Windows 10, Nvidia being left in the dust.

In any case, MSAA performance is a moot point in modern games thanks to deferred rendering, thanks to consoles.

CONSOLES STOLE OUR MSAA!!! :p


That is not the reason for deferred rendering :rolleyes:

Here is a very small account about forward and deferred rendering http://www.anandtech.com/show/5458/the-radeon-hd-7970-reprise-pcie-bandwidth-overclocking-and-msaa/3 and they do mention the memory constraints of MSAA on deferred rendering.

They also talk about Forward Plus rendering, proposed by AMD and used at least on the Forza series on the xb1.
 
I have 2 7990s crossfired. For Windows 10 the drivers got a lot better. The original first driver i couldn't even install without the system crashing during the installation. Had to wait awhile. Now its good though. 2.0 wddm seems a lot better than 1.2 it was before. Online is just smoother. Better hardware acceleration. I could tell right off the bat. Glad they are updating the 7000s still :).
 
That is not the reason for deferred rendering :rolleyes:

Here is a very small account about forward and deferred rendering http://www.anandtech.com/show/5458/the-radeon-hd-7970-reprise-pcie-bandwidth-overclocking-and-msaa/3 and they do mention the memory constraints of MSAA on deferred rendering.

They also talk about Forward Plus rendering, proposed by AMD and used at least on the Forza series on the xb1.

Interesting.

"One of the biggest technical culprits (as opposed to quickly ported console games) is the increasing use of deferred rendering, which makes MSAA more difficult for developers to implement."
 
The biggest reason to use deferred rendering is for lighting.

Deferred rendering separates out the geometry, which means lighting is much less cosly as it doesnt have to render the geometry.

The side effect of course is now hardware based AA is confused so it takes a lot more work to get it implemented. Rather than forcing it from the control panel any time you want, the engine itself has to support it.

Forward plus should take care of both problems.
 
WCCF picked up the Overclock thread:
http://wccftech.com/amd-r9-fury-x-performance-ahead-nvidia-980-ti-latest-drivers/

There was also a summary of changes in TPU's benchmarks which can be found here -- along with a list of games that changed in their test suite.

So TPU removed a bunch of games Nvidia does well in from the composite, then the braintrust at WCCF cherrypicked the summary and reworked it into the perfect clickbait narrative "AMD gets performance boost!"

Remind me again why you're always linking to wccftech?
 
Last edited:
So they removed a bunch of games Nvidia cards did well in from the composite, and deduced "AMD gets performance boost!"

What total idiocy.
WCCF jumped straight to drivers and ignored all other changes in TPU's test suite. That's just fucked up. Even I noted the OS and line-up of games. :rolleyes:
I should check the Overclock thread and see if they made the same erroneous conclusion.
 
So TPU removed a bunch of games Nvidia does well in from the composite, then the braintrust at WCCF cherrypicked the summary and reworked it into the perfect clickbait narrative "AMD gets performance boost!"

Remind me again why you're always linking to wccftech?

Uh what? the selection of games is pretty broad, your statement is unfounded.

You're upset that AMD released drivers that increase the performance of their cards? I don't think you even read the summary or the article.

I swear, I think these forums need infractions for purposely misleading fanboy posts, stupid posts, and troll posts.
 
Uh what? the selection of games is pretty broad, your statement is unfounded.
You're upset that AMD released drivers that increase the performance of their cards? I don't think you even read the summary or the article.

Straw man more. I wasn't implying that the game selection was biased. The point was more variables changed between tests than just AMD's driver. My comment was squarely focused on the idiocy of wccftech taking something and twisting it around for the clickz. My GPU is at the top of the heap in any test, so the minutiae of which lesser cards beat which others hardly concerns me.

I swear, I think these forums need infractions for purposely misleading fanboy posts, stupid posts, and troll posts.

Agreed.
 
Last edited:
I really don't understand how the fanboy argument can even be used here when someone argues this article. They are clearly using way too many different variables to "compare" the results.
 
when you use a larger sample size ( number of games ), the impact of adding and removing games, is less noticeable, and lets be practical, I don't buy 1 videocard for 3 games, and another video card to play a different set of games, I don't think there are any or many people that would.
 
Back
Top