Jump to content

Unigine Heaven 4.0 Benchmark! Updated 14.02.2013 + BIG GPU test


  • Please log in to reply
125 replies to this topic

#41
mitch_de

mitch_de

    InsanelyMacaholic

  • Local Moderators
  • 2,884 posts
  • Gender:Male
  • Location:Stuttgart / Germany
Then the huge fps drop cames from drivers or not working AGPM (i prefer guessing).
Also it may happen that the drivers are OK but OpenGL benchmark has problems with it = fps lost.
Have you test Cinebench on DP2 vs 10.7.3 ? If also huge diff : drivers / AGPM if normal diff : Unique has problems.

#42
Graebags

Graebags

    InsanelyMac Geek

  • Members
  • PipPipPip
  • 129 posts
  • Gender:Male
  • Location:Canberra
@mitch_de
With Cinebench 11.5 on Lion 10.7.3 I get 29.09, and on DP2 30.51.
With Galaxies on Lion 10.7.3 I get 646 GFlops, and on DP2 621.
With Luxmark v2 on Lion 10.7.3 I get 958, and on DP2 1096, for the default render.

With OpenGL Extensions Viewer my scores for the basic Cube range from 6718 to 7309 with Lion, and 731 to 749 with ML DP2, ie a HUGE drop.

Is this a Fermi and later Nvidia gpu problem?

#43
mitch_de

mitch_de

    InsanelyMacaholic

  • Local Moderators
  • 2,884 posts
  • Gender:Male
  • Location:Stuttgart / Germany
OK - so we found out thats not the cpu clock and not the gpu clock - otherwise Cinebench would also be much slower in DP2.
So the DP2 drivers seems to be an problem for some OpenGL apps - but not for all. Some functions an OpenGL App uses / other not or the way it is using them gives problem or not.
It can be a bug in the drivers OR the problematic OpenGL Apps have some OpenGL code which isnt good for that OPenGL driver (Nvidia ) but not in general (, AMD seems not affected, also other Nvidia gputypes may not affected).

I also have NVIDIA (9600GT) and tested DP2:
For me NO FPS lost by DP2 vs DP1 - near same results as DP1 (which is also near 10.7.3).
So your FPS lost must be not an general problem with Nvida drivers but may come by your gpu type.
HINT: After clicking RUN (benchmark) you see SETTINGS menue : You could try to disable sound (its musc in the bench)- perhaps thats the problem?
Also check if you setup windowed (disable fullscreen!!) - perhaps somethings changed by VSYNC problems. In windowed mode vsync never matters. In fullscreen VSYNC does matter - even if OpenGL App has an setup for no vsyc . The drivers may NOT disable such in App disabled VSYNC if tehre is some bug in their vsyc driver part.
Also OpenGLExtensionsviewer works MUCH different in speed if VSYNC is working - to avoid vsync problems use window mode.

For me it doesnt matter sound on/off near same fps.
PIC shows usual 800x600, windowed, FSSA*2 , Shader = High, OS X ML DP2

Attached Files



#44
surfcaster

surfcaster

    InsanelyMac Sage

  • Members
  • PipPipPipPipPipPip
  • 447 posts

I think that you liked mich_de

I would like to see the punctuation of something like if run on OS X

#45
oSxFr33k

oSxFr33k

    InsanelyMac Legend

  • Members
  • PipPipPipPipPipPipPip
  • 826 posts
  • Gender:Male
  • Interests:Sound and Graphic Design. Electronics in general.
I deleted my post a few up since those screenshots where attachments and made screenshots this time. Very unhappy with my results. I had to delete AGPM because I was getting random freezing. 800 x 600 gives me the best scores!!

Attached Files



#46
mitch_de

mitch_de

    InsanelyMacaholic

  • Local Moderators
  • 2,884 posts
  • Gender:Male
  • Location:Stuttgart / Germany
I would not change so much settings like Shaders M/H or Ani beside resolution and FSAA - last parameters are the main FPS parameters-
Leave Shaders always at HIgh and Ani at 4* - otherwise we have even more different settings to compare. Different res + different FSAA is enough to bench ;)

Would also interesting_
Bench you results in non FSAA + windowed mode (800x600 or 1280x1024) vs same res but fullsreen.
Normally both sould run near same speed - otherwise vsync (only happens in fullscreen modes) limits FPS.
But normally vsync is switched off in such benchmarks - but may handled wrong by some gpu / driver combinations.

#47
^Andy^

^Andy^

    InsanelyMac Legend

  • Members
  • PipPipPipPipPipPipPip
  • 659 posts
  • Gender:Male

@Aivarz, These may make your day:

dp1 dp2


Edit:

My geekbench score was round 9500 for both Lion 10.7.3 (9483) and ML DP2 (9557).



I get similar results with my GTX570 DP1 v DP2 - Unigine scores in DP2 is roughly half the score i get in DP1. Cinebench and GLViewer reults are pretty much identical between DP1 and DP2 though (run GLViewer in full screen - for some reason it defaults to windowed on DP2?).

#48
mitch_de

mitch_de

    InsanelyMacaholic

  • Local Moderators
  • 2,884 posts
  • Gender:Male
  • Location:Stuttgart / Germany
Can you compare windowed 800x600 no FSAA against 800x600 no FSAA fullscreen for DP2 ?
(Or FSAA *2 if you have benched 800x600 in DP1 already to compare also DP1)
So we can see if vsync handling may be the fps down (in fullscreen mode).
If ysync is off normally windowed vs fullscreen (same 800x600!) should not give much differences.
Also compare GLViewer results fullscreen vs windowed (same res!). Also should give not much diff if vsync is off .

#49
oSxFr33k

oSxFr33k

    InsanelyMac Legend

  • Members
  • PipPipPipPipPipPipPip
  • 826 posts
  • Gender:Male
  • Interests:Sound and Graphic Design. Electronics in general.
I tried as you suggested and I saw no difference between windowed and full screen.

After many many benchmarks scenarios I realize where I start to suffer from FPS loss. In my case its screen resolution and not so much FSAA.

I am using 10.7.3 clean install and my graphics card is Mobile. I don't see any benchmarks other than mine on Mobile. It seems like most of the benchmark results posted so far are from Desktop Graphics Cards.

Even the GTX 460 I thought at first was mobile is why I was unhappy with my original results. I should be happy with these results I suppose unless I see someone post there mobile GTX460M with better results over my GTX560M. I doubt that will be the case though.

Attached Files



#50
^Andy^

^Andy^

    InsanelyMac Legend

  • Members
  • PipPipPipPipPipPipPip
  • 659 posts
  • Gender:Male

Can you compare windowed 800x600 no FSAA against 800x600 no FSAA fullscreen for DP2 ?
(Or FSAA *2 if you have benched 800x600 in DP1 already to compare also DP1)
So we can see if vsync handling may be the fps down (in fullscreen mode).
If ysync is off normally windowed vs fullscreen (same 800x600!) should not give much differences.
Also compare GLViewer results fullscreen vs windowed (same res!). Also should give not much diff if vsync is off .


I'll do this when i get some time but i know off hand that windowed glviewer results in dp2 are way way slower than fullscreen. I cant remember the exact figures but in Windowed mode the results are less than 1000 whilst in fullscreen they are over 5000.

#51
mitch_de

mitch_de

    InsanelyMacaholic

  • Local Moderators
  • 2,884 posts
  • Gender:Male
  • Location:Stuttgart / Germany
Yep, at least OpenGLViewer (Test/Benchmark) results are critical for VSYNC :
Its not DP only - also in 10.6 times , as i changed Monitor 120 Hz to TFT 60 Hz i got , as same resolution (fullscreen) + same cpu/gpu much lower FPS.
Its because OGLViewer will not switch off VSYNC and so the lower 60 Hz limits FPS against 120 Hz or (even better) windowed mode FPS.
As i said, only in windowed mode its garantied that VSYNC is OFF (because not used in windowed mode).

Thanks for testing Unigine fullscreen vs windowed mode = same FPS. SO we can be sure that Unigine didnt use VSYC.

EDIT: I tested OGLViewer 4.x (after long pause) again with 10.7.4 dev. Funny, now windowed mode (even a bit less res) is much slower than fullscreen.

And i seen that you can test 3.1, 3.2 OpenGL bench if you use CORE and not COMPATIBILITY (in the TEST Menue selectable).
Seems that default mode checks if all functions of 3.x / 4.0 are implemented / available by HW and if not (even only 1 of 10) isnt it disables the newer 3.1...4.2 Tests.
I now got running 3.1 and 3.2 Tests using CORE with my 9600 GT. Seems that for that tests the partly implemented 3.x features are enough to bench at least 3.1 + 3.2 OpenGL. :thumbsup_anim:

Attached Files



#52
oSxFr33k

oSxFr33k

    InsanelyMac Legend

  • Members
  • PipPipPipPipPipPipPip
  • 826 posts
  • Gender:Male
  • Interests:Sound and Graphic Design. Electronics in general.
@mitch_de,

And to compound the issue I noticed if I move the window around in OpenGLViewer my scores go way up from the 100's to several thousand, Try this see if you get the same result. Just move the window around immediately when the program starts and watch what happens.

#53
mitch_de

mitch_de

    InsanelyMacaholic

  • Local Moderators
  • 2,884 posts
  • Gender:Male
  • Location:Stuttgart / Germany
Does this, window moving rise FPS from hundrets to tousands also needed if you use Multisampling *8 ? (produce much more GPU load )
For sure the FPS are lower but maybe fast at max. possible without moving window.
If Multisampling *8 (OpenGLviewer TEST windowed) has NOT same effect : moving window rises FPS much (many times) than i guess the AGPM works but needs too much GPU load to rise the gpu MHz.
Also you can try to use KING insted of CUBE, also produce some more gpu load.
CUBE is very easy - perhaps to easy for fast gpus ;)

EDIT: If i move window my FPS LOWERS not rises, what should be normal.

#54
oSxFr33k

oSxFr33k

    InsanelyMac Legend

  • Members
  • PipPipPipPipPipPipPip
  • 826 posts
  • Gender:Male
  • Interests:Sound and Graphic Design. Electronics in general.

Does this, window moving rise FPS from hundrets to tousands also needed if you use Multisampling *8 ? (produce much more GPU load )
For sure the FPS are lower but maybe fast at max. possible without moving window.
If Multisampling *8 (OpenGLviewer TEST windowed) has NOT same effect : moving window rises FPS much (many times) than i guess the AGPM works but needs too much GPU load to rise the gpu MHz.
Also you can try to use KING insted of CUBE, also produce some more gpu load.
CUBE is very easy - perhaps to easy for fast gpus ;)

EDIT: If i move window my FPS LOWERS not rises, what should be normal.


Interesting it should rise I beleive. I went to a Mac Store and tried this on the latest Macbook Pro sandybridge with the latest AMD graphics card with the same results as my Graphics card. When I started it I was around 600-700FPS and moved it over a bit and all of a sudden went to 2500-3500FPS.

I have not changed any settings in OpenGLViewer. I have to try some of the settings you suggested to see when mooving the window if this actually lowers the FPS count with my system.

#55
mitch_de

mitch_de

    InsanelyMacaholic

  • Local Moderators
  • 2,884 posts
  • Gender:Male
  • Location:Stuttgart / Germany
I would temporary remove the AGPM (System/Library/Extensions/AppleGraphicsPowerManagement.kext).
Copy it to deskop, delete whole system cache /System/Library/Caches/com.apple.kext.caches (will be rebuild automatic)
Reboot and try again if that 600 > 3000 also happens or not. If also, its not an AGPM problem.
You can use kextwizard to easy install AppleGraphicsPowerManagement.kext again + reboot.

#56
oSxFr33k

oSxFr33k

    InsanelyMac Legend

  • Members
  • PipPipPipPipPipPipPip
  • 826 posts
  • Gender:Male
  • Interests:Sound and Graphic Design. Electronics in general.

I would temporary remove the AGPM (System/Library/Extensions/AppleGraphicsPowerManagement.kext).
Copy it to deskop, delete whole system cache /System/Library/Caches/com.apple.kext.caches (will be rebuild automatic)
Reboot and try again if that 600 > 3000 also happens or not. If also, its not an AGPM problem.
You can use kextwizard to easy install AppleGraphicsPowerManagement.kext again + reboot.


Just got home and tried without AGPM no difference at all. I don't think you have to actually only move the window but clicking in the window increases the FPS count as well.

WIth King around 2000 FPS with CUBE around 2500 FPS. Windowed or FUll screen minimal difference. Changing Multisampling from default 2 to 8 no difference maybe a couple hundred FPS. If I do not move the window or click in the window the FPS average around 600 FPS.

Must be a bug in the OpenGLViewer Software maybe?

If I change Compatibility mode to CORE the program just stops after the first couple tests. I tried the default SGI and Nvidia no difference in scores

#57
mitch_de

mitch_de

    InsanelyMacaholic

  • Local Moderators
  • 2,884 posts
  • Gender:Male
  • Location:Stuttgart / Germany
Yep, OpenGLViewer may have an bug handling some tests and maybe its only for some driver / gpu combinations. Dont worry, simple remember that problem and dont try this TEST with your GPU if other tests are more valide.
The CORE Test Mode , which tries to use uncomplete (Drivers for gpu type hasnt 100% impelemnted of that) OpenGL 3.1 features may depend also on gpu type / drivers.
For me CORE (insted of default compatibility) with NV 9600 GT runs at least OpenGL 3.1 and 3.2 without errors. Result around 410 FPS KING.
PS: OGLV is Version 4.0.5, maybe you check for update. For me same results with .05 as with .02.

Screenshoot shows CORE Test OpenGL 3.2
benching 3.1+3.2 then it stops , because in CORE Mode OpenGL 3.3+ havent all functions implemented for my GPU (less than 100%)

Attached Files



#58
dMopp

dMopp

    InsanelyMac Protégé

  • Members
  • Pip
  • 1 posts
Hmm, my 470 GTX has only OpenGL 2.1 Compatibility ? WTF ?

#59
nyolc8

nyolc8

    InsanelyMac Legend

  • Members
  • PipPipPipPipPipPipPip
  • 802 posts
  • Gender:Male
  • Location:Budapest, Hungary
Sapphire Radeon HD 4830 512Mb > 800x600 | 2XAA | window | Shader:High | Ani:4X > 1237
Sapphire Radeon HD 4830 512Mb > 1920x1080 | 2XAA | fullscreen | Shader:High | Ani:4X > 464

Attached File  1.jpg   76.26KB   18 downloadsAttached File  2.jpg   79KB   21 downloads

Not the best, but it does the job for me :) (for now)

#60
mitch_de

mitch_de

    InsanelyMacaholic

  • Local Moderators
  • 2,884 posts
  • Gender:Male
  • Location:Stuttgart / Germany
Yep - and for gaming (not benching) in that high res (1900x1080) i think you would get near playable min FPS if you would use NOFSAA with that engine.





0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users

© 2014 InsanelyMac  |   News  |   Forum  |   Downloads  |   OSx86 Wiki  |   Mac Netbook  |   PHP hosting by CatN  |   Designed by Ed Gain  |   Logo by irfan  |   Privacy Policy