Today, Futuremark released 3DMark 11, the latest in their long-running series of benchmarking software. 3DMark, as the name suggests, tends to be aimed towards benchmarking gaming PCs and provides users with a score that can be used to compare the relative performance of different systems for gaming. Futuremark has kindly provided us with press copy of 3DMark 11 Professional for a few days, so we can go through some of the features of the new software and reveal some of the first 3DMark11 scores. Unlike some previous versions, the 11 in 3DMark11 stands for ‘DirectX 11′ rather than ’2011′.
Features and Specifications
As stated, this is a DirectX11 benchmark, which means it will probably thrash even current bleeding edge hardware; in the sample tests below, frame rates were pretty low. As with previous versions, there will be be a free Basic Edition (no expiry date or run limits) and an Advanced Edition ($19.95 RRP, with some extra toys), as well as a Professional version (licensed for commercial use, coming in at a whopping $995). As well as a battery of tests and online results submission, 3DMark11 also features a demo mode (based on the ‘Deep Sea’ and ‘High Temple’ tests) with a soundtrack. While the Basic Edition only allows use of the ‘Performance’ preset (1280 x 720 resolution; scores are designated Pnnnn) the Advanced and Professional versions allow use of the Entry (1024 x 600 resolution; Ennnn) and Extreme (1920 x 1080 resolution; Xnnnn) presets, as well as custom settings.
The only really massive difference, to overclockers, in the Professional Edition is the Image Quality: this allows high quality renders of any given point in any test. Futuremark has kindly provided us with a Professional license, which should allow us to pick out image quality differences between cards, operating systems and drivers in the future.
The Tests
There are six tests: four test the graphics capability of the machine, one is a physics test designed to put the CPU through its paces and one tests the CPU and GPU for physics-related capability. Realistic physics is one of the major benefits of modern games, but do require some extra processing power to deal with properly.
The graphics tests all use a lot of post-processing to add various camera lens effects, such as altering the depth of field.
- Test 1 is heavy on light sources, shadows and volumetric lighting.
- Test 2 has some lights, and tessellation of many parts of the scene.
- Test 3 employs tessellation and some volumetric lighting and shadows.
- Test 4 also features tessellation and volumetric lighting but casts a large number of shadows.
The following YouTube videos are courtesy of Futuremark.
The two physics tests will stress different parts of the PC.
- The Physics Test deals with rigid-body physics on a large number of objects, at a fixed resolution. Eye-candy is kept to a minimum here.
- The Combined Test also tests the CPU with rigid-body physics calculations, but puts the GPU to use on volumetric lighting, tessellation, post-processing and soft-body physics-simulation.
As you can see, all of these tests will be particularly heavy on the GPU; only one test is really designed to assess your CPU. However, this is appropriate given the incredible capabilities of modern graphics chips. Although the minimum specifications suggest you need a 1.8 GHz dual-core CPU and 1 GB RAM you’re definitely going to need more horsepower, as our sample results below illustrate.
How Will Your Rig Perform?
Thanks to my colleague Jeremy (who has more computer hardware than your local store) we have some sample scores, on both Intel and AMD platforms.
Note: 3DMark11 apparently works best when run as ‘Administrator’, although the stability issues will probably be worked out in the first patch.
Pre-DirectX 11 Cards
These don’t work. The program will refuse point blank to start: below is what I was greeted with when trying to start 3DMark11 on my Lenovo ThinkPad W510 (nVidia Quadro FX880M).
AMD Hex-Core Build
- CPU: AMD Phenom II X6 ‘Thuban’ 1100T (overclocked)
- Motherboard: Asus Crosshair IV Formula
- RAM: 4 GB of G.Skill Flare DDR3-2000
- GPU: Asus Matrix 5870 Platinum
| CPU @ 4.0 GHz | CPU @ 4.26 GHz | |
| GPU @ 900 MHz / 1200 MHz | P4433 | P4444 |
| GPU @ 975 MHz / 1225 MHz | P4713 | P4731 |
Intel Quad-Core Build
- CPU: Intel Core i7-870 (overclocked)
- Motherboard: EVGA P55 FTW
- RAM: 4GB of G.Skill DDR3-2400
- GPU: AMD HD6870
| CPU@ 4.0 GHz | CPU @ 4.4 GHz | |
| GPU @ 900 MHz/1050 MHz | P4369 | P4391 |
| GPU @ 1025 MHz / 1175 MHz | P4884 | P4908 |
Conclusions
These sorts of benchmarks are intended to be valid for some time to come, so it’s no surprise that these two very up-to-date machines struggle with this initial version of 3DMark11. It will probably be one or two generations of cards (GT780 anyone?) before this benchmark becomes even close to trivial for hardware. The GPU influence on these scores is massive: a mere 25 MHz on the core and RAM (for a 5870) nets you ca. 300 3DMarks, and 125 MHz on the core and RAM on a 6870 scores an extra 500+ 3DMarks. In stark contrast, 5 to 10% extra CPU speed gets you only 10 – 30 extra 3DMarks. These are just preliminary runs, but I’m sure there will be a lot more trends emerging in the near future.
According to the published formulae, in order to hit 10,000 ‘P’ 3DMarks, you’ll need 40 FPS across the board, or 45 FPS in all graphics tests plus 35 FPS in the physics tests.
Clearly, our benching team will have to get their GTX580s and the like under some liquid nitrogen to see some boost in 3DMarks!
If you’ve downloaded and run 3DMark11 (and we hope you do), please post your score below (‘Leave a Comment’) along with your hardware specification.
- David
Tags: 3dmark11, Futuremark












12-07-10 02:12 PM
Screenshots???
12-07-10 02:14 PM
Hokie had the hardware for the tests but didn't have time to write the article, so I wrote it. However, I didn't realise the FX880M isn't DirectX11....
If anyone wants to run it and grab some screenshots, send me them as png or jpeg ideally, and 800 pixels on the longest edge, to david [at] overclockers doot com I'll add them to the article and credit you appropriately.
12-07-10 02:53 PM
12-07-10 04:04 PM
12-07-10 04:06 PM
12-07-10 04:11 PM
Thats like saying a new dragster is competing with Honda's minivan.
People playing games are not worried about the release of a benchmark. Just as most benchers could care less about the new release of WoW. Sure there is some crossover, but I just cant think of a way these would compete being such different applications.
12-07-10 04:18 PM
I'll try & take a screenshot or two at lunch. Looking for anything specific?
12-07-10 04:21 PM
12-07-10 04:34 PM
For anyone who saw this post and had no idea what we were talking about, the front page article is now visible to the public. I blame David.
12-07-10 04:41 PM
12-07-10 05:06 PM
Edit, Word at Futuremark's forum is that nobody's able to get SLI to work.
12-07-10 05:11 PM
12-07-10 05:18 PM
12-07-10 06:10 PM
12-07-10 06:21 PM
Graphics Test 1
Graphics Test 2
Graphics Test 3
Graphics Test 4
Physics Test (CPU only)
Combined Test (Physics on CPU, with the rest on the GPU)
Now, these can be made a lot prettier. This is the way they appear with the Performance preset; went with that since it's how most people will see it.
12-07-10 06:22 PM
12-07-10 06:25 PM
12-07-10 06:32 PM
that what you're looking for maybe?
12-07-10 06:45 PM
12-07-10 06:48 PM
12-07-10 07:44 PM
12-07-10 07:51 PM
12-07-10 07:56 PM
12-07-10 07:58 PM
12-07-10 10:43 PM
3.6ghz quad
http://3dmark.com/3dm11/25275
P4353
http://3dmark.com/3dm11/25746
X1595
12-07-10 11:00 PM
Weird also that 10.10 drivers are not supported? I may revert back to 10.9.
12-08-10 12:19 AM
http://3dmark.com/3dm11/33947
X1648.
http://3dmark.com/3dm11/34308
Stock CPU (3.3GHz) no HT, GPU @ 775/1550/1800.
EDIT: Scaling for CPU and HT sucks. I raised my clocks to 4Ghz, added HT, bumped GPU up 25 on core, and scored only 280 more points. Bumped GPU up again (825) and scores another 200 or so. Sitting at 5528 now with this 470.
http://3dmark.com/3dm11/38091
Bleh 5636 @ 850/1700/1800. Thats all shes got on air captain! Hurry up benching station so this little heater can get under water!
12-08-10 08:57 AM
http://3dmark.com/3dm11/30031
What can I say ( at least about basic version ) is that test is ugly, short and I was expecting something better ...
Using physx won't change score ( I got 1 mark difference after turning off physx on my card ), all based on direct compute so better for rankings. Still it's better to use physx, why ? ... because it's not using cpu so much in last tests and at least I had to lower cpu clock a bit so it wouldn't crash ( seems like it's not using multi cores in most tests except physics but I didn't have time check all ).
12-08-10 12:05 PM
http://forums.techpowerup.com/showpo...&postcount=158
12-08-10 01:18 PM
12-08-10 01:36 PM
12-08-10 01:45 PM
OP updated with results. Preliminarily, seems the generational difference from C2Q to i7 (and Thuban) -and thus either double the threads or 50% more cores- does make a difference, even if higher clocks don't help all that much.
EDIT - To make things easier for whichever mod edits the OP, only the top score will be listed. Thanks!
12-08-10 01:50 PM
To the contrary...sort of.
The jump I made from 3.3Ghz no HT to 4Ghz w/HT (and a 25 core clock speed increase on the 470), net me about 280 points. When I jumped another 25 core and left it as is, I jumped 150 points. So 700Mhz of clock speed and HT was good for 100 points.
EDIT: Now I realize that is wholly unscientific and probably not right, but one would think such a drastic difference would net a bit more than 2%...
EDIT2: See post 28 for actual details. Apparently Im tired as all get out and butchering my numbers.
12-08-10 01:59 PM
12-08-10 02:00 PM
12-08-10 02:01 PM
That had to be the fastest reply I have ever witnessed! Nice!
12-08-10 02:12 PM
12-08-10 07:33 PM
P3600
http://3dmark.com/3dm11/63935
Q9300 @ 3.33 ghz
Evga 460 @ 900/1800/2200
12-08-10 10:20 PM
Plus, now Vantage is corrupted.
12-09-10 02:05 AM
1. Does not recognize SLI with 260.99 drivers.
2. Does not recognize independent PhysX card.
3. Fubars Vantage......
Score for now....
12-09-10 03:55 AM
12-09-10 04:21 AM
http://3dmark.com/3dm11/74278
12-09-10 04:36 AM
ANyhoo on the SLI issue...
OR
http://www.nvnews.net/vbulletin/show...2&postcount=34
Really, dont worry about overclocking the CPU...
12-09-10 04:36 AM
When heaven is added to the list of scoring benches at the bot, I'll have to make some submissions.
12-09-10 04:38 AM
EDIT: Grrrrrrr Woomack... killing me on that 470 man!!! THat 920 core is huge. Vmodded (bios and OCP)? water?
12-09-10 04:45 AM
12-09-10 04:49 AM
12-09-10 04:55 AM
12-09-10 05:10 AM
EDIT - It's looking like he should be around 10,000. There appears to be a beta driver b/c high scorers are using a different version that's not FM approved.
12-09-10 05:16 AM
EDIT: Woo was also on 260.99
12-09-10 12:04 PM
Somewhat off-topic, but how long has FutureMark used such blantant branding? (The msi submersibles and antec jeeps, in this case.) Is the branding present in the paid licensed software as well?
12-09-10 12:32 PM
12-09-10 12:45 PM
Coincidentally 06 was the last tolerable 3dmark.
12-09-10 01:07 PM
12-09-10 01:10 PM
12-09-10 01:17 PM
12-09-10 01:41 PM
12-09-10 01:43 PM
12-09-10 01:56 PM
12-09-10 02:08 PM
12-09-10 02:20 PM
12-09-10 02:20 PM
12-09-10 02:27 PM
12-09-10 02:34 PM
Version:
263.09 WHQL
12-09-10 02:38 PM
Stock GTX580 should have score near 6k or at least I saw many 5800-6100.
about drivers I was reading somewhere that latest betas are fixing most problems with sli and other things but sometimes are weird visual effects
most beta drivers for nvidia are on guru3d
12-09-10 02:43 PM
12-09-10 02:51 PM
12-09-10 02:55 PM
Other thing, for nvidia better use physx on so it won't keep your cpu under full load in last tests and will let you oc cpu higher ( at least works for me ). Seems like there is some difference between 2 and 4 cores but almost nothing with more cores and ht seems useless.
12-09-10 02:57 PM
Looks like 1366 platform definitely has the upper hand on others
12-09-10 02:59 PM
12-09-10 02:59 PM
12-09-10 03:00 PM
I downloaded 3D11 on Monday and still haven't had time to get around to benching it
12-09-10 03:01 PM
5870 at 950/1300
920 is at 4.2 Ghz
http://3dmark.com/3dm11/86306
12-09-10 03:16 PM
Do you have the advanced version? If so, go to the advanced tab and press the performance button and make sure you disable the audio and demo so its only the 6 benches.
Is PhysX enabled on the card?
12-09-10 03:57 PM
12-09-10 05:27 PM
12-09-10 05:28 PM
12-09-10 08:06 PM
12-09-10 08:18 PM
Gah.
12-10-10 12:47 AM
W3680 at 4.5 Ghz
GTX 580 at 910
http://3dmark.com/3dm11/99576
12-10-10 12:54 AM
12-10-10 12:59 AM
and my card wont run the same settings as vantage.
12-10-10 01:01 AM
12-10-10 01:23 AM
12-10-10 05:15 AM
P5044 3DMarks
http://3dmark.com/3dm11/101868
12-10-10 05:30 AM
With Brolloks posting a similar system you should be more in the 4800-5000 range.
My score was only 43xx partly because I had all of my apps still open, including anti-virus and other cpu related programs. A couple more runs on a clean system and no oc on my 5870 (875/1225) put me closer to 4400 @ 3.6ghz.
12-10-10 05:37 AM
12-10-10 05:41 AM
http://3dmark.com/3dm11/101640
12-10-10 06:56 AM
4724
i7 920 @ 4ghz / Radeon 5870 900/1300
12-10-10 09:27 AM
http://3dmark.com/3dm11/97312
single GTX470@915/2060MHz , i7 950@4.2GHz
260.99 driver / all auto settings with performance profile
12-10-10 09:37 AM
I get through all the graphics tests no problem but after it loads the first physics test, the screen just goes blank and nothing happens... I've tried downclocking my cpu slightly, turning off HT, bumping vcore a notch... nothing works....
Fixed... oddly enough it was a driver issue, updated to 10.11 and can run the cpu tests while overclocked... but the dreaded overclocked GPU dual screen flicker is back!! C'mon AMD!!
Anyway...
P7802
http://3dmark.com/3dm11/106391;jsess...v8Bt5jDtpp2bEt
Increased to 4GHz, P7891... not a very big jump... 5970 is at 900/1200
http://3dmark.com/3dm11/106879;jsess...pcm05Results=0
Bumped my core clock to 950MHz and got a bigger jump... P8329 (i7 920@4GHz with HT on, 5970 @950MHz core and 1200MHz RAM)
http://3dmark.com/3dm11/107398;jsess...pcm05Results=0
12-10-10 12:22 PM
Athlon II 620 x4 @ 3.2ghz DDR2 @ 1000 w/ 5830 = 2653
Just playing with it, and a system I built for a family member.
:edit:
Wow...
Oc'ed my 5830 to 900/1200 and got 2856.
Man, too bad I cant crossfire the 5770 with my 5830 for a bit of more fun.
12-10-10 04:09 PM
P6464
http://3dmark.com/3dm11/110809
12-10-10 04:39 PM
christmas presents put me all out of my money
12-10-10 04:48 PM
12-10-10 04:50 PM
nah, pot comes in for christmas. so christmas break i will officially join!
12-10-10 05:28 PM
12-10-10 06:20 PM
First time retail 980X 3013A489 clocks with single HD5870 at 4.0Ghz vs 4.2Ghz @1025/1263 - 6pts diff LOL
2x HD5870 CFX
Tried 3x HD5870 @1000/1250 at 4.2ghz first
Extreme Preset
12-10-10 07:34 PM
12-10-10 10:57 PM
Is this hindering my score? or did that only show up on the report and not effect the benchmark itself?
If it shouldnt be that way, what setting do i change?
12-11-10 12:51 AM
i7 960 on 4.2 ghz,
matrix platinum + xfx 5870 crossfire 900/1200
**** my xfx 5870 is downgrade my platinum matrix :-(
with itracker2 i can hit stable 1000 gpu on matrix but i can not use itracker because freezing pc with mi xfx in crossfire
i try msi afterburner but can not put voltage to card and i also in msi after... enabled unoficial overvoltage but not function
what can i do to overclock these to card with one soft ?
any advice is welcome, please send private message thanx
12-11-10 04:40 AM
Why dont you leave aferburner up and run the bench. Then you can look at the graph of the clockspeeds of the card while you were benching. I would bet good money they are just wrong. But double check to confirm, it should only take a second.
12-11-10 10:34 AM
http://3dmark.com/3dm11/125857
Core i7 965 @ 4GHz
HD5970 @ 900/1100MHz
Hmm, makes me feel like my system is slow again
12-11-10 01:19 PM
Demi post #63
http://www.yougamers.com/forum/showt...=138917&page=7
yay might be able to play now
edit yay i can play but not after the above it did work but then another error showe up easier to restore
http://3dmark.com/3dm11/136203
12-11-10 07:42 PM
anyhow pic to boot
12-11-10 09:19 PM
12-11-10 09:32 PM
off to bed now up 21 hrs zzzz
12-12-10 02:33 AM
http://3dmark.com/3dm11/139630
i7 920 @ 4.0
GTX 580 @ stock (772/2004)
12-12-10 02:44 AM
hey dejo.. been busy with work and medical health issues for family and myself but all cleared now
better scaling now
http://3dmark.com/3dm11/133584
12-12-10 04:46 AM
I'm almost ashamed of posting mine
P1542 --->
What else would anyone expect from a sub $500 setup!!!!
12-12-10 06:58 AM
X2702 http://3dmark.com/3dm11/131755
P6702 http://3dmark.com/3dm11/111209
12-12-10 05:34 PM
lack of psu power holding me back full report at
http://i4memory.com/f54/my-first-ret...70/#post138654
12-12-10 08:33 PM
4 5870's??
No need for a heater for that room..
12-12-10 09:09 PM
12-13-10 01:41 AM
http://3dmark.com/3dm11/157056
12-13-10 08:27 AM
GTX470@ 920/2100, i7 950@4.2GHz and I think it's max on my cooling
12-13-10 05:28 PM
sorted my psu issues by adding 2nd psu http://i4memory.com/f54/my-first-ret...70/#post138699
14,619 http://3dmark.com/3dm11/164326
12-13-10 05:35 PM
Great score!
12-13-10 07:11 PM
12-13-10 07:20 PM
12-14-10 06:33 AM
12-14-10 09:12 AM
One day I forgot to change gpu clock after benching ( usually I'm keeping this card underclocked
12-14-10 01:44 PM
12-14-10 02:50 PM
Edit: checked I have OC version
12-14-10 05:41 PM
12-14-10 06:41 PM
great stuff eva
12-15-10 07:45 AM
very nice
12-15-10 03:14 PM
http://3dmark.com/3dm11/101893
Eva come catch my 480's
12-15-10 03:15 PM
12-15-10 04:14 PM
12-15-10 05:32 PM
12-15-10 05:34 PM
12-15-10 05:37 PM
12-15-10 05:43 PM
12-15-10 05:45 PM
12-15-10 05:47 PM
Up to you. The first benches are just to get in the door. They dont need to break records or score boints.
12-15-10 05:52 PM
12-15-10 05:52 PM
12-15-10 06:35 PM
12-15-10 07:34 PM
finally, broke 15k 3dmark11 with HD5970 + 2x HD5870 http://i4memory.com/f54/my-first-ret...70/#post138789
3dmark11 = 15,049 http://3dmark.com/3dm11/189647
i7 980X @4.7ghz
HD5970@995/1297 + 2xHD5870 @1030/1282
Asus P6T6 WS Revolution
HX1000 + OP1200
Powermate measured 785 watts from HX1000 psu, so ~340-350w from OP1200 = 1,135 watts for system
12-15-10 10:19 PM
great score though
12-16-10 08:53 PM
I wonder if thats a bug or just Crossfire doing its thing in the 10.5's though...
12-18-10 06:42 PM
12-20-10 01:34 AM
http://3dmark.com/3dm11/231769
i7-750 @ 4ghz
Asus 5850 @ 986/1181
I gained over 1000 marks just from the overclock on the video card (stock clocks in 5850 yielded P3739)!
12-20-10 03:00 AM
p4492, q9550 - 3.6ghz, 6950 2gb, Stock
12-20-10 02:56 PM
12-20-10 02:58 PM
Jeremy, I can go ahead and start an 'official' type thread for this like the other 3D's...
12-20-10 02:58 PM
12-20-10 03:01 PM
12-20-10 03:07 PM
12-20-10 03:15 PM
12-20-10 03:34 PM
Any chance, when you have a moment, you want to link it up to the first post?
12-20-10 03:46 PM
12-20-10 03:55 PM
12-20-10 03:59 PM
12-20-10 04:03 PM
http://3dmark.com/receivingfailed?er...pcm05Results=5
12-20-10 04:14 PM
12-20-10 06:23 PM
12-20-10 06:54 PM
12-23-10 04:55 AM
5970 - 900/1200 (stock is 725/1100)
12-23-10 08:29 AM
Man, good job with that 5970. Glad it worked out for you, I was worried this morning!
12-23-10 04:54 PM
12-23-10 09:17 PM
Deal of the year!
12-23-10 10:44 PM
12-24-10 02:58 AM
12-26-10 10:36 PM
12-27-10 02:12 PM
12-27-10 03:58 PM
12-27-10 03:58 PM
12-28-10 01:33 AM
Yeah, I tried to look for 3DMark11 on HWbot did not see a place for a submission, am I missing something?
12-28-10 02:06 AM
06-07-11 01:04 PM