Printable View
Source: http://forum.beyond3d.com/showpost.p...postcount=4535
http://www.megaupload.com/?d=DXLT34HS
http://rapidshare.com/files/36724306...-RC1.part1.rar
http://rapidshare.com/files/36724679...-RC1.part2.rar
Super bloated Win7 install, Cat 10.1, Stock HD5870, i7@3.6
1920x1200, everything enabled, Tessellation at 1.0, 4xAF, no AA.
Min: 6.6
Max: 92.3
AVG: 32 FPS
It's about bloody time, that thing needed patching. Your forum link is dead, however.
This is not a patch, it's a new benchmark. There is a new zeppelin and stuff!
oRly. This would be the new version shown at GDC then?
why didnt u use cat 10.3a?
because I'm happy with my card's performance and I don't need anything else :D
I don't overclock the card even 1mhz...
Is there any change list of some sort available?
The zeplin is pretty cool I must admit. Did the older versions have an extreme option for tessellation or is that new to version 2? Anyways here is a default run through at 1920x1200. Details in signature.
http://img46.imageshack.us/img46/375/43659379.th.jpg
I'm going to run one with extreme tessellation and 4xAA 16AF to see how badly it rapes my machine :rofl:
so this is a pure DX11 bench ? Or can it do DX10 mode or something.
The new version adds new tessellation options and more DX11 features such as enhanced DoF and ambient occlusion. Suposedly, this is more representative of what the engine is capable of when used in upcoming games (Afterfall and Primal Carnage).
gtx295 default 1920x1200 DX10 4xAF 0AA
FPS:65.8
Scores:1658
Min FPS:23.1
Max FPS:137.9
197.13 forceware.
Holy crap... with extreme tessellation and 4x AA 16AF this thing destroys current hardware :eek:
http://img138.imageshack.us/img138/1638/72650842.th.jpg
Brutal!
One thing I wish they'd changed is the cobblestones. Every time I look at them my ankles cringe :p:
woow that bad i'll try with the 5850 quad cf but my friend does not let me install such things on the crunching computer...Quote:
Originally Posted by Chickenfeed
Anyone with a 5970 pls try this with all extrema settings like Chickenfeed did
Basically each setting performs roughly 2x worse than the last ( DX10 default 2x ---> DX11 default 2x ---> DX11 extreme with AA/AF ) Obviously it is just a benchmark showcasing new technologies ( no different than Futuremarks stuff ) as it isn't realistic for a game to have the same rendering cost. It still should provide some insight though.
What replication does?
BTW, a quick run
HD 5970 stock / Cat 10.3a
http://i40.tinypic.com/bwoxv.jpg
It sucks that Min FPS doesn't really increase from 5870 to 5970
my 5770 @ 935Mhz and i7 920 @ 4GHz gets 370 with everything on max 1920x1200 except AA and AF turned completely off for anyone wondering.
More importantly Furmark just got updated too: Furmark 1.8
As I mentioned, the Unigine engine will be used in several games. So, it is much more than "just" a benchmark.Quote:
Originally Posted by Chickenfeed
Tesselation works on 4850 with opengl mode :D
So it's using 4-series fixed-mode tesselator
Thats not with Tesslation ExtremeQuote:
Originally Posted by japamd
Quick comparison: Unigine Heaven 1.0 vs 2.0 RC1
1.0 - 33 FPS
2.0 RC1 - 51 FPS
I didn't know Tesselation was THAT awesome :eek:
You're right, it's not with extreme setting.Quote:
Originally Posted by Gorguls
If min FPS with tesselation doesnt improve, then I cant see any point getting better cards than 5770s except for Eyefinity setups.
That is very weird, that even a 5970 with extreme tesselation gets knocked down to 9 FPS minimum.
Looks like Fermi will actually have a huge advantage with tesselation enabled.
Thanks for this thread annihilat0r :up:
Yes I fully realize this. The fact is a games rendering cost won't have the resource budget of a benchmark as the benchmark load isn't ment to be realistic ( ie it is not indictive of the engines use in a game using the games resources ) That said, expect to see actual games using the engine to run better.Quote:
Originally Posted by SKYMTL
I'd be most interested seeing how GF100 does at the extreme setting with AA. Given the Nvidia internal benchs were with 1.1 and no AA with the less demanding tessellation, who knows.
Extreme Tesselation - 1920x1200 - NoAA - AF 16x - HD5970 @ 925/5000 - Cat 10.3a
http://i43.tinypic.com/wod41.jpg
Replication (aka _gs at end of bat file name) gives 10% Score hit.
wait till you try 8xAA
http://i42.tinypic.com/iqg8xh.jpg
Anybody else find it kind of fishy that it releases right before FERMI?
no.Quote:
Originally Posted by 96redformula
http://translate.google.com/translat...p%3Ft%3D164761
Can anyone check that? Apparently the dragon scene from 1.0 -> 2.0 improved by 40% for the 5800's...
This is also where the Fermi cards had their biggest lead in the benchmark
I noticed this myself... Now the question is will this gain be seen on GF100 hardware as well... if not someone did a no no :confused: He is claiming that Nvidia used their own version of the benchmark for internal testing purposes which would be a stretch if you ask me but if true... tisk tisk.
There are no such things as no no's in PR... :rofl:Quote:
Originally Posted by Chickenfeed
I am on a 46" Samsung LCD, so no 1920x1200 for me, damn you guys :shakes:
A bit...Quote:
Originally Posted by 96redformula
Quote:
Originally Posted by zerazax
I've posted about it in previous page, got ~50% improvement at dragon scene.Quote:
Originally Posted by WeiT.235
Quick comparison: Unigine Heaven 1.0 vs 2.0 RC1
1.0 - 33 FPS
2.0 RC1 - 51 FPS
Default Settings
1920x1080 Resolution
HD5850 785core /1275 memory
Athlon 555 Unlocked 4 cores 3.2GHZ @ 1.25V
http://img401.imageshack.us/img401/926/58154425.jpg
So it looks like 1gb is crapping out again.
I doubt it. Even at lower resolutions ( less than 1680x1050 ) min fps still is low with 4x AA with the extreme setting. Add the fact I havent seen any stuttering that would indicate otherwise.
HD5970 850/1200 everything at maximum (Extreme tessellation, AA8x, AF16x and 1920x1200):
20.3fps / 510 points. These are the most impressive graphics i've ever seen. Bye Crysis, you're second.
Given the similar single digit min FPS using 5850, 5870 and 5970 that's quite possible. Please, somebody post the Vmem usage graphic.Quote:
Originally Posted by Iconyu
except crysis is actually a game, so it's still first.Quote:
Originally Posted by NaMcO
Sure, it can.Quote:
Originally Posted by Zloyd
Extreme tessellation mode doesn't look much better from what I can tell, it's only good for killing FPS...Quote:
Originally Posted by Chickenfeed
It should, maybe that's why it's RC1...Quote:
Originally Posted by annihilat0r
Off topic, but why does it say "Windows XP / Vista 32-bit"?Quote:
Originally Posted by initialised
Well, to be honest, it looks good, but not really mind blowing, not that we couldn't achieve similar stuff without it...Quote:
Originally Posted by FischOderAal
What I don't understand is why FPS goes down so much during zoomed in scenes... There's barely anything to render yet the card is starting to choke. Not that the stuff looks so amazing when you're too close either...
:rofl: Points of view, they're still CG Graphics :p:Quote:
Originally Posted by Helloworld_98
CG? Hey, I thought there were actually people and palm trees in my monitor. :p: I also thought my chute is gone, MY DAMN CHUTE IS GONE!!Quote:
Originally Posted by NaMcO
I can't download it from that link.
link down.
I would love to upload it to another site, but I have 385k upload speeds or something like that, would take me too long. Sorry :)
I added a Rapidshare link.
Direct Link Final Version.
http://unigine.com/download/files/Un...Heaven-2.0.msi
Tessellation is leveraged to dynamically adjust the detail level based on the scene. Scenes where less detail can be seen ( eg from a far ) have their triangle count reduced thus the performance increase were as scenes with high detail ( close up ) increase their triangle count thus the performance reduction. Its quite simple really. If the wide shots were rendered at the same quality as the close shots, it would be a power point presentation.Quote:
Originally Posted by zalbard
@Gorguls
Nice :)
Thank you! :up:Quote:
Originally Posted by Gorguls
I know.Quote:
Originally Posted by Chickenfeed
The thing is, even with reduced level of detail scenes taken from far have 2-3x more FPS yet look much better and more detailed than close up scenes the engine seems to stutter at...
Maybe it's just me... :shrug:
Whoops, seems like when media center is open Powerplay doesn't release the clocks back to 3d speeds.
Here is the corrected run with it at full clocks 785/1275 :)
HD5850 785/1275
Athlon 555 Unlocked 4 cores 3.2ghz @ 1.25V
Benchmark Settings:
Default settings
1920x1080 Resolution
http://img401.imageshack.us/img401/926/58154425.jpg
ThanksQuote:
Originally Posted by Gorguls
Average FPS: 34.9
Score: 880
Min FPS: 5.8
Max FPS: 127.8
Settings
1920x1080
noAA
MaxAF
Extreme Tess.
No need to check i know that guy :DQuote:
Originally Posted by zerazax
Yep, that's right. There's definitely improvement for my 5870 with the dragon scene.
This is with my Crossfire 5770s @ 1000 / 1445, no AA and 4x AF:
http://img28.imageshack.us/img28/3820/heavenextreme.png
The Tessalation Extreme setting isn't really a good real world comparable setting. On Extreme, tessalation is way over the top, slowing down the cards, while not giving any noticeable extra eyecandy.Quote:
Originally Posted by Gorguls
Its a good power comparison between different cards though.
Power comparison for what? What it's most likely going to show, is that Fermi will run all over 5870, yet with real world usage, you don't need such excessive amounts of tessellation. Why benchmark something you aren't going to use in real world applications and games? Just for higher numbers? It looks like Fermi is only going to be a tad faster than 5870, while extreme tessellation might give an extremely skewed view.Quote:
Originally Posted by bhavv
Yup, thats what benchmarks are for, better numbers.
But how can you be so sure how much tesselation games are going to use? Someone might decide to use a crazy amount of it.
Link got pulled.....
Why use a crazy amount of it if it's not going to make a difference in picture quality? Just to bring the FPS down? :shrug: Just run the benchmark with the different tessellation options, and try to see the difference. It's extremely hard to spot the differences, so it isn't going to look better if you have to actually search and compare screen shots, just to try and find differences :shrug:Quote:
Originally Posted by bhavv
Tessellation is not only there to make games look better. It's also there to make games run better. Instead of using multiple fixed models, swapping them as you move away/towards those models for either higher/lower detailed models, tessellation can actually increase polygon counts on the fly. This saves space, as you will only need 1 model per item you want to render, avoids seeing models pop into more/less detailed models as you move and can be easily scaled down and up to improve performance or to improve visuals. Enough polygons is enough, though. If you can't see the difference anymore, because you are already using enough polygons for all the details, using more won't actually net you a better looking model. It's just going to slow the game down. That's why I think using and benchmarking an extreme amount of tessellation isn't going to be really needed, and will only skew the results, as Nvidia's and ATI's architecture both handle it differently.
Nvidia uses it's shaders to produce tessellation. This means it can allocate more resources towards tessellation as needed. ATI uses a fixed function tessellator, however. This mean as you increase the amounts of tessellation, the tessellator will be bottlenecking the rest of the architecture. That's why it's unfair to benchmark using excessive amounts of tessellation, as it's obviously going to make nVidia look better than ATI, while it's not going to be any benefit to real world games.
ATI's tessellator does look to a bit too slow, though. I guess we'll find out if it's good enough once people start benchmarking heaven with the normal tessellation mode, metro 2033 and grid. I'm really looking forward to comparisons between Fermi and 5870! :up:
First run, CPU @4,5, 5970 and 2x 5870's @stock.
http://img442.imageshack.us/img442/8978/heaven2.jpg
Could you possibly have a faster rig? :eek:
Nice score, is the 5970 clocked to 5870 levels? Should improve performance by quite a bit I'd say.Quote:
Originally Posted by Mydog
I usually run the 5970 @1000/1220 and the 2 5870 @1050/1330 with th 10.3 preview driver, but for some I just got "display drivers stopped..". Took the OC down to 5970 @950/1200 an 2x 5870 @1000/1300 and got this.Quote:
Originally Posted by Musho
http://img32.imageshack.us/img32/3637/heaven21.jpg
Wow, more than DOUBLE the min FPS by overclocking! That's very good indeed!Quote:
Originally Posted by Mydog
Great point, so why are we all using the extreme setting again????Quote:
Originally Posted by Musho
He CAN play Crysis :shocked: /jkQuote:
Originally Posted by annihilat0r
Impressive scores Mydog!
And I can play Metro 2033:yepp:Quote:
Originally Posted by Lightman
Thanks Lightman
Haha, yeah dude only the best aye? Respect. :toast:
Ah but if they had flat, able-to-walk-on cobblestones then that wouldn't show the power of tessellation would it :pQuote:
Originally Posted by SoulsCollective
Theres a clue in the name of this website - www.xtremesystems.org :pQuote:
Originally Posted by 96redformula
Games dont matter here when you can get higher points in a benchmark with an extreme setting ^^
Just like no one is going to be gaming on a LN2 cooled CPU at 5.5 Ghz, but they can do it just because its Xtreme.
I just knew somebody was going to say that, so I set you up for that one :ROTF:Quote:
Originally Posted by bhavv
Any way to tell what mode is being used on specific cards or are we relying on what the engine programmer is telling us?Quote:
Originally Posted by netkas
Haven't tested this myself, but the difference between no tessellation and tessellation enabled is huge. If they are reporting they are seeing tessellation, they probably are.Quote:
Originally Posted by STEvil
here, tesselation ON ( EXTREME )
http://i44.tinypic.com/23icbwm.jpg
Let's count the polygons. I'm starting
Unigine Heaven 2.0 FINAL
Tesselation Normal vs Extreme
1920x1200 - AA 0x - AF 1x
http://www.framebuffer.com.br/sites/...vs_extreme.jpg
Gaul, now the same image but with normal tessalation please! Thanks a lot guys for sharing those numbers. Hell, if I buy a 5970 and get those min FPS, for sure I will ask my granny to make me a chocalate cake to end with the frustration, too much money for ridiculous slow minimal fps.
Its kind of insane, so much difference between minimal and max fps, also with extreme and normal.
@japamd: So this means that the min FPS numbers were indeed momentary drops probably due to HDD usage, and are not to be taken seriously.
Exactly why I am now even happier with a pair of super clocking V1 5770s :).Quote:
Originally Posted by Pekalion
I were actually thinking of trying to sell them at an inflated price and getting 5850s, but the price difference seriously isnt worth the performance difference in DX11, and DX9 / 10 games are fast enough at 1920x1200.
How do you make that FPS graph btw? I want to make one for comparison.
well AS u WISH.....:up:Quote:
Originally Posted by Pekalion
TESSELATION OFF
http://i40.tinypic.com/2wnsoz4.jpg
TESSELATION ON ( extreme )
http://i39.tinypic.com/358cuur.jpg
That's terrific news for 5870 owners with the fps bump (40% wow). Looks like the card which was initially a beast is now even more of a beast as time goes on..
Quote:
Originally Posted by Gaul
on this shoot, can fermi touch 30 FPS ?? :p: 5870 def only 6 FPS
well see later :D
I think it was just an issue with RC version, with the final version I got 20.3/15.1 min fps in normal/extreme tesselation mode.Quote:
Originally Posted by annihilat0r
Tesselation Normal
http://i42.tinypic.com/33xhcaw.jpg
Tesselation Extreme
http://i44.tinypic.com/vxc677.jpg
Unigine
Heaven Benchmark v2.0
FPS:
21.7
Scores:
547
Min FPS:
3.2
Max FPS:
58.3
Hardware
Binary:
Windows 32bit Visual C++ 1500 Release Mar 7 2010
Operating system:
Windows 7 (build 7600) 64bit
CPU model:
Intel(R) Core(TM)2 Quad CPU @ 2.40GHz
CPU flags:
2400MHz MMX SSE SSE2 SSE3 SSSE3 HTT
GPU model:
NVIDIA GeForce 7900 GT/GTO 8.16.11.9107 256Mb
Settings
Render:
direct3d9
Mode:
640x400 windowed
Shaders: low
Textures:
high
Filter:
trilinear
Anisotropy: disabled
Occlusion:
enabled
Refraction:
enabled
Volumetric:
enabled
Replication: disabled
Tessellation: disabled
Unigine Corp. © 2005-2010
:rofl::rofl::rofl::rofl::rofl::rofl::rofl::rofl::r ofl::rofl::rofl::rofl:
Thanks Gaul! That is one of an epic shoot... Well, this weekend is the waiting over, we will see if Fermi can handle this Rev better than Ati at min fps, hope so or we are really furk up all of us, at least with this Dx11 gen card, maybe future gen can handle tessalation better.
If the GTX480 can get at least 25FPS at min FPS, we are good to go, if not, then, oh granny, chocolate cake please! Woha, what the hell are those mountains in my stomach? :O
I mean in reference to tesselation mode used: Fixed (4870/4890), other, DX11 (57/58/59 series)?Quote:
Originally Posted by Musho
Wireframe is killing the fps iirc.Quote:
Originally Posted by Gaul
Hopefully people are using the final version, since the RC seems to be causing min FPS to be in the 6-10 fps range, and final is higherQuote:
Originally Posted by Pekalion
Eh, even with the RC version this version seems a lot smoother than 1.0 Min FPS must be for a very short amount of the benchmark's running time.
http://img534.imageshack.us/img534/9458/00002d.jpg
it's tesselated(opengl/dx10.1 hw) :D
btw, can anyone with dx11 hw show tesselation settings (click on tesselation button on top) with moderate tesselation?
Just tried doing a run with 295 quad sli, and no support? Just ran on one core.
More here!Quote:
If you go into the heaven 2.0 directory, you will find heaven.zip. If you unpack that, you can see all the mesh files. If you have the free SDK, you will find optimized code that looks for specific hardware before defaulting to a secondary engine.
It doesn't name what it is looking for, but it looks for a specific hardware string, and when it is not found, it just falls back to the old render.