NVIDIA 375.10 vs. Linux 4.8 + Mesa 13.1-dev AMD GPU Benchmarks

[zakhrov]

For those who want some closed source numbers. I have benchmarked my R7 250X with xonotic. Will post Heaven and Valley numbers soon. Bear in mind that my rig is way less powerful than Michael's http://openbenchmarking.org/result/1...ZAKH-161026972

[johnc]

Dude, I can post a dozen more links proving it -again- and the truth still wouldn't matter one tiny little bit to you. So keep up the bliss.

There is not a single, respectable benchmarking site out there that considers its benchmarking results unreliable due to nvidia lacking "visual fidelity".

[zakhrov]

For those who want some closed source numbers. I have benchmarked my R7 250X with xonotic. Will post Heaven and Valley numbers soon. Bear in mind that my rig is way less powerful than Michael's http://openbenchmarking.org/result/1...ZAKH-161026972

Those results didnt upload properly. First time using PTS here are the R7 250X results http://openbenchmarking.org/result/1...ZAKH-R7250XF46

[smitty3268]

Not sure what you're getting at here.

He has a Fury in the benchmarking results.

Right, and most people glance at the charts and assume it's a Fury X since that's what all the Windows review sites use. Because they got the Fury X for free, while Michael went with the cheaper (still expensive) option since he had to pay for it.

So everyone sees results for a card that they think is 10% faster than it really is, and assume the driver is worse than it really is.

It's like when a marketing team releases a bunch of slides with charts in it that aren't to scale. Anyone who takes a real look will clearly see what they did and roll their eyes a bit. But those marketing teams are doing it for a reason - a large chunk of people just glance at the charts and fall for it.

Smart marketing by Nvidia. Poor decision by AMD. There's a reason they sent out Fury X hardware to all the windows sites. It's because they didn't want to see charts like the one Michael is showing with the non-X going against higher-end competition.

[Passso]

Except when it comes to benchmarking and reviews what matters is how fast each GPU produces the same image with the same settings. Thus these differences lead to skewed results.

As I quoted : if you cannot see the difference it is the same image. If you want to have the "real" speed of your hardware make asumptions based on clock speed, number of cores, memory etc. like all "specialist" love to do. But in the end the FPS win.

[oooverclocker]

If someone can't see a difference or some people just don't notice a difference it doesn't make it the same image. There are people in the world that would hardly decide between a 20- times re-encoded 96kbit/s MP3 played on their cellphone and lossless 5.1 audio played on a 100.000$ Burmester sound system. It's the same for food, fashion and of course visual fidelity.
Do you really think that many people notice they are playing mainly on low textures in CoD:BO3 when they excessively exceed the VRAM on their GTX 970?
No! They just say: "It just stuttered a bit but now I have a nice framerate and I can play on Ultra."
They just assume that Ultra is Ultra cause they don't even know that Ultra is just variable without someone notifying you when the quality has been reduced because of your deficient hardware.

[artivision]

If most people cannot see the difference it cannot be considered as "cheating", but more "optimization", and those who do not make it are "brute forcers"

This is how the IT world improves!

You understand that we can do the same cheating with Mesa and some dozens linens of code right? But we don't want to. NV is not that superior as you think. So for NV goes like this: smells like shit, tastes like shit, looks like shit, feels like shit - what is it?

[tomtomme]

So exactly you ingored there GPUtest results for some reason and didn't bother answering why

Why?
You said it yourself - those are there to ignore them

[Passso]

You understand that we can do the same cheating with Mesa and some dozens linens of code right? But we don't want to. NV is not that superior as you think. So for NV goes like this: smells like shit, tastes like shit, looks like shit, feels like shit - what is it?

A explained I only judge by results: I see the same visuals with more FPS.