Unigine Valley Benchmark

[biostud]

Extreme
Score: 2249
FPS: 53.8
Min FPS: 26.9
Max FPS: 99.6

ExtremeHD
Score: 1738
FPS: 41.5
Min FPS: 22.5
Max FPS: 74.7

 

[Phynaz]

I'm going to have to call this - Valley 1.0 runs much better on Nvidia hardware.

Of course it does, it's sponsored by Nvidia.

 

[notty22]

Of course it does, it's sponsored by Nvidia.

You have some kind of background to Nvidia sponsoring 'Valley' exclusively?

http://unigine.com/company/
Partners

UNIGINE Corp. is a registered developer of Apple Inc., AMD Inc., NVIDIA Corporation and Creative Technology Ltd. As a registered developer we work closely with our ecosystem partners in order to achieve the best level of performance and compatibility for all of our products.

http://www.phoronix.com/scan.php?page=article&item=unigine_valley_benchmarks&num=1

Benchmarks Of Unigine Valley

Published on February 13, 2013

http://www.phoronix.com/review2forums.php?view=unigine_valley_benchmarks
Continuing in the exclusive coverage of the yet-to-be-released Unigine Valley, here are some initial performance results for this visually-amazing multi-platform tech demo / benchmarks when using the OpenGL 3.2 Core renderer on Ubuntu Linux. A range of NVIDIA GeForce and AMD Radeon graphics cards were used for this initial testing of Unigine Valley. There's also benchmarks in this article of Unigine Heaven 4.0, which was just released yesterday.

​

This weekend I was allowed by Unigine Corp to spill the beans on Unigine Valley 1.0 and Unigine Heaven 4.0 (there's lots of screenshots in that article for those interested). Heaven 4.0 was released on Tuesday while Valley 1.0 is coming for a public release on Thursday to Windows, Mac OS X, and Linux users. For the past few days I have been busy running OpenGL benchmarks on Valley 1.0 and Heaven 4.0 to see how the performance stacks up.

​

Unigine Heaven 4.0 is much more demanding than the Unigine Heaven 3.0 release from last year with the frame-rates for NVIDIA and AMD graphics processors both being sharply lower. Unigine Valley 1.0 is also able to drop all current-generation graphics cards to their knees. For reference, on the same assortment of graphics cards, I also re-ran Unigine's older tech demos -- Unigine Tropics and Unigine Sanctuary.

​

The graphics cards tested for this comparison on the NVIDIA side were a GeForce 9800GTX, GT 220, GT 240, GTX 460, GTX 550 Ti, and GTX 680. On the Radeon side there was the Radeon HD 5750, HD 6570, HD 6770, HD 6870, HD 6950, and HD 7950. The latest proprietary graphics drivers (NVIDIA 310.14 and AMD fglrx 9.1.11) were used from the Ubuntu 13.04 development system backed by an AMD FX-8350 Eight-Core processor. Only the proprietary AMD/NVIDIA graphics cards were tested since Mesa/Gallium3D isn't yet in a state to support the latest Unigine Engine.

 

[blackened23]

Very odd. With the old 3dmark11 and unigine benchmarks, nvidia reigned 3dmark11 while AMD did better in the unigine DX11 heaven benchmark.

Now it appears that the exact opposite has happened - AMD better in the new 3dmark, while nvidia is better in the new unigine. Although i'm sure both sides will tweak drivers.

 

[Rhezuss]

Glad you got it figured out!

I now realize that the reason you had a 30fps cap in extreme HD but hit 40fps in extreme is that the extreme benchmark was actually jumping between the 30fps and 60fps VSync caps. Hence why both improved without VSync.

Well I forgot to thank you and Russian for the advices. Greatly appreciated.

 

[tviceman]

Of course it does, it's sponsored by Nvidia.

A friend told me Nvidia wrote Valley themselves and tessellated the dirt underneath the grass to purposefully hurt performance on AMD cards. BOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOM!

 

[Phynaz]

You have some kind of background to Nvidia sponsoring 'Valley' exclusively?

Go look it up from when Heaven was released. Hint: It's in their blog.

 

[Tookie123]

gtx 670@ 1201/7000

FPS: 46.5
Score: 1947
Min FPS: 25.0
Max FPS: 84.0

System
Platform: Windows 8 (build 9200) 64bit
CPU model: Intel(R) Core(TM) i7-2600K CPU @ 3.40GHz (3399MHz) x4
GPU model: NVIDIA GeForce GTX 670 9.18.13.1396 (2047MB) x1

Settings
Render: Direct3D11
Mode: 1920x1080 8xAA fullscreen
Preset Extreme HD

 

[notty22]

Go look it up from when Heaven was released. Hint: It's in their blog.

Ok , that's a no. Both companies were involved. And Heaven was one of the only programs/benchmarks that could display dx11 after the 5870 launched. Which was the first and only dx11 hardware, for months, besides the 5850. The other dx11 software was Dirt2.


edit: Heaven 1.0
Post Date: Oct 23, 2009 6:54am PDT

//////



http://benchmarkreviews.com/index.p...sk=view&id=477&Itemid=72&limit=1&limitstart=3

• Native support of OpenGL, DirectX-9, DirectX-10 and DirectX-11​
• Comprehensive use of tessellation technology​
• Advanced SSAO (screen-space ambient occlusion)​
• Volumetric cumulonimbus clouds generated by a physically accurate algorithm​
• Dynamic simulation of changing environment with high physical fidelity​
• Interactive experience with fly/walk-through modes​
• ATI EyeFinity support ​

edit added link of review before Fermi launched.

 

[Keysplayr]

Some fun with a GTX660

 

[YBS1]

Didn't bother to crank up the cpu, assumed it was just like Heaven and default vs. 4.6GHz+ wouldn't show a bit of difference. Might try later...

 

[Xenphor]

Does anyone else get a slight hiccup at scene 10 with the closeup of the rocks? Sometimes it'll be fine but other times I get a slight pause as if it's loading something.

 

[Rhezuss]

I must confess, the Gigabyte WF3 is one hell of a nice cooler. It's silent and highly effective. I was "lucky" enough to get a card with BIOS FZ1 with a locked voltage of 1.25v and a factory OC of 1000MHz...you could call it a HD 7950 GHz Edition.

Although I could get a bit more OC out of it I stayed in the comfort and safe zone at 1150/1450 which give me an awesome gaming experience in all my games (Crysis 2, Thw Witcher 2, Guild Wars 2, WoW, Diablo 3, Dishonored, etc). On top of that I got Crysis 3 and Bioshock Infinite for free. I highly recommend this card to any potential buyers.

 

[Soulkeeper]

Just ran mine in linux with pts

30fps here
1920x1200 no AA/AF
7850 @ 1000/1200

 

[omeds]

 

[moonbogg]

Serious? Did you just beat my score? OK, lock this thread right now.

 

[Protomize]

HD 7970 @ Stock | i5-3570k @ 4.7GHz

HD 7970 GE @ Stock | i5-3570k @ 4.7GHz

HD 7970 @ 1220/6900 | i5-3570k @ 4.7GHz

 

[hyrule4927]

Gave it a try on my laptop, which honestly performed better than anticipated.

GTX460M @ 675/1250

GTX460M @ 800/1600

Going to have to see what my 7950 can do, but clearly the laptop has set a very high standard already.

 

[Red Hawk]

Well something is fishy about my card specifically since a HD7870 OCed outperforms my HD7950 OCed...

The 7870 and 7950 have had an odd relationship since release. The 7870 is made to clock higher than the 7950, and they both have the same dual geometry engine (Pitcairn and Tahiti have the same geometry engine in general), and they have the same amount of ROPs. This can lead to the occasional situation where the 7870 pulls ahead of the 7950 in geometry-limited benchmarks.

 

[Shephard]

don't you have to pay for these benchmark programs?

 

[Protomize]

don't you have to pay for these benchmark programs?

The basic editions are free.

 

[Shephard]

The basic editions are free.

what are the reasons for these benchmarks anyway? Are the numbers just for bragging rights?

Isn't it better to test a gpu on a real game to see fps and temperatures?

 

[blastingcap]

what are the reasons for these benchmarks anyway? Are the numbers just for bragging rights?

Isn't it better to test a gpu on a real game to see fps and temperatures?

Stop making so much sense.

 

[Keysplayr]

what are the reasons for these benchmarks anyway? Are the numbers just for bragging rights?

Isn't it better to test a gpu on a real game to see fps and temperatures?

This particular benchmark, as well as 3DMark, are in it to make money. Why else would they continually come out with newer benchmarks? For the fun?

 

[Termie]

what are the reasons for these benchmarks anyway? Are the numbers just for bragging rights?

Isn't it better to test a gpu on a real game to see fps and temperatures?

Benchmarks are good for testing behavior with overclocking, or to check whether a stock system is set up properly. Games, and even the built-in benchmarks in games, aren't quite as precise, and often rely on significant amounts of CPU power, making GPU comparisons difficult.