Unigine Heaven 2.0 is out

[1h4x4s3x]

http://unigine.com/download/files/Unigine_Heaven-2.0.msi

http://translate.google.com/transla...rabhardware.net/forum/showthread.php?t=164761

Increase of up to 40%

 

[GodisanAtheist]

404 not found.

Edit: Still 1.0 on their D/L page.

 

[mhouck]

downloading currently. thanks for the link!

edit: slow as hell though. must be swamped

 

[Kenmitch]

Gotta find a mirror as it's estimating 65hrs left for me....Aborted for now

From a couple post's below

I would use Techpowerup to download it right now:
http://www.techpowerup.com/downloads/1775/mirrors.php

 

[Janooo]

http://translate.google.com/transla...rabhardware.net/forum/showthread.php?t=164761

AMD is up 40%? Nice gain.

 

[vj8usa]

You mean Heaven 2.0. The title had me confused for a second; Unigine's the company.

 

[Sylvanas]

The main thing we are interested in from the 'Heaven' bench is Tesselation performance, I hope this latest release finds it's way into some reviews come Fermi launch day.

 

[NYHoustonman]

Crashes about 30 seconds in for me (HD 5870)

EDIT: Seems to only happen when I enable the wireframe view...

 

[RussianSensation]

The official website has Version 2.0.
http://unigine.com/download/

I would use Techpowerup to download it right now:
http://www.techpowerup.com/downloads/1775/mirrors.php

 

[blanketyblank]

Not bad.
30.9 fps average on a 5770 with 4x antisotropic and normal tesselation at 1360 x 768 with catalyst 10.2. Minimum was like 16 and max was 90 something.
If this were a game I'd consider that borderline playable unlike the 24 or so I got before.

 

[spinejam]

 

[Dark4ng3l]

So they release this the week we are supposed to see Fermi. That and Heaven was the hand picked benchmark that their marketing drones picked to show us first. Let me speculate that we will see Fermi do real good in this benchmark for some reason....

 

[HeXen]

so whats different with this version?

 

[dguy6789]

Q9550 @ 3.4Ghz
4870 512MB
Catalyst 10.3a
1680x1050
DX10
Default settings
Min: 24.4
Avg: 41.7
Max: 80.5

 

[happy medium]

Not bad.
30.9 fps average on a 5770 with 4x antisotropic and normal tesselation at 1360 x 768 with catalyst 10.2. Minimum was like 16 and max was 90 something.
If this were a game I'd consider that borderline playable unlike the 24 or so I got before.

30.2 avg. with a 5750 @ 890,1325 with 4x and normal tess. @ 1280x1024 with cat 10.2. Min 6.5, max 81.

Thats odd? My minimums sucked. My resolution was higher.
Mabe your 80 more sp's? Is your card overclocked at all?

 

[marmasatt]

I scored 179 with this (FPS of 6.9) with an i7 920 and 5850, lol. . Think I have a problem anywhere.

 

[Drakcol]

Overclocked 5850 800/1200
Resolution 1920 x 1080
Catalyst 10.3a

Yay for charts

Normal Stock
FPS: 29.2
Scores: 736
Min FPS: 11.7
Max FPS: 82.1

Normal Overclocked
FPS: 32.6
Scores: 821
Min FPS: 11.0
Max FPS: 92.2

Extreme Stock
FPS:17.7
Scores: 446
Min FPS: 6.3
Max FPS: 64.9

Extreme Overclocked
FPS: 19.7
Scores: 497
Min FPS: 6.4
Max FPS: 72.4

* Normal / Extreme refer to tessellation settings

 

[1h4x4s3x]

You mean Heaven 2.0. The title had me confused for a second; Unigine's the company.

Yes, fixed.

Nice chart Drakcol.

 

[mhouck]

so whats different with this version?

"Main changes are: added more content, improved engine, introduced "moderate" and "extreme" tessellation modes, added Linux version (32/64 bit)."

 

[blanketyblank]

30.2 avg. with a 5750 @ 890,1325 with 4x and normal tess. @ 1280x1024 with cat 10.2. Min 6.5, max 81.

Thats odd? My minimums sucked. My resolution was higher.
Mabe your 80 more sp's? Is your card overclocked at all?

Nope, my card was at 850/1200 for that run. I just ran it again to see how precise the test is, and got the same score though minimum was 16.2 that time.
Are you setting AA? 4x was my antisotropic not AA.

Also try running your test without overclocking and check your minimums. Maybe the OC isn't quite as stable as you thought.

Strangely enough just a small overclock to 900/1200 (just 50 Mhz clock) and my minimum is at 10.2 though my average went to 31.5 at the same settings. Maybe this test is very good at bring about an error correction for overclocking?

 

[spinejam]

How long do you let it run through to get your scores? Is it like 3dmark where it goes until it stops? I stopped it after a few laps through and then didn't see a score....

click "benchmark" in upper left screen.

 

[Dark4ng3l]

30.2 avg. with a 5750 @ 890,1325 with 4x and normal tess. @ 1280x1024 with cat 10.2. Min 6.5, max 81.

Thats odd? My minimums sucked. My resolution was higher.
Mabe your 80 more sp's? Is your card overclocked at all?

I get 29.0 fps(stock settings) and 33.6 @ 1015/1370 at those settings with a 5770.

 

[Sylvanas]

So they release this the week we are supposed to see Fermi. That and Heaven was the hand picked benchmark that their marketing drones picked to show us first. Let me speculate that we will see Fermi do real good in this benchmark for some reason....

It's my understanding that Unigine is a 3rd party with no affiliation to Nvidia or AMD. It's no worse than 3Dmark is, so I don't see the problem. I'm sure nvidia probably will show us Heaven 2.0 benches at release because due to the parallel (geometry) nature of Fermi I expect Fermi to do better than a 5870 with that extra 1 billion transistors at it's disposal. However we already have seen slides from Nvidia with this bench and they only took 'the most intensive' 'parts' of the bench and then demonstrated what a lead it had over the 5870 (it was on techpowerups news page, too lazy to find it). However when given the entirety of the they both performed pretty much equal (I think the full bench is in techreports GF100 preview, may be mistaken).

 

[RussianSensation]

Q9550 @ 3.4Ghz
4870 512MB
Catalyst 10.3a
1680x1050
DX10
Default settings
Min: 24.4
Avg: 41.7
Max: 80.5

My 4890 with 880 GPU clocks and 1GB is barely faster. :thumbsdown: I wonder if cranking 4AA/16AF has any impact on that 512mb of ram.

Catalyst 10.2
1680x1050
DX10
Default settings
Min: 28.0
Avg: 48.1
Max: 94.7

 

[mhouck]

spinejam: the top benchmark you posted shows you running in dx11 w/ tessalation off. From your signature, I thought you were running a 4870x2. I'm confused as to how you were able to run it with that setting.