6950 vs GTX 460 768MB - Why does Nvidia beat the Radeon in Civ5???

[AtenRa]

That is it?

http://translate.googleusercontent....le.com&usg=ALkJrhiHa5XXx-4A_fr8JeQGsJ5L1baVZg

Click on any of the Civ 5 benchmarks. In fact, look at all of them - they are all exactly what you would expect to see.

The only problem is with Civ 5, and only because the 480 beats the 580 in that. To me thats a clear case of nVidia not bothering to fix the actual game issues when it was easier and more profitable to fix the benchmark instead.

oh, my mistake

didn't see the links in the benchmakrs, thx

 

[Lepton87]

I want the url of the REVIEW this diagram was taken from
http://pclab.pl/zdjecia/artykuly/majkel/GTX590/wykresy/civ_1920_aa8.png

not the GTX590 review

Actually I took that diagram from GTX590 review.

 

[AtenRa]

This mast be a mistake, they probably misplaced the numbers and GTX580 is 10% faster than GTX480.

 

[cusideabelincoln]

I want the url of the REVIEW this diagram was taken from
http://pclab.pl/zdjecia/artykuly/maj...v_1920_aa8.png

not the GTX590 review

He gave it to you.

 

[bryanW1995]

Hopefully one day NVIDIA will allow me to explain what they did to improve Civ V so much. I found out what they did, however I'm not allowed to talk about it (and boy I'm dying to). It makes all the CPU limitations make sense though, and it somewhat reshaped my view on DX11. Honestly I'm surprised the eggheads over at Beyond 3D haven't already figured this one out; it seemed kind of obvious in retrospect.

I assume that once AMD figures it out then there won't be any reason to hide what they did. I'm extremely curious now.

edit: HA, I guess this is a good reason to spend a few days in RL instead of at AT... it's like I entered a time machine and got to see the answer right away!!

 

[Black96ws6]

Great info Ryan, and to everyone else with the suggestions, thank you.

I hope more games implement the multithreaded rendering feature!

Side note, I tried out AVP last night on my 1090t, using the stock cooler so can't go higher than this - oc'd to 3.8ghz, turbo mode on and set to 4ghz, NB oc'd to ~2600Mhz - and AVP flies on the 460 OC'd to ~830mhz on high settings. Granted I just tried out a few maps in Survivor mode @1920res (masoleum and cell 5 or something, can't remember), however it really looks like I don't need to upgrade after all.

I just seem to have the upgrade bug lol. Everything's fine but just have the urge! Can anyone else relate? Civ 5 is very fast for me, it's just on the large maps in the late game it slows down, however, in reading the comments here, it looks like that's true for most as well, so I think I'll just enjoy what I have right now and if things start to bog down in future games, then upgrade...

 

[bryanW1995]

Thanks Ryan, good stuff to digest and give some life back in to dx11 relevance. Now hopefully Crytek is working on some of this to implement in Crysis 2.

haha, crytek has sold out. We'll be lucky to get much of anything in dx11 out of Crysis 2, it was built/designed to be used in dx9 and now is getting ported over to dx11.

 

[bryanW1995]

But isn't that the whole point of this entire thread?
NV has fixed its driver problems -> performance increases -> the odd capped situation where the GTX460 is faster than the 480 is solved -> performance should now reflect, well, performance -> GTX580 should be ahead of GTX480.

470 and 460 faster than 480 on old (not fully functional) driver.

580 > 570 > 480 > 560 > 460
Properly functional driver. Properly reflective scores because they aren't capped by driver limitations.

The benchmark quoted was suggesting that AMD outscores NV even with the fixed driver because the performance of the built in benchmark doesn't reflect game performance in terms of NV vs AMD difference.
The implication of the other graph is not that NV has a lead, but that the fixed driver is in fact not fixed, or the benchmark is wrong, because with a fixed driver it should be that the GTX580 is faster than the 480, especially when the 580 and 570 have different performance to each other and it scales with the GTX590, meaning performance isn't capped.

Therefore the high GTX480 is completely unusual and not what you would expect under any circumstance. If it was because the cards were capped, the GTX590 shouldn't scale and the 580 and 480 should be at about the same level, the 480 shouldn't have a 10% lead.

They also have some odd results in other Civ 5 benchmarks, like the 6970 being soundly beaten by the 5770 and sometimes almost the 5750. Since they use the same architecture, and have the same functional units, and the 6790 is improved in many ways, it's odd that it would ever be slower, especially when you crank up AA and res (1920x1200/8xAA and the 5770 is faster than the 6790, despite the massive bandwidth advantage of the 6790 and the improved tess performance and the equal in every other aspect specs, on paper at least).

6970 is vliw4, 57x0 was vliw5. It should still beat those much older cards obviously, but it's defintely a newer arch.

edit: ok, are you talking about 6790 or 6970?

Great info Ryan, and to everyone else with the suggestions, thank you.

I hope more games implement the multithreaded rendering feature!

Side note, I tried out AVP last night on my 1090t, using the stock cooler so can't go higher than this - oc'd to 3.8ghz, turbo mode on and set to 4ghz, NB oc'd to ~2600Mhz - and AVP flies on the 460 OC'd to ~830mhz on high settings. Granted I just tried out a few maps in Survivor mode @1920res (masoleum and cell 5 or something, can't remember), however it really looks like I don't need to upgrade after all.

I just seem to have the upgrade bug lol. Everything's fine but just have the urge! Can anyone else relate? Civ 5 is very fast for me, it's just on the large maps in the late game it slows down, however, in reading the comments here, it looks like that's true for most as well, so I think I'll just enjoy what I have right now and if things start to bog down in future games, then upgrade...

I would tell you to upgrade the cpu, but with the improved multicore scaling I don't know how much better it will get on fewer/faster cores like an SB system. Keep your eyes peeled on the bulldozer release in june, it looks like many AM3 mobos could end up supporting a drop in upgrade.

 

[Ryan Smith]

Ryan, from what it sounds like. Is this multi-threaded rendering like out of order execution on the CPU side?

Eh, if you want an analogy I'd go with something closer to a waitress simultaneously taking orders from several tables on one order sheet, submitting them to the cook at once, and managing to deliver each order to the right person at the right table in the end.

 

[Idontcare]

Making this a sticky for a while, post #28 from AnandTech GPU Editor Ryan Smith on the topic of multi-threaded rendering deserves more exposure for the benefit of the community. Check it out.

Idontcare
Super Mod

 

[SirPauly]

I am glad it was stickied, because there are some damn good posts in this thread and would of missed them, thanks!

 

[Powermoloch]

Great read, thanks to Ryan :thumbsup:

 

[3DVagabond]

The important info in this thread is about the multithread rendering. That one post is going to get lost mixed in with all of the other posts.

Change the title to "The benefit of multi-threaded rendering", and delete all of the other posts. Or, better yet, have Ryan do an article on it and sticky it instead. As it is now, we've got a stickied thread because of performance in one game being better with one vendor.

 

[insurgent]

Is mjolnir (11.4 preview) the driver that will start supporting multi-thread rendering? AMD claims up to 70% performance gains in Civ V.

 

[Spjut]

I would like to see benchmarks showing what gains DX10/10.1 hardware get with the multithreaded rendering

"The power of DirectX 11" article mentions that even if AMD/Nvidia don't do some driver work to get the multithreaded rendering working to its full extent, it will still "work"
I hope both AMD and Nvidia work on the drivers for their DX10/10.1 cards to get the multithreaded rendering working as good as possible though

 

[mnewsham]

Is mjolnir (11.4 preview) the driver that will start supporting multi-thread rendering? AMD claims up to 70% performance gains in Civ V.

That doesn't mean it will actually increase 70%

 

[wahdangun]

wow, this is what crytek should do, damm I will give more respect to fraxies now, maybe next moth i will buy Civ5 games just to appreciate the developer. and shame on you AMD

 

[insurgent]

That doesn't mean it will actually increase 70%

I know and that's not even my question. I was asking if mjolnir is the driver that supports multi-thread rendering since they claim huge gains, and I don't think I need an english lesson even if it's not my native language.

wow, this is what crytek should do, damm I will give more respect to fraxies now, maybe next moth i will buy Civ5 games just to appreciate the developer. and shame on you AMD

My sentiments exactly.

 

[LiuKangBakinPie]

What about Flight Sim X. Another highly intensive game.
For the Amd Nvidia issue drivers behaving differently to games. This is a feature I enjoy with nvidia. The profile system they have for most games. Ok they hide most of the settings from the end user for some strange reason but thanks to that there's Nhancer. Lot of compatibility settings you can change which can improve how the driver behaves with the game. Been using Nhancer for years great app.

Then there's some games aka bad ports *cough GTAIV cough* that's just plain junk coding horrors which I think should never be played. But me maxing it out are great for bragging about you hardware in a forum

 

[1h4x4s3x]

After reading through the thread it seems that Nvidia optimized its drivers for the integrated bench but not so much for the actual game?
I'm surprised reviewers wouldn't mention that.

 

[LiuKangBakinPie]

After reading through the thread it seems that Nvidia optimized its drivers for the integrated bench but not so much for the actual game?
I'm surprised reviewers wouldn't mention that.

You mean with the profiles? No they use the quality setting when they bench. The specific profile won't be use then unless you choose it.

 

[1h4x4s3x]

It's worth to add that the advantage that nVidia holds in CIV 5 is only apparent when you use the built-in benchmark. It mostly disappears when you test in actual game-play with Fraps.

http://pclab.pl/zdjecia/artykuly/majkel/GTX590/wykresy/civ_1920_aa8.png

I mean this.

 

[notty22]

After reading through the thread it seems that Nvidia optimized its drivers for the integrated bench but not so much for the actual game?
I'm surprised reviewers wouldn't mention that.

It is mentioned at times, but more importantly the benchmarks still measure GPU PERFORMANCE.

Civilization V
In addition to the compute shader test we've already covered, Civ V has several other built-in benchmarking modes, including two we think are useful for testing video cards. One of them concentrates on the world leaders presented in the game, which is interesting because the game's developers have spent quite a bit of effort on generating very high quality images in those scenes, complete with some rather convincing material shaders to accent the hair, clothes, and skin of the characters. This benchmark isn't necessarily representative of Civ V's core gameplay, but it does measure performance in one of the most graphically striking parts of the game. As with the earlier compute shader test, we chose to average the results from the individual leaders.

http://techreport.com/articles.x/20629/6

 

[railven]

It is mentioned at times, but more importantly the benchmarks still measure GPU PERFORMANCE.


http://techreport.com/articles.x/20629/6

So it measures real time non-playable scene? A cutscene? We're benching cutscenes now!?

 

[Bill Brasky]

I don't know how I missed this thread the first time around (being a sticky and all), but what a great read. I enjoyed it enough I thought it was worthy of a bump!

Thanks for the input, Ryan.