 |
|
11-04-2009, 07:35 AM
|
#1
|
|
Xtreme Addict
Join Date: Jul 2008
Location: Shimla , India
Posts: 1,129
|
First DirectCompute Benchmark Released
"...A forum member by the name of Pat has recently released a new benchmark tool for DirectCompute. This tool allows you to benchmark DirectX 11 latest general-purpose computing feature by calculating tons of FFT-like data and some memory transfers. DirectCompute is an application programming interface (API) that takes advantage of the massively parallel processing power of a modern graphics processing unit (GPU) to accelerate PC application performance. Be advised that DirectX 11 and the latest display drivers are required to run this benchmark...."
http://www.ngohq.com/news/16710-firs...-released.html
Downloading now to test with GTX275...
Try the benchmark and tell what score do you get, also mention the config of the system?
|
|
|
11-04-2009, 07:42 AM
|
#2
|
|
Xtreme Mentor
Join Date: Dec 2008
Location: Russia (for now)
Posts: 3,173
|
From the included readme file:
Quote:
Sample results:
GeForce 9800GT ECO OC (740/1800/1010) : M15670/S3600
GeForce 9800GT ECO OC (600/1500/900) : M13195/S3220
ATI Radeon HD5850 (900/1200) : M11799/S11841
ATI Radeon HD5870 (850/1200) : -/S10961
S-single threaded result
M-default dx11 device settings
|
I'm not sure this release is worth bothering with... 9800GT beating HD5850? Can't be right IMO.
|
|
|
11-04-2009, 08:12 AM
|
#3
|
|
Xtreme Member
Join Date: Jun 2005
Location: Bulgaria, Varna
Posts: 280
|
My 5870, clocked at 900/5000 is getting like M11666 score.
There's something fishy here. With some other DirectCompute apps available (fractal generators), 5870 at stock clocks would reach ~2 TFLOPS effective performance.
__________________
Q9450 @ 8*475MHz (Zalman CNPS-9700) :: DFI LANParty UT P45-T2RS :: 2*2GB G.Skill F2-8800CL5 @ 1141MHz (5-5-5-15-PL7) :: WD VelociRaptor 300GB :: Sapphire HD 5870 :: CHIEFTEC 850W :: Win7 x86
|
|
|
11-04-2009, 08:28 AM
|
#4
|
|
Xtreme Addict
Join Date: Apr 2007
Posts: 1,195
|
Quote:
Originally Posted by fellix_bg
My 5870, clocked at 900/5000 is getting like M11666 score.
There's something fishy here. With some other DirectCompute apps available (fractal generators), 5870 at stock clocks would reach ~2 TFLOPS effective performance.
|
It completely depends on the ILP extracted from the code. If the developer doesn't go out of their way to ensure maximum utilization of AMD's hardware it could suffer. Same goes for Nvidia hardware of course but to a somewhat less extent. But even with that taken into consideration these results are ridiculous.
|
|
|
11-04-2009, 08:48 AM
|
#5
|
|
Xtreme Addict
Join Date: Oct 2006
Posts: 1,138
|
Well you have to remember, the Nvidia shaders are clocked significantly higher than ATI shaders...
Kinda weird that ATI is obliterating Nvidia in single threaded results though, ATI single threaded is about the same as their multi-thread score...
__________________
Quote:
|
Originally Posted by Shintai
Imagine Microsoft said that Windows 7 would only work on Acer/HP/Dell/Lenovo machines.
Its just Apple thats full of BS
|
Tyan S2932G2NR-SI
AMD Opteron quads @ 2.6GHz (Shanghai) x2
16GB of Kingston DDR-800 ECC RAM
ATI 4890 1GB
ASUS M3A79-T Deluxe
Phenom II 940 @ 3.75GHz w/ 1.4v
NB and HTT @ 2.75GHz w/ 1.4v & 1.3v
G.Skill PI @ DDR1150 w/ 2.1v
ATI 4890 @ 1022MHz core
|
|
|
11-04-2009, 08:59 AM
|
#6
|
|
Xtreme Member
Join Date: Jun 2005
Location: Bulgaria, Varna
Posts: 280
|
Quote:
Originally Posted by EniGmA1987
Well you have to remember, the Nvidia shaders are clocked significantly higher than ATI shaders...
|
...and ATi has a lot more of them.
This is not a sustainable argument for the case.
__________________
Q9450 @ 8*475MHz (Zalman CNPS-9700) :: DFI LANParty UT P45-T2RS :: 2*2GB G.Skill F2-8800CL5 @ 1141MHz (5-5-5-15-PL7) :: WD VelociRaptor 300GB :: Sapphire HD 5870 :: CHIEFTEC 850W :: Win7 x86
|
|
|
11-04-2009, 09:27 AM
|
#7
|
|
Xtreme Enthusiast
Join Date: Jun 2005
Posts: 774
|
The way it's meant to be benchmarked?
|
|
|
11-04-2009, 09:32 AM
|
#8
|
|
Xtreme Mentor
Join Date: Dec 2008
Location: Russia (for now)
Posts: 3,173
|
Quote:
Originally Posted by Piotrsama
The way it's meant to be benchmarked?
|
Nah, just buggy / broken release I guess. Will see if it gets better later on.
|
|
|
11-04-2009, 09:34 AM
|
#9
|
|
LOVE=MC²
Join Date: Dec 2002
Posts: 3,444
|
|
|
|
11-04-2009, 09:39 AM
|
#10
|
|
Registered User
Join Date: Oct 2009
Posts: 39
|
Hmm this benchmark is half done , the whole time my GTX 295 was running in low 2d mode and produced this score
__________________
.:. Bloomfield i7 920 4300Mhz .:. eVga GTX 295 Plus .:. Patriot DDR3 6GB .:. Silverstone 800W PSU .:. Asus P6TD Deluxe .:. Intel G2 SSD 80GB .:.
|
|
|
11-04-2009, 09:51 AM
|
#11
|
|
Registered User
Join Date: Dec 2008
Posts: 56
|
no directcompute support for 4850 with streamSDK beta2 driver (9.11 beta) O.o
|
|
|
11-04-2009, 11:43 AM
|
#12
|
|
Xtreme Addict
Join Date: Apr 2008
Location: Texas
Posts: 1,138
|
Quote:
Originally Posted by netkas
no directcompute support for 4850 with streamSDK beta2 driver (9.11 beta) O.o
|
REALLY?! I was looking forward to running this on my 4870 when I got home. BAH!
__________________
Phenom II 940 @3.2Ghz 1.35v(CACVC AC 0849GPAW)| NB@2.4Ghz 1.3v| 4GB Corsair Dominator@1066mhz 5-5-5-15-22-2T 2.1v + 4GB OCZ Reaper 1066 kits| ASUS M3A79-T Deluxe 1001 BIOS| Coolermaster Hyper 212 | PCP&C Silencer 750w | Diamond HD 4870 512mb | Fatality X-Fi | Areca ARC-1210 + RAID0 4xOCZ Solid Series 30GB SSDs | RAID0 2xWD 750GB Caviar Blacks |TV Wonder HD 650 PCIe
Quote:
Originally Posted by phelan1777
Hail fellow warrior albeit a surat Mercenary. I Hail to you from the Clans, Ghost Bear that is (Yes freebirth we still do and shall always view mercenaries with great disdain!) I have long been an honorable warrior of the mighty Warden Clan Ghost Bear the honorable Bekker surname. I salute your tenacity to show your freebirth sibkin their ignorance!
|
|
|
|
11-04-2009, 11:47 AM
|
#13
|
|
Xtreme Addict
Join Date: May 2007
Posts: 1,478
|
Doesn't seem like the benchmark is complete
__________________
*ASUS Rampage Formula * Intel Core 2 Extreme QX9650 (10 x 400.0) * 4 x 1 GB Crucial Ballistix DDR2 1066 * Sapphire Radeon HD 4870 X2 w/ ASUS TOP BIOS (811/1000) * Western Digital Raptor X * Seagate 7200.11 1TB * Seagate 7200.10 500GB * Corsair HX1000 PSU * CM Stacker 830SE Black * Dell 3007WFP-HC * Dell 2007FP (Portrait Mode) *
Dual WC Loops:
Loop 1: Swiftech MC-RES Reservoir -> Laing DDC 3.2 w/ XSPC Top -> Swiftech MCR-220 QP Rad -> D-Tek FuZion v2 w/ Quad Insert
Loop 2: Laing DDC 3.2 w/ XSPC Reservoir Top -> Thermochill PA120.3 w/ Shroud -> EK NB S-MAX Acetal Block -> EK 4870X2 FC Block
Fans: Yate Loon Mediums
|
|
|
11-04-2009, 01:35 PM
|
#14
|
|
Xtreme Addict
Join Date: Apr 2008
Location: Texas
Posts: 1,138
|
Quote:
Originally Posted by zerazax
Doesn't seem like the benchmark is complete
|
version 0.15
You are correct sir.
__________________
Phenom II 940 @3.2Ghz 1.35v(CACVC AC 0849GPAW)| NB@2.4Ghz 1.3v| 4GB Corsair Dominator@1066mhz 5-5-5-15-22-2T 2.1v + 4GB OCZ Reaper 1066 kits| ASUS M3A79-T Deluxe 1001 BIOS| Coolermaster Hyper 212 | PCP&C Silencer 750w | Diamond HD 4870 512mb | Fatality X-Fi | Areca ARC-1210 + RAID0 4xOCZ Solid Series 30GB SSDs | RAID0 2xWD 750GB Caviar Blacks |TV Wonder HD 650 PCIe
Quote:
Originally Posted by phelan1777
Hail fellow warrior albeit a surat Mercenary. I Hail to you from the Clans, Ghost Bear that is (Yes freebirth we still do and shall always view mercenaries with great disdain!) I have long been an honorable warrior of the mighty Warden Clan Ghost Bear the honorable Bekker surname. I salute your tenacity to show your freebirth sibkin their ignorance!
|
|
|
|
11-09-2009, 06:52 PM
|
#15
|
|
Xtreme Member
Join Date: Jul 2004
Posts: 269
|
version 0.25 is up. My 5870 (900/1215) now scores M35555. I7 @ 3.8 only scores M3038 though....
|
|
|
11-09-2009, 09:44 PM
|
#16
|
|
Xtreme Guru
Join Date: Oct 2004
Location: U.S.A.
Posts: 4,280
|
What I got with 285.
__________________
Watercooling parts:
Pa120.3 rad (CPU)|PA120.3 (for video card)|GTZ on CPU
|stock cooling on chipset|Laing DDC w/ Petra'sTech DDCT-01s Top on cpu loop|aqua extreme 50z pump on gpu l
cpu i7 960 @???
old cpui7 920 3836A762 @ 3.7Ghz
Gigabyte EX58-UD5 F9e bios
12GB of G.skill (6 x 2GB)DDR3 perfect storm F3-16000CL8-2GBPS 7-7-7-21
2x bfg 285
new Dell 30"LCD 3008wfp oh the colors!
LG GGW-H20L
COOLER MASTER Real Power Pro RS-C50-EMBA-D2 1250watt for the cpu/mobo & video
OCZ600MXSP 600W for the hard drives
areca 1280ML with 2GB ecc of cache running in other 16x pci-e slot with battery backup running two raid 10 arrays 2TB boot array 6TB storage
sound card: Auzentech Forte 7.1 | Gigaworks S750 7.1 speakers
OSes: debian Linux unstable, and Windows XP pro x64 w/sp2| testing windows 7 x64 ultimate
|
|
|
11-10-2009, 01:33 AM
|
#17
|
|
Registered User
Join Date: Nov 2006
Posts: 30
|
gtx295 and i7 920@3.8 with the 0.25 version.
|
|
|
11-10-2009, 03:59 AM
|
#18
|
|
Xtreme Addict
Join Date: Apr 2008
Location: Texas
Posts: 1,138
|
Quote:
Originally Posted by hurleybird
version 0.25 is up. My 5870 (900/1215) now scores M35555. I7 @ 3.8 only scores M3038 though....
|
WOW! That is a high score. Did they actually improve the way it functions on the card or just mess with how it scores different devices? The sharp drop in CPU score bothers me.
__________________
Phenom II 940 @3.2Ghz 1.35v(CACVC AC 0849GPAW)| NB@2.4Ghz 1.3v| 4GB Corsair Dominator@1066mhz 5-5-5-15-22-2T 2.1v + 4GB OCZ Reaper 1066 kits| ASUS M3A79-T Deluxe 1001 BIOS| Coolermaster Hyper 212 | PCP&C Silencer 750w | Diamond HD 4870 512mb | Fatality X-Fi | Areca ARC-1210 + RAID0 4xOCZ Solid Series 30GB SSDs | RAID0 2xWD 750GB Caviar Blacks |TV Wonder HD 650 PCIe
Quote:
Originally Posted by phelan1777
Hail fellow warrior albeit a surat Mercenary. I Hail to you from the Clans, Ghost Bear that is (Yes freebirth we still do and shall always view mercenaries with great disdain!) I have long been an honorable warrior of the mighty Warden Clan Ghost Bear the honorable Bekker surname. I salute your tenacity to show your freebirth sibkin their ignorance!
|
|
|
|
11-11-2009, 04:19 AM
|
#19
|
|
Xtreme Enthusiast
Join Date: Jan 2005
Location: Stafford, UK
Posts: 808
|
Hmmm, why is my stock E8600 getting a CPU score of M617?!
__________________
MAIN PC: DFI LP LT X38-T2R ~ C2D E8600 @ 3.33GHz ~ 4GB OCZ Blade LV PC9200 ~ XFX 8800GTS 320MB ~ 2 x WD Raptor 74GB (RAID-0) ~ Noisetaker 600W ~ Dell 2001FP ~ Vista Ultimate x64
HOME PC: Abit I-N73HD ~ PDC E5200 @ 3.33GHz ~ 2GB OCZ SLI-Ready PC8500 ~ PNY 8800GT 256MB ~ Samsung Spinpoint 250GB ~ OCZ Modstream 450W ~ Hyundai Q17 ~ Win 7 RC7100 x64
NETBOOK: Dell Inspiron Mini 9 ~ Atom N270 @ 1.60GHz ~ 2GB PNY PC6400 ~ Intel GMA950 ~ STEC 8GB SSD & 16GB SDHC Card ~ Intel 5100N & Bluetooth ~ Win 7 RC7100 x32
E-MAIL ~ FULL SPEC ~ 3DMark 01 03 05 06
|
|
|
11-11-2009, 08:32 PM
|
#20
|
|
LOVE=MC²
Join Date: Dec 2002
Posts: 3,444
|
air cooled at that
|
|
|
11-11-2009, 08:50 PM
|
#21
|
|
Xtreme Enthusiast
Join Date: Jul 2007
Posts: 800
|
Quote:
Originally Posted by NapalmV5
http://i35.tinypic.com/oaww9y.png
air cooled at that 
|
Why doesn't yours have "GPU Score" above the GPU score and no feature level listed? I'm not saying it is fake, it just seems odd and disproportionate with the other scores posted here.
|
|
|
11-11-2009, 11:41 PM
|
#22
|
|
LOVE=MC²
Join Date: Dec 2002
Posts: 3,444
|
i removed one of the vmods.. and reran it several times.. getting constant 113k
|
|
|
11-12-2009, 12:11 AM
|
#23
|
|
Xtreme Enthusiast
Join Date: Jul 2007
Posts: 800
|
I just noticed that the adapter properties box is missing in the first shot too. Also you lowered the clock on the core and shaders and the score went UP? Yeah, I officially don't trust this benchmark one bit.
|
|
|
11-12-2009, 02:36 PM
|
#24
|
|
LOVE=MC²
Join Date: Dec 2002
Posts: 3,444
|
what adapter properties box missing ??
temp and voltage have just as much affect on scores/benchmark as clocks do
this bench behaves just like cuda factorial does: http://www.xtremesystems.org/forums/...d.php?t=233280
if clocks pushed too hard/too much voltage/temp ^ then you get this wacky score
get to know the bench better before you call fake on someone
|
|
|
11-12-2009, 02:46 PM
|
#25
|
|
Xtreme Enthusiast
Join Date: Jul 2007
Posts: 800
|
I guess box isn't really the right term. The line enclosing the bottom 3 fields is missing in your 108K score shot.
And I'm not calling your shots fake. I'm saying that it doesn't seem to make for a very accurate/reliable benchmark. But you sure seem defensive about it, which makes me suspicious.
|
|
|
 |
|
| Thread Tools |
|
|
| Display Modes |
Rate This Thread |
Linear Mode
|
|
Posting Rules
|
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts
HTML code is Off
|
|
|
All times are GMT -8. The time now is 03:31 PM.
|