3DMark (PC): Download, benchmark results & release trailer

[Hyperhorn]

The PC version of the new 3DMark can be downloaded now from one of the download mirrors or Steam and I'm honestly pretty amazed one of my work mates managed to test the performance of 22 graphic cards in the test scenes Cloud Gate, Fire Strike Extreme and Ice Storm from 720p with no AA/AF up to 1600p with 8xSGSSAA 16:1 HQ-AF on day 1. If you're interested how HD 7970 (GE), HD 7950 (Boost), GTX 680, GTX 670, GTX 580 etc. perform please have a look at the results from PC Games Hardware:

- German version: http://www.pcgameshardware.de/3DMark-Software-122260/Tests/3DMark-im-Test-Benchmarks-Download-1047157/

- English version: http://translate.google.de/translate?sl=de&tl=en&js=n&prev=_t&hl=de&ie=UTF-8&eotf=1&u=http%3A%2F%2Fwww.pcgameshardware.de%2F3DMark-Software-122260%2FTests%2F3DMark-im-Test-Benchmarks-Download-1047157%2F

 

(Note: You can switch between the results by clicking the "tabs" at the head of the benchmark result box. As usual high quality driver settings were enabled, so you will likely get higher results with your default bench-settings.)

 

If you look at the results it's easy to see why AMD suggested Fire Strike as a good test scene in their press release.

 

Also here's the official Futuremark trailer released today:

 

What results have you got and what do you think about the new 3DMark from Futuremark? Please also post links to other GPU/CPU benchmarks results you find on the internet, so we get more data to get an idea about the scaling, what scores can be considered normal etc.!

[SF3D]

Here is my Fire Strike sample video with maximum details and 2560 x 1440 resolution:

 

[hokiealumnus]

What results have you got and what do you think about the new 3DMark from Futuremark? Please also post links to other GPU/CPU benchmarks results you find on the internet, so we get more data to get an idea about the scaling, what scores can be considered normal etc.!

 

Quick comparison with some high end cards:

 

 

Best score so far:

 

 

More details and such in our article. It's a fun bench. Hard on hardware as 3DMark benchmarks are wont to be and it's quick, which is a pleasant change.

[Christian Ney]

Tested it too: New 3DMark Tested - 25 GPU, CPU and Memory Scaling

 

There are all screenshots and "Run details"

 

 

[hokiealumnus]

Good lord Ney, we've only had the thing since Thursday. Did you lock yourself in you room with red bull and a bunch of GPUs all weekend?

[Christian Ney]

Almost ,

 

Started friday morning till friday night. Saturday I was at Splmann's place to overclock FM2/Trinity so I wasn't able to bench 3DMark anymore.

Then I started again on sunday morning and I finished a few hours ago that's why the article isn't available in french yet, need to sleep

 

Will do that tomorrow morning.

Edited February 4, 2013 by Christian Ney

[T0lsty]

win7 or win8 is better ?

[hokiealumnus]

I know right? Seriously Ney, how could you not test Win 7 vs Win 8???

 

[Christian Ney]

I hate windows 8

 

But I could do it for tomorrow if enough people ask for it. (enough => 100000000+)

[hokiealumnus]

CPU & RAM Frequency testing added to ours (courtesy MattNo5ss).

[FM_Jarnis]

So, how are ya liking it? Fried any GPUs yet? Any info if the temp data actually works with sub-zero temps? (we don't have liquid nitrogen setups to test... code says it should work)

 

Specifically interested in feedback on the "view run details" and how the result validation works (note that in Pro you need to enable online validation or manually click the "?" and hit revalidate. In Advanced it is always on as long as you have network connection)

 

Also there is now an "anti-cheat" validation link on top of the "view run details" window - it can say once and for all if the result is photoshopped or not (the only way you won't get the result from that link is if the user hides it on purpose and wants to block you from viewing it)

 

Also just a heads up; you really want to use betas. 13.1 drivers do not render everything correctly (so they are not approved) - some particles are missing from Fire Strike demo and GT2. Use 13.2 beta3/4 to get correctly rendered scenes (well, except for some minor flickering in crossfire)

 

310.90 NVIDIAs on the other hand do not have SLI profile and perform considerably less well than 313.95/96 betas... Okay, even these are not absolutely 100% perfect in SLI - again, some flickering there. Complex particle effects and smoke simulation seems to be hard .

 

13.2 betas and 313.95/96 betas are approved. Approval also shows in validation and in the future we aim to have less than 48h gap from "new driver" to "approved".

Edited February 4, 2013 by FM_Jarnis

[FM_Jarnis]

Also a small note that many review seem to get wrong - tho I must say we also haven't communicated this well. 3DMark actually does use DX11.1. In a very small and technical way.

 

From the FAQ (which has also lots of other good info);

 

Does 3DMark use DirectX 11.1?

 

Yes, but only in a minor way and with a fallback for DirectX 11 to ensure compatibility with the widest range of hardware and to ensure that all tests work with Windows Vista and Windows 7 as well as Windows 8.

 

DirectX 11.1 API features were evaluated and those that could be utilized to accelerate the rendering techniques in the tests designed to run on DirectX 11.0 were used.

 

Discard resources and resource views

 

In cases where subsequent Direct3D draw calls will overwrite the entire resource or resource view and the application knows this, but it is not possible for the display driver to deduce it, a discard call is made to help the driver in optimizing resource usage. If DirectX 11.1 is not supported, a clear call or no call at all is made instead, depending on the exact situation. This DX11.1 optimization may have a performance effect with multi-GPU setups or with hardware featuring tile based rendering (sometimes found in tablets and entry level notebooks).

 

16 bpp texture formats

 

The 16 bpp texture formats supported by DirectX 11.1 are used on Ice Storm game tests to store intermediate rendering results during post processing steps. If support for those formats is not found, 32 bpp formats are used instead. This optimization gives a noticeable performance effect on hardware for which the Ice Storm tests provide a suitable benchmark workload (tablets, entry level notebooks).

 

There are no visual differences between the tests when using DX11 or DX11.1 in 3DMark and the practical performance difference from these optimizations is generally limited to Ice Storm on very low end Windows hardware, and on Windows RT.

 

So.. in theory Win8 vs Win7 is a small difference due to these DX11.1 optimizations but mostly they matter with tile-based renderers (ie. PowerVR based tablets and netbooks, some WinRT tablets) and in some cases multi-GPU scenarios (not sure how well AMD/NVIDIA drivers handle such scenarios yet)

 

But no, no special eye candy. Because, frankly, DX11.1 doesn't offer anything much in that department.

Edited February 5, 2013 by FM_Jarnis

[TASOS]

Thanks for the info Jarnis

 

3DMARK looks good and runs smooth , no problems here.

Good info,nice graph.

 

You have done a very good job !!!

 

I see this as the best 3DMARK ever.

[Gunslinger]

Quick comparison with some high end cards:

 

Best score so far:

 

 

More details and such in our article. It's a fun bench. Hard on hardware as 3DMark benchmarks are wont to be and it's quick, which is a pleasant change.

 

Looks like Ivy/Z77 will own this benchmark. Same GPU's and clocks with a 4.7GHz 3930K is almost 2k lower, even though I crushed your Physics score.

 

 

http://www.3dmark.com/fs/15355

[hokiealumnus]

Wow yea, you definitely crushed the physics there.

 

At those GPU clocks though, you should have scored better. My 3960X scored ~100 points higher at the same clocks as the 3770K. What driver are you using? Make sure you're on 13.2 beta 4.

[dafour]

The submissions are all with pre 313.95/96 drivers,scores are way off..

[Gunslinger]

Wow yea, you definitely crushed the physics there.

 

At those GPU clocks though, you should have scored better. My 3960X scored ~100 points higher at the same clocks as the 3770K. What driver are you using? Make sure you're on 13.2 beta 4.

 

Pretty sure I'm using 13.2 beta 3 as it was listed on the "approved" list.

[DrWeez]

Any info if the temp data actually works with sub-zero temps? (we don't have liquid nitrogen setups to test... code says it should work)

 

sadly with my MSI 7970 and 680 Lightning the temp's did not drop past 0 on the chart.

There are a few new things to learn, 3DMark 11 is fine and runs no issues. Fire strike with the same settings would crash :D

[Guest cowgut]

Is beta5 approved?

 

Yup they are

Edited February 5, 2013 by cowgut

[FM_Jarnis]

Is beta5 approved?

 

Yup they are

 

We actually haven't tested but I think they will be.

 

Looks like AMD is not changing the version number while updating the betas

[FM_Jarnis]

sadly with my MSI 7970 and 680 Lightning the temp's did not drop past 0 on the chart.

There are a few new things to learn, 3DMark 11 is fine and runs no issues. Fire strike with the same settings would crash :D

 

Okay. Could be the chart drawing code... Will report this as a bug, will try to fix in future updates.

 

Fire strike is way harsher than anything in 3DMark 11 so it crashing on heavy overclocks where 3DM11 gets through is not unexpected. The sheer number of complex compute shader calls per frame in Fire Strike is... umm... we did hear from some vendors (no names) that there were too much compute shaders

 

But we like to torture GPUs to the limit. Only way to see what the're made of.

[GENiEBEN]

Okay. Could be the chart drawing code... Will report this as a bug, will try to fix in future updates.

 

Easy fix: invert sign before drawing into chart.

[FM_Jarnis]

If someone has a result file from sub-zero run, send one to info [at] futuremark.com and we can test our fix

[hokiealumnus]

Pretty sure I'm using 13.2 beta 3 as it was listed on the "approved" list.

 

FYI, I ran it with beta 4 and got the same score you did (roughly). I think the 14K score was on 13.2 beta 2. Or it was a fluke. Either way, 12K+ is more accurate for two 7970s and an Ivy chip at 4.9. Not sure what led to the 14K score unless it was the beta version.

 

Unfortunately, the time has come to pack up the ARES II and send it back, so my ability to test just ran out. Easy come, easy go.

[Lucky_n00b]

@ FM_Jarnis :

The Fire Strike Combined test(the fight scene) looks soooo damn cool !!! I hoped it would be a bit longer though :D

 

That said, the tests were shorter ,but it feels 'heavier' than 3dmark11. My 7970 can pass 3d11 on 1300mhz core easy, but fail to do the same on 3DMark Fire Strike Extreme.

 

Also, I found that the 3DMark Ice Storm benchmark is quite okay for benching some low-end netbook/notebook and windows-based tablet and the like, its definitely better than have to use the old '05 and '06.

 

Kudos to Futuremark Team, 3DMark really is your best benchmark until now

(a minor note, no offense, but as for the demos , I still prefer the old 3DMark99/2000/2001SE demo, can't beat that awesome soundtrack :p)