A Four Way Comparison For Alien: Isolation Shows Off The Performance Difference

Quoting: MblackwellI wonder what the difference is if you run Nvidia driver 355.11, since that's what's recommended in the README.

There is no difference, not with 355.11 or any other driver.

dubigrasu, how much performance is lost between SteamOS 1.0 and 2.0? How many FPS? Do you think the brand new NVIDIA 358.13 driver might make any difference?


Last edited by Xaero_Vincent on 19 November 2015 at 5:51 am UTC

Quoting: XaeroVincentdubigrasu, how much performance is lost between SteamOS 1.0 and 2.0? How many FPS?

I haven't tested Alien Isolation with SteamOS 2.0 if that's what you're asking.
But 2.0 has a constant loss in performance with other games I tried, so I just stopped using it.

QuoteDo you think the brand new NVIDIA 358.13 driver might make any difference?

Maybe for other games, IDK, but it didn't for AI.


Last edited by dubigrasu on 19 November 2015 at 6:04 am UTC

Quoting: dubigrasu

Quoting: XaeroVincentdubigrasu, how much performance is lost between SteamOS 1.0 and 2.0? How many FPS?

I haven't tested Alien Isolation with SteamOS 2.0 if that's what you're asking.
But 2.0 has a constant loss in performance with other games I tried, so I just stopped using it.

Well I mean more like a ballpark figure... is it like 10 or 20 FPS slower across the board or even more? Mainly curious about an ballpark figure across any game from your testing.

It's quite upsetting to see the flagship 2.0 being worse than 1.0, especially since Valve discontinued 1.0. Valve needs to kick it into overdrive and get serious here if they intend to keep this going. I know it's not reasonable to expect all games to run as good as Windows but things can certainly improve if they've gotten worse than before! I hope you decide to show Valve benchmark results between 1.0 and 2.0 soon.


Last edited by Xaero_Vincent on 19 November 2015 at 6:09 am UTC

Quoting: XaeroVincentWell I mean more like a ball park figure... is it like 10 or 20 FPS slower across the board or even more? Mainly curious about an ballpark figure across any game from your testing.

Well, Interstellar Marines is the first game where I noticed this. The difference in FPS is 10 at most but stutters like crazy.
Dying Light has more or less the same FPS but is too choppy to play.

double post


Last edited by dubigrasu on 19 November 2015 at 6:19 am UTC

Quoting: dubigrasu

Quoting: XaeroVincentWell I mean more like a ball park figure... is it like 10 or 20 FPS slower across the board or even more? Mainly curious about an ballpark figure across any game from your testing.

Well, Interstellar Marines is the first game where I noticed this. The difference in FPS is 10 at most but stutters like crazy.
Dying Light has more or less the same FPS but is too choppy to play.

Thanks for the info. Hmm. Have you by chance tried these games on other modern distributions? Could it be the more modern Linux OS stack (compared to SteamOS 1.0) causing problems? I wonder how games would perform doing a SteamOS 1.0 vs 2.0 vs Ubuntu 15.10 vs Archlinux / Antergos benchmark with all the same latest Nvidia drivers.


Last edited by Xaero_Vincent on 19 November 2015 at 6:27 am UTC

Quoting: XaeroVincentThanks for the info. Hmm. Have you by chance tried these games on other modern distributions? Could it be the more modern Linux OS stack (compared to SteamOS 1.0) causing problems? I wonder how games would perform doing a SteamOS 1.0 vs 2.0 vs Ubuntu 15.10 vs Archlinux / Antergos benchmark with all the same latest Nvidia drivers.

Don't know about Arch, but Ubuntu 15.10 also has a somewhat lower performance.

While it's lower on the Linux side it is still respectable framerates on the Nvidia side. Hopefully AMD can get some more money to spend on their driver teams.

Unlike many I understand why they are behind on the driver side of things. AMD had a profit in 2014 of $51 million while Nvidia had a profit of $147 million in just the fourth quarter of last year. When you make more than 7 or 8 times the money of the competition it is easier to outspend them in R&D. I'm not taking sides, just pointing out a fact.

I prefer AMD's more open type of innovations such as tressFX so I will continue to choose them when they can deliver a competitive product. I wouldn't take a Fury X over a GTX 980 Ti, i'm not blind, but I would take a 390X over a GTX 980 and a 390 over a GTX 970.

Quoting: PublicNuisanceWhile it's lower on the Linux side it is still respectable framerates on the Nvidia side. Hopefully AMD can get some more money to spend on their driver teams.

Unlike many I understand why they are behind on the driver side of things. AMD had a profit in 2014 of $51 million while Nvidia had a profit of $147 million in just the fourth quarter of last year. When you make more than 7 or 8 times the money of the competition it is easier to outspend them in R&D. I'm not taking sides, just pointing out a fact.

I prefer AMD's more open type of innovations such as tressFX so I will continue to choose them when they can deliver a competitive product. I wouldn't take a Fury X over a GTX 980 Ti, i'm not blind, but I would take a 390X over a GTX 980 and a 390 over a GTX 970.

Those profit figures are interesting, given that AMD now has a monopoly over console GPUs and consoles are doing pretty well. How come nVidia makes so much more profit?