fanboy or stay impartial and deliver the latest news/benchmarks to his readers?

I'm seeing this trend a lot on here, what's the big deal? I understand AMD has open-source drivers, but you're still using proprietary firmware so I don't want to hear it. Let's just be happy performance is improving, regardless if it's NVIDIA or AMD.

[nuetzel]

Just when AMD was closing in on the performance gap, this had to happen...
Oh well, I don't see this as a bad thing.

To my recollection, a lot of these games were already getting higher framerates in Linux than they were in Windows. I'd be a little curious about the Linux vs Windows performance too.

Yes, please - Linux vs Windows performance!

[Zeioth]

I'm getting +15 FPS in QuakeChampions.

[coder]

I'm confused what you're asking Michael here. Are you expecting him to be an AMD fanboy or stay impartial and deliver the latest news/benchmarks to his readers?

No, it's about accurate reporting. The headline should explain that they fixed a performance regression.

Trying to make that a partisan issue creates the impression you're spoiling for a fight. Not everything is a Green vs. Red issue, you know?

[perpetually high]

No, it's about accurate reporting. The headline should explain that they fixed a performance regression.

Trying to make that a partisan issue creates the impression you're spoiling for a fight. Not everything is a Green vs. Red issue, you know?

I’m not interested in any of that, trust.

Michael said performance boost, but you want “they fixed a performance regresssion.” To which I say—who gives a f*ck? Probably just you guys. Many may've skipped the driver revision if Michael didn’t provide the benchmarks. Don’t lose sleep over it.

[coder]

Michael said performance boost, but you want “they fixed a performance regresssion.” To which I say—who gives a f*ck?

So, what if I was running a driver prior to the regression, saw the headline, updated my driver, and was left asking: where's my performance boost?

There's a meaningful distinction, here. To the extent this subject matters at all, I think it's worth getting right.

Don’t lose sleep over it.

Don't post while high.