An uncertain moment of truth: Post your Unigine Heaven benchmark score/screenshots

[Voo]

Bold ^
Ermm... Isn't that what you might want? A card that holds up when the going gets rough?

Oh and I thought most people would want a card that runs actual games smooth and not benchmarks that use a feature in a way it was never intended to be used - well, de gustibus non est disputandum

There's one thing for sure: Nobody would ever use tesselation in the same way as Unigine does, because there are much better ways to obtain similar visual effects for lots of things. As far as I know Tesselation should be there to increase the number of polygons at runtime to make structures look better if you zoom in and not to create stairs from a handful polygones the way Unigines does.

So all that matters is if the advantage of Nvidia is also noticeable in real games and there we have to wait for some real benchmarks I fear.

 

[GaiaHunter]

I've read a few different places that Nvidia didn't use AA for the test, assuming that is indeed the case, you have to wonder why that is. Most people who buy this level of video card will want AA. Maybe the going gets a little rough for Fermi with AA in this test? Maybe it's not true, they did use AA? Maybe something else? I'm sure Nvidia will use the settings that showcase Fermi in the best light.

You don't have to read anything - just see for yourself.

http://www.youtube.com/watch?v=vpdPSZB8A8E&feature=related

Unigine Heaven v1.1

1920 x 1080 0xAA 1xAF

At around 1:06 you can see all this information.

Again I don't see the point of comparing results of cards using different versions of the benchmark.

The rumour was that the GTX 480 had image issues using the v1.0 - don't know if that is accurate or not.

 

[GaiaHunter]

With a 5870 in the rig in my sig, at 1920x1200 all settings maxxed, I
Get a score of 682 at 22.2 fps. This card does take a serious hit turning tesselation on. Example would be the dragon in the courtyard scene. Tesselation off, I get around 44 fps. Turn it on, framerate drops to 15-17 fps. I'll post screenshots later. On my blackberry right now.

What version of the benchmark are you running? And if you have both versions is there a difference?

And what is the FPS in the dragon scene @ 1920 x 1080 0xAA 1xAF?

Ty.

 

[Keysplayr]

What version of the benchmark are you running? And if you have both versions is there a difference?

And what is the FPS in the dragon scene @ 1920 x 1080 0xAA 1xAF?

Ty.

1.0

Hold on, I'll post 1920 x 1080 0xAA 1xAF results in a few minutes.

Allright, here we go:

Unigine Heaven Benchmark 1.0
1920x1200
5870 Cat 10.2.
Windows 7 64
Rig in sig.

1920x1200, DX11, Shaders=High, tesselation=enabled, Anistropy=1x, AA=off

1920x1200, DX11, Shaders=High, tesselation=disabled, Anistropy=16x, AA=8x

1920x1200, DX11, Shaders=High, tesselation=enabled, Anistropy=16x, AA=8x

The score I gave before was from memory. I got the 22.2 fps right but was generous on the score.

 

[jaggerwild]

Specs:
E5200 AT 2510Mhz
TP45HP Biostar
G-Skill1100@802
ATI RADEON 4870 X2


Need more ask oh Version 1.0

 

[Termie]

5850 @850/1200

1440x900, Dx 11, High settings/Tessalation, AA 2x AF 8x
FPS: 39.0
Score: 982

1920x1200, Dx 11, High settings/Tessalation, AA 0x AF 4x
FPS: 34.6
Score: 872

 

[Kindjal]

My new Sapphire HD 5870 Vapor-X seemed to handle the Heaven benchmark well.

 

[Keysplayr]

Specs:
E5200 AT 2510Mhz
TP45HP Biostar
G-Skill1100@802
ATI RADEON 4870 X2


Need more ask oh Version 1.0

10x7? Can your monitor go higher?

 

[jaggerwild]

10x7? Can your monitor go higher?

Yeah I just ran it at native settings to see as I've never run it before and there's so many adjustments too it. BRB.

 

[james1701]

These were done on a Q6600 @ 3.6ghz using an 8X multi and 4890.

 

[SlowSpyder]

My new Sapphire HD 5870 Vapor-X seemed to handle the Heaven benchmark well.

I have a similar system to you (though I have a Phenom II @ 3.6GHz). My Vapor-X 5870 is up to 930MHz on the core so far, I just keep pushing it 5-10MHz at a time and testing. I wonder if this test uses CPU power at all? You should score a bit higher with your CPU if it does. I'm gong to run the test at your settings and see how I do. I'll also run with my 5870 at your speed (875MHz) and see how my score compares, just to see how much the CPU matters in this test. I'll post later, have to run off to work. :/

*Edit: 43.5FPS / 1095 @ 1680x1050 / 4xAF with the system in my sig as is. Same settings you ran. Just for comparrison. So, the extra 55MHz over the stock 875MHz VaporX is worth a few FPS. If I ca get another ~50MHz and the FPS scales the same as it did with the current overclock, then that would be quite worth while if it translates to a similar boost in games. Anyway, just wanted to compare. I have the screen shot saved, just have to find a place to host it so I can link it if anyone cares. Off to work now though.

 

[MrK6]

I don't trust the frame counter. Despite it being well over 60fps the whole time the rendering looked choppy to me. I tried everything maxed out at 2560x1600 and it barely hits 2fps. FAIL!

Yea, one thing I noticed was that no matter what the frame rate, I could see some 'hitching' here and there... like there's be an instant where the rendering would get choppy despite the FPS showing a rate that should look smooth.

Sounds like you're both overclocked too high. Take your cards to stock clocks and see if it does it again. Usually it's the RAM being overclocked to highly that produces the "hitching" effect.

EDIT: Here's a run:
1920x1200, 4xMSAA, 16xAF, DX11 w/ all options on "high"

5850 Crossfire @ 1020/1250

 

[GaiaHunter]

Nvm - I thought that was a run of your 5870 MrK6

 

[SlowSpyder]

Sounds like you're both overclocked too high. Take your cards to stock clocks and see if it does it again. Usually it's the RAM being overclocked to highly that produces the "hitching" effect.

EDIT: Here's a run:
1920x1200, 4xMSAA, 16xAF, DX11 w/ all options on "high"

5850 Crossfire @ 1020/1250

I haven't touched memory speed in either my system or video card. My system memory is at 800MHz and the rated 2.1 volts, my video card memory is at the factory setting of 1250MHz (5GHz effective). So I don't think it would be the memory.

A game I play, Titan Quest, it also does this regardless of CPU/GPU/Motherboard/Memory, etc. I've also read other people complain about this all over the internet, so I don't think it's my system (at least not with Titan Quest). It might be the game/bench retreiving data from the hard drive, though I have 8GB of ram, if it's not coded to use all that space it may still have to read from the drive from time to time and cause a hiccup. But, the frame rate readings seem unaffected even though I see it happen from time to time. Just one of those odd things I guess.

 

[jaggerwild]

10x7? Can your monitor go higher?

Yeah I just ran it at native settings to see as I've never run it before and there's so many adjustments too it. BRB.

Here it is run at 1080, also with 2X AA, 800/1000 on the GPU All the rest as per my other posting.

 

[Apocalypse23]

Some folks have posted DX10 tests and some have used lower filtering settings. I think we need to see more of the tests using exactly these settings or higher for a direct comparison:

1440x900/1920x1080/1920x1200
Dx11 -- High
2x AA
8x AF
Tessellation enabled

 

[jaggerwild]

Some folks have posted DX10 tests and some have used lower filtering settings. I think we need to see more of the tests using exactly these settings or higher for a direct comparison:

1440x900/1920x1080/1920x1200
Dx11 -- High
2x AA
8x AF
Tessellation enabled

Cool I didn't know or didn't read the whole thread I'll post back...........
Here you go, DX11 Tessellation enabled, 2XAA, Anisotropy @8X, all the rest the same settings

 

[Apocalypse23]

Cool I didn't know or didn't read the whole thread I'll post back...........
Here you go, DX11 Tessellation enabled, 2XAA, Anisotropy @8X, all the rest the same settings

Nice score! Cool thanks!

 

[Apocalypse23]

Heaven Catalyst 10.3 a on 5850@925/1175

 

[Kindjal]

I have a similar system to you (though I have a Phenom II @ 3.6GHz). My Vapor-X 5870 is up to 930MHz on the core so far, I just keep pushing it 5-10MHz at a time and testing. I wonder if this test uses CPU power at all? You should score a bit higher with your CPU if it does. I'm gong to run the test at your settings and see how I do. I'll also run with my 5870 at your speed (875MHz) and see how my score compares, just to see how much the CPU matters in this test. I'll post later, have to run off to work. :/

Figuring that the Vapor-X card had a bit of headroom left, I ran the CCC sliders to 900MHz and reran the Heaven benchmark. Unfortunately, I had serious artifacting at that speed. I saw hundreds of sparkling squares lined up diagonally accross the monitor. I jumped out of the Heaven benchmark and ran Furmark and had the same problem. I dropped back to stock clocks and rebooted and everything was fine again.

I reran the Heaven with 2x and 8X AA results below:

 

[SlowSpyder]

Figuring that the Vapor-X card had a bit of headroom left, I ran the CCC sliders to 900MHz and reran the Heaven benchmark. Unfortunately, I had serious artifacting at that speed. I saw hundreds of sparkling squares lined up diagonally accross the monitor. I jumped out of the Heaven benchmark and ran Furmark and had the same problem. I dropped back to stock clocks and rebooted and everything was fine again.

I reran the Heaven with 2x and 8X AA results below:

Yea, I really seem to have lucked out with my Vapor-X. It seems a lot of them struggle to get past 900MHz since they barely have a bump in voltage over reference and cannot have the voltage tweaked. Reference 5870's can usually get to near 1GHz (or even above) with a bump in voltage, but because of non-reference VRM's the Vapor-X is stuck at it's factory voltage. I'll drop mine back down to 875MHz and test, that way we can see how much CPU matters, just out of curiosity.

*edit: With my PhII at 2.4/3.6GHz (NB and L3/Core Speed) and my Vapor-X 5870 at 875/1250 and all the setting identical to your 2xAA run I scored 36.7FPS / 924. So actually a bit higher than your i7, but that really just points out that CPU speed probably doesn't matter too much in this test. That's most likely with in the margin of error, and could be due to other differences in the system. Anyway, I just wanted to see how much CPU mattered, and it looks like as long as your CPU is up to task, having a faster CPU won't matter much for this test.

 

[Rubycon]

Anybody set the camera mode and walk around?
Use F3 to toggle tessellation ON and OFF. It sure does kill the frame rate (like cut in HALF!) but IMO it affects quality more than any other parameter on the board! Even running as fast as you can to kill everything that moves in a typical FPS game this is a feature that you can really notice in gameplay.

 

[Termie]

5850 @850/1200

1440x900, Dx 11, High settings/Tessalation, AA 2x AF 8x
FPS: 39.0
Score: 982

Note nearly 10% improvement with cat 10.3:

5850 @850/1200

1440x900, Dx 11, High settings/Tessalation, AA 2x AF 8x
FPS: 42.8
Score: 1079

1920x1200, Dx 11, High settings/Tessalation, AA 2x AF 8x
FPS: 32.3
Score: 813

This article confirms my finding of ~10% improvement with new drivers: http://www.legitreviews.com/article/1253/7/

 

[Kindjal]

Here are my results with the 10.3a driver - 5870@875/1250.

My Heaven score looks to have improved between 10-12%. (Previous score of 41.4FPS and a score of 1044)

I saw a 3% improvement in 3DMark Vantage and no change in 3DMark06.

 

[Bushwicktrini]

Unigine
Heaven Demo v1.0
FPS: 22.0
Scores: 554

Unigine
Heaven Demo v1.0
FPS: 29.2
Scores: 736

Hardware
Binary: Windows 32bit Visual C++ 1500 Release Oct 22 2009
Operating system: Windows 7 (build 7600) 64bit
CPU model: Intel(R) Core(TM) i7 CPU 920 @ 2.67GHz
CPU flags: 2672MHz MMX SSE SSE2 SSE3 SSSE3 SSE41 SSE42 HTT
GPU model: ATI Radeon HD 5800 Series 8.703.0.0 1024Mb

Settings
Render: direct3d11
Mode: 1680x1050 2xAA fullscreen
Shaders: high
Textures: high
Filter: trilinear
Anisotropy: 8x
Occlusion: enabled
Refraction: enabled
Volumetric: enabled

Everything is stock and the card is a 5850 using the cat 10.2 drivers