• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Nvidia’s GameWorks program usurps power from developers, end-users, and AMD

UrrrGHHh. The 290 had barely been on the market when Arkham Origins launched Matthew, they fixed most of the performance gripes the following day after release. What more do you want?

It's the same game engine. It COULDN'T BE MORE relevant. It's only not relevant to you because you've convinced yourself that GameWorks is the soul reason for it. When I'm showing you another Unreal Engine 3 title that NV cards are dominant in. If I install Arkham City, I bet you I could emulate near exactly the same scenario when using 8XAA.

Enough with the corporate conspiracies.

No they didn't. Joel found exactly the same results when he tested a 770 vs a 290X long after that Techspot review. Its not normal for a 770 to beat out a 290X. Same as its not normal for a 660 to beat out a 7950 boost. That scales down to the low end cards as well.

The numbers don't lie though Matt and look at all the other game bench threads and nVidia own in all of them, except Batman....

You have a pair of 290's, but if you want to play with FXAA, be my guest, I think MSAA looks great :p

What about Sleeping Dogs? :D

290X competes well with the titan in most things, just the 780TI that is normally a bit ahead, but then that was released after the 290 cards.

In most of the game bench threads there are very few water cooled 290 entrants, where as all the top Nvidia cards are generally water cooled. If Ranger (or other WC 290 users) bothered to post some game benchmarks im sure he'd give some of you a run for your the money looking at the 3dmark scores he just posted. Beating out the fastest 780TI's in gpu score by a hefty margin. :p

That has nothing to do with this though and does not explain what is going on. :)

EDIT

And on this note my new motherboard has arrived so off to install it, wish me luck. :D
 
FXAA (At least from what I recall) is an Nvidia tech, it could be down to their architecture that FXAA doesn't run so well on the GCN (Or it could be Nvidia crippling, but I think that's jumping the gun, hence the performance favour at the highest settings).

Dirt Showdown advanced effects are slower on Nvidia 6XX etc (Titan's fine, but the 680 gets trounced by a 7970), and I don't think sleeping dogs max AA is exactly great on 6XX either :p?

http://images.anandtech.com/graphs/graph6774/53358.png

FXAA costs 1 fps. :D

Dirt Showdown uses compute, 680 has gimped compute, titan does not. Can't be compared to this. Sleeping dogs also uses compute. Compute is part of dx11, GameWorks gimping is not.
 
If Nvidia were gimping, they'd give their cards the performance edge at the highest of settings, I'm happy I'm getting far better frames than the prior game (Which wasn't gameworks) whereas you're not happy on the basis of your own performance, you want it to be inline with Nvidia hierarchy (I can understand, but given City, don't agree with, I'm happy I'm getting better performance)

FXAA is crap anyway.

The majority don't have hardware capable of X8AA to overcome gimping though. The majority have low to mid range gpus where only FXAA is possible. Now if that's you you're at a serious disadvantage there, facing a huge performance penalty when in actual fact FXAA should cost 1 fps. Its not an issue for me or you obviously, but sometimes you have to look at the bigger picture rather than just the hardware you own. If this happened in more games, it would be a concern. If that's what you can expect from GameWorks title when you have an AMD card, its worrying. Those are my thoughts anyway.
 
But the performance across the board is still better than the none gameworks Batman Arkham City, regardless of Nvidia's performance, as an AMD GPU owner, regardless where you are on the hierarchy, you're getting better performance than Arkham City.

There's a 7770 giving playable performance at high details 1080p, that's a low range GPU.

Which to me makes it a none point currently.

Makes no difference. Why is performance gimped for AMD cards only when using FXAA, the least demanding form of AA? I do not consider 38 fps playable personally. Without the gimping the fps would be much higher than that. A 560 TI is averaging over 60 at 1200P and a 7770 is a t 38 fps yet the HD 7770 is a much faster card.
 
Because almost every U3 Engine game favours Nvidia cards? Doesn't matter which or how-which way it's always been that way. GameWorks is nothing but a convenient excuse. If it has propriatary features within it, then AMD will just use their own extensions anyway.

Maybe, if you're sooo convinced it's an issue, you should take it up with the developers willing to use these libraries in the first instance?

I can play the game just fine. Seemingly, so can Kaap. With everything enabled bar PhysX too. What a crying shame.

You have to look past your own hardware. 38 fps is not really acceptable on a 7770 which is one of the most common used gpu's for typical gamers. I have one in another pc and it can run bf4 1080P at high settings around 60 fps, so i know its a capable card.

The engine has nothing to do with it either, if that was the case then x8AA would not shift the results so dramatically.

No point taking it up with the devs, just raising a valid point. No one can argue against it other than 'it works fine for me with my high end gpu'.
 
Yes, it WOULD shift the results. How can you not understand this. The memory bus on Hawaii is so much better, so any advantage is lost. This would happen in the Thief thread as well if we applied more. Sleeping Dogs also favours NV cards, but it has a weird overlaying technique so it's not directly comparable.

You're just fixiated on this GameWorks library stuff because it's convenient. You need to draw a line somewhere, it could be that it's just not very well optimised for AMD cards anyway. Although I'm sure if someone did some tests now, performance would be improved.

In no circumstance should a 770 beat a 290X. Neither should a 660 beat a 7950 boost, or a 560TI beat a 7770. Not without serious gimping going on.
 
The HD7770 is by no stretch of the imagination a much faster card than the 560Ti, when it comes to gaming, overall the 560Ti is the superior card.

People on those graphics cards would fine 38 FPS perfectly playable, hell people game on FPS figures we'd find unbearable.

While in an ideal situation, the AMD cards shouldn't be gimped at all, I'd much rather get better performance, but less than Nvidia (With FXAA, which I repeat, is crap), but better than the prior game, than have City, which was just crap FPS all around.

Sorry i meant the 550TI. The 7770 is just 1 fps faster than it but its a superior card. Anyway, it matters not. There is no way a 770 should be beating a 290X at 1080P but 5fps no matter how anyone tries to dress it up.
 
Back
Top Bottom