|02-08-06, 07:58 AM||#1|
Join Date: Mar 2003
64-bit benchmark variations
I was having a hard time finding some 64 bit benchmarks. I am posting some findings.
While there have been many discussions on the benefit of 64-bit, it would be presumptious to conclude that things automatically run better. In fact, I've had a difficult time getting some games to run at all. This is usually do to an issue with the installer and less to do with compatibility. While I have observed apps and games do run faster, I do not feel we are anywhere near realizing the potential of the 64 bit variations or the associated hardware.
I am limited to 64 bit information as well as Windows 2003 Enterprise Server 64-bit.
82.12 default clocks cards and 2 dual core opties 285 cores
On Doom 3 -timedemo demo1 usecache - I get 136 FPS with Ultra Quality and 2x AA at 2560x1600.
On CF 64-bit - Video Stress test - I get 171 FPS with Reflect all, high, high, high, 4xAA, 8xAF, at 2560x1600.
On Lost Coast 64-bit HDR - Video Stress test - I get 122 FPS with Reflect all, high, high, high, 4xAA, 8xAF, at 2560x1600.
On PainKiller - benchmark C5L1 - I get 229 FPS Maxed settings at 2560x1600
On PainKiller - benchmark C5L2 - I get 199 FPS Maxed settings at 2560x1600
My 3dMark 06 composite score is 10488, but professes that the driver is not approved.
82.12 with clocks at 600/1800 cards and default clocks for 2 dual core opties 285 cores
On Doom 3 -timedemo demo1 usecache - I get 148 FPS with Ultra Quality and 2x AA at 2560x1600.
On CF 64 bit - Video Stress test - I get 180 FPS with Reflect all, high, high, high, 4xAA, 8xAF, at 2560x1600.
On Lost Coast 64-bit HDR - Video Stress test - I get 135 FPS with Reflect all, high, high, high, 4xAA, 8xAF, at 2560x1600.
On PainKiller - benchmark C5L1 - I get 241 FPS Maxed settings at 2560x1600
On PainKiller - benchmark C5L2 - I get 206 FPS Maxed settings at 2560x1600
My 3dMark 06 composite score is 10720, but professes that the driver is not approved.
Except for 3dmark 06, all benchies were done at 2560x1600
7800gtx 512 clocks were modified at the vga bios, not using coolbits.
Same identical bios image was loaded on both cards.
Win Ent Server 2003 64-bit was optimized for workstation and gaming.
I recently moved from Opt 280 to Opt 285 (dev) and have seen some slight increases in performance, but
most of the performance gains come from overclocking the two 7800gtx 512's
Yes, I can and have overclocked the dual opties and have gotten significant increases in the scores, however I am such a huge fan of stability that I don't run it in that mode often, so these scores are with the 4 cores at stock speeds. I did break the 11K mark on 3dmark 06 when I did clock up the cores, but that result is not part of my normal configuration. In addition I don't believe 3dmark 06 takes advantage of more than two cores and that optimization seems slight at best.
Hope this helps those seeking 64 bit benchies on 82.12
Dual Opteron Dual Core 2222s (OC 3.17GHz)
ASUS KFN32-D SLI SAS (OC 1.1GHz HPT)
2 evga Superclocked 8800 GTX SLI (650-1550-2080 OC)
16GB Ram DDR-2 ecc reg
8 terabytes (8 x 1T drives)
ex8350 raid controller card
Dual Tagan 1100Watt Power Supplies (2200W)
Sony Blue Ray DVD Burner (Vrs 3)
Apple 30" Cinema LCD at 2560x1600 (version 2)
Windows Enterprise Server 2003 64-bit
Suse 10 64-bit, Vista 64-bit Ultimate
Stacker CM server case
Deck Legend Keyboard
...& plenty of blue lights.
Last edited by Venturi; 02-08-06 at 08:46 AM.
|Thread||Thread Starter||Forum||Replies||Last Post|
|Understanding the Bulldozer Architecture through the LINPACK Benchmark||News||Latest Tech And Game Headlines||0||06-26-12 11:30 AM|
|Passion Leads Army: DX 11 and GPU PhysX Benchmark||News||Latest Tech And Game Headlines||0||05-28-12 08:00 PM|
|Anyone benchmark KT266, KT266A, KT333 and KT400?||Needa916||CPUs, Motherboards And Memory||7||10-09-02 08:34 AM|
|nForce2 "First Look" and fly-by benchmark||GamblerFEXonlin||Feedback Forum||4||10-01-02 07:48 PM|
|3d benchmark se||cricket||Benchmarking And Overclocking||3||08-30-02 11:02 PM|