Share your 4850/4870 Benchmarks w/ 8.7 drivers

rafer

Distinguished
Jul 15, 2008
38
0
18,530
For anyone posting benchmarks, if you'd be so kind as to state which driver you were using previously and what benchmarks you got with that (assuming you know that info), in addition to the benchmark with the new official 8.7 drivers.
 

SpinachEater

Distinguished
Oct 10, 2007
1,769
0
19,810
Oh yeah, forgot about them coming out. Probably just 3DMark score increases and xfire fixes...I don't have any expectations for the singe GPU. This will give me something to do with my insomnia tonight :)

P.S. Guru3D used RivaTuner 2.09 in their 4870x2 review....does it support the R700 family now? I missed that announcement if it does...the last I heard it wasn't supporting them. Plus it doesn't say that it supports them...huh?



Performance Improvements

The following performance gains are noticed with this release of Catalyst™.

* 3DMark Vantage: Performance increases of up to 20% is noticed in the Performance Preset on single card configured systems when using ATI Radeon™ HD 36xx or ATI Radeon™ HD 34xx products
* Company of Heroes DX10: Performance increases of 3% to 12% is noticed in specific maps on single card configured systems when using either an ATI Radeon™ HD 48xx, ATI Radeon ™ HD 38xx, or an ATI Radeon™ HD 36xx product
* Lost Planet DX10: Performance increases of 4% to 15% is noticed in specific maps on single card configured systems when using either an ATI Radeon™ HD 48xx, ATI Radeon™ HD 38xx, or an ATI Radeon™ HD 36xx product
* Call Of Duty 4 DX9: Performance increases of up to 4% in specific maps is noticed on single card configured systems when using an ATI Radeon™ HD 48xx product
* Lost Planet DX9: CrossFire™ scaling improves up to 1.7x and performance increases up to 80% across all Radeon™ products when AA and AF is enabled



Well, off the bat my temps are lower.
40cwe9.png


Huh...I didn't believe this at first but the air coming out of the back isn't even hot. I can't find my dig thermometer to measure the exhaust but it is hardly warm.
40loadyt5.png
 

ovaltineplease

Distinguished
May 9, 2008
1,198
0
19,280
Did you mod the fan speed or is that by default in the driver now?

I had to boost the fan speed on the 4850 with the 8.6 driver in order to get the temps down - my other PC isn't as cool running as my main one though.

Haven't tried 8.7 yet
 
Wow you are running a 4GHz memory clock. Crazy. I would like to see how far the GPU will go OCing on air. The highest I ever got my 2900Pro 1GB was 815MHz. But I decided that was not a very good idea to run consitently.
 

SpinachEater

Distinguished
Oct 10, 2007
1,769
0
19,810
Yeah I did mod the fanspeed. I like setting it to a select speed rather than have it bump up to 80% then drop to 5%. I didn't check though to see if the fan management was improved on the default profile.
 

SpinachEater

Distinguished
Oct 10, 2007
1,769
0
19,810
I wasn't even paying attention to the memory clocks in GPUz. That is not correct. I slipped in the updated version 2.6 and it corrected it. This is with the AMDtool OC hack + fan fix. Let the card frying begin!

ocamdqs2.png
 

AdioKIP

Distinguished
Jul 10, 2008
789
0
19,160
Running 2 4850's in Crossfire. Ran the Performance test at 1680x1050. With 3DMark Vantage I get a GPU Score of 7122 and a CPU Score of 4425. Previoulsly my GPU score was in the area of 4500 but I dont beleive i had things configured right when I originally ran it, I dont think all of the boost was from the drivers as much as me tweaking a few things.

Asus M3a32 Deluxe
Athlon 64 X2 6000+ (No Overclocking)
2x Radeon 4850'2 (3 Monitors, Crossfire for gaming)
4 Gigs DDR 2 800
Vista Ultimate X64
 

ovaltineplease

Distinguished
May 9, 2008
1,198
0
19,280
you just make a profile and then adjust the fan speed in the profilename.xml

set fan control to manual and speed to whatever

Generally, that will bring the temps of the gpu die down a lot
 

SpinachEater

Distinguished
Oct 10, 2007
1,769
0
19,810
I put OCZ freeze on when I first got the card and am running at 50% fan speed. Having a constant fan speed helps a lot. The auto setting cycle between 5% and 80% speeds...it is really loud and annoying when it vents the hot air at 80%. I can deal with a constant 50% when I am gaming though.
 

ovaltineplease

Distinguished
May 9, 2008
1,198
0
19,280
Hm..

If you have them...

Grid with CSAA edge detect, Crysis with any AA setting - if you don't have Crysis then maybe CoD4 CSAA ED

Those are the ones i'm most curious about I think.
 

SpinachEater

Distinguished
Oct 10, 2007
1,769
0
19,810
Ok, here is a quick preview for the Crysis bench....need some sleep so I will cut it short for now. I can't get it in x64 for some reason and it says demo-benchmark....whatever.

Beginning Run #1 on Map-island, Demo-benchmark_gpu
DX9 1680x1050, AA=No AA, Vsync=Disabled, 32 bit test, Windowed
Demo Loops=1, Time Of Day= 7
Global Game Quality: High
==============================================================
TimeDemo Play Started , (Total Frames: 2000, Recorded Time: 111.86s)
!TimeDemo Run 0 Finished.
Play Time: 47.32s,
Average FPS: 42.26
Min FPS: 32.86 at frame 1300,
Max FPS: 53.50 at frame 1011

Average Tri/Sec: -37931272, Tri/Frame: -897514
Recorded/Played Tris ratio: -1.02
TimeDemo Play Ended, (1 Runs Performed)



Beginning Run #1 on Map-island, Demo-benchmark_gpu
DX9 1680x1050, AA=2x, Vsync=Disabled, 32 bit test, Windowed
Demo Loops=1, Time Of Day= 7
Global Game Quality: High
==============================================================
TimeDemo Play Started , (Total Frames: 2000, Recorded Time: 111.86s)
!TimeDemo Run 0 Finished.
Play Time: 56.90s,
Average FPS: 35.15
Min FPS: 24.41 at frame 146,
Max FPS: 43.93 at frame 999

Average Tri/Sec: -30706284, Tri/Frame: -873669
Recorded/Played Tris ratio: -1.05
TimeDemo Play Ended, (1 Runs Performed)

 
G

Guest

Guest
spinach do you have a 4850 or 4870?

I searched your posts and your profile and couldn't find it...
 

AdioKIP

Distinguished
Jul 10, 2008
789
0
19,160
My Crysis Benchmark. Kinda dissapointing running in crossfire and this is all I get:


7/23/2008 8:17:34 PM - Vista 64
Beginning Run #1 on Map-island, Demo-benchmark_gpu
DX9 1680x1050, AA=2x, Vsync=Disabled, 64 bit test, FullScreen
Demo Loops=3, Time Of Day= 7
Global Game Quality: High
==============================================================
TimeDemo Play Started , (Total Frames: 2000, Recorded Time: 111.86s)
!TimeDemo Run 0 Finished.
Play Time: 70.23s, Average FPS: 28.48
Min FPS: 18.29 at frame 1930, Max FPS: 40.74 at frame 1569
Average Tri/Sec: -17759584, Tri/Frame: -623596
Recorded/Played Tris ratio: -1.47
!TimeDemo Run 1 Finished.
Play Time: 62.08s, Average FPS: 32.21
Min FPS: 16.48 at frame 1940, Max FPS: 47.11 at frame 91
Average Tri/Sec: -20033294, Tri/Frame: -621877
Recorded/Played Tris ratio: -1.47
!TimeDemo Run 2 Finished.
Play Time: 61.81s, Average FPS: 32.36
Min FPS: 16.48 at frame 1940, Max FPS: 48.79 at frame 72
Average Tri/Sec: -20115148, Tri/Frame: -621629
Recorded/Played Tris ratio: -1.47
TimeDemo Play Ended, (3 Runs Performed)
==============================================================

Completed All Tests

<><><><><><><><><><><><><>>--SUMMARY--<<><><><><><><><><><><><><>

7/23/2008 8:17:34 PM - Vista 64

Run #1- DX9 1680x1050 AA=2x, 64 bit test, Quality: High ~~ Overall Average FPS: 32.285
 

ovaltineplease

Distinguished
May 9, 2008
1,198
0
19,280



Would you say this was an improvement over 8.6 catalyst or just the same more or less?