- Joined
- Nov 1, 1998
-
Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!
FRONTPAGE Futuremark Launches 3DMark Cross-Platform Benchmark
- Thread starter Overclockers.com
- Start date
As with all new benchmarks, it takes time. People are going to try to find ANYTHING to get the best score out of it so that stuff needs to be tested... so, I imagine it will be, just dont expect it soon if their history is anything to go off of.
Id imagine so, it depends on their requirements for it (as far as GPUz, CPUz, details of the bench, etc) but why not just run it when it is actually a boint receiving benchmark.
EDIT: I do wonder if the SLI scaling would be better with like cards as opposed to the Hybrid config tested though? Clearly its an issue from the AMD side, I just wonder if the Hybrid 'hack' has anything to do with it.
EDIT: I do wonder if the SLI scaling would be better with like cards as opposed to the Hybrid config tested though? Clearly its an issue from the AMD side, I just wonder if the Hybrid 'hack' has anything to do with it.
- Joined
- Oct 14, 2007
In my brief conversation with Futuremark, NVIDIA needs as much work on their drivers as AMD does to get multiple-card scaling to work properly.
One thing I think I forgot to mention in the article that's nifty - 3DMark has the ability to detect when you turn Tessellation off in Catalyst. If you do so, it will tell you your run was invalid. Thus, all runs listed here were run with Tessellation (and detail, for that matter) at default.
The same cannot be said about LoD. I don't have a green card at the moment, so I can't test whether LoD makes any difference, but 3DMark can't detect that because of the way it's implemented.
One thing I think I forgot to mention in the article that's nifty - 3DMark has the ability to detect when you turn Tessellation off in Catalyst. If you do so, it will tell you your run was invalid. Thus, all runs listed here were run with Tessellation (and detail, for that matter) at default.
The same cannot be said about LoD. I don't have a green card at the moment, so I can't test whether LoD makes any difference, but 3DMark can't detect that because of the way it's implemented.
- Joined
- May 10, 2009
Futuremark's been able to tell about tessellation for a while now, 3d11 has that ability too. It just shows up on the Orb though, not in the main window.
For myself, I'd rather they didn't add yet another benchmark you have to pay for to HWBot, especially when it's not testing any new features (3d11 for instance added DX11, Vantage added DX10).
For myself, I'd rather they didn't add yet another benchmark you have to pay for to HWBot, especially when it's not testing any new features (3d11 for instance added DX11, Vantage added DX10).
- Joined
- Oct 14, 2007
I doubt they'll be adding this to the bot very soon.
Side note - we received an email with pricing info and (most importantly) announcing a 25% discount for the advanced edition if you already have a 3DMark 11 license.
Side note - we received an email with pricing info and (most importantly) announcing a 25% discount for the advanced edition if you already have a 3DMark 11 license.
- Joined
- Aug 11, 2008
Still have to do some RAM timing, GPU core, and GPU vRAM tests. Here are a couple graphs and a table...
CPU speed from 4.0-4.8GHz increased the overall score by 0.83%
RAM speed from 1066-2133MHz increased overall score by 0.5%
From the table below combined with the CPU speed chart from above, we can tell that the Graphics Score doesn't care about the CPU, and that CPU cores/threads are more important than CPU speed for the Physics score. Also, it looks like the CPU doesn't start bottlenecking the GTX670 in the Combined Test until we get down to 1c/2t or 1c/1t.
CPU speed from 4.0-4.8GHz increased the overall score by 0.83%
RAM speed from 1066-2133MHz increased overall score by 0.5%
From the table below combined with the CPU speed chart from above, we can tell that the Graphics Score doesn't care about the CPU, and that CPU cores/threads are more important than CPU speed for the Physics score. Also, it looks like the CPU doesn't start bottlenecking the GTX670 in the Combined Test until we get down to 1c/2t or 1c/1t.
my scores:
Operating system
64-bit Windows 7 (6.1.7601)
Motherboard
BIOSTAR Group TA785GE 128M
Memory
8,192 MB 4x2 Module 1
2,048 MB Corsair DDR2 @ 400 MHz
Detailed scores
3DMark Score
4670 3DMarks
Graphics Score
5211
Physics Score
6178
Combined Score
2179
Graphics Test 1
24.34229278564453 FPS
Graphics Test 2
21.19338035583496 FPS
Physics Test
19.615468978881836 FPS
Combined Test
10.136276245117188 FPS
Catalyst 13.2beta
Phenom II X6 1065t
Gigabyte 7950
8GB RAM
Operating system
64-bit Windows 7 (6.1.7601)
Motherboard
BIOSTAR Group TA785GE 128M
Memory
8,192 MB 4x2 Module 1
2,048 MB Corsair DDR2 @ 400 MHz
Detailed scores
3DMark Score
4670 3DMarks
Graphics Score
5211
Physics Score
6178
Combined Score
2179
Graphics Test 1
24.34229278564453 FPS
Graphics Test 2
21.19338035583496 FPS
Physics Test
19.615468978881836 FPS
Combined Test
10.136276245117188 FPS
Catalyst 13.2beta
Phenom II X6 1065t
Gigabyte 7950
8GB RAM
- Joined
- Dec 10, 2011
- Location
- Cádiz (Spain)
3DMark for Steam is not working for me, complaining about Steam not being on. It obviously is. I think it's something on Valve's side, as 3DM11 isn't working either.
PS: I finally finished my yearly watercooling rebuild (Of all the things, I ran out of TIM. I usually buy 2x30g MX-2 syringes. I was not expecting this
). Feels great to use a mechanical keyboard and a 470 SLi again...
PS: I finally finished my yearly watercooling rebuild (Of all the things, I ran out of TIM. I usually buy 2x30g MX-2 syringes. I was not expecting this
). Feels great to use a mechanical keyboard and a 470 SLi again...Nice review Hokie, does anyone find the picture quality lacking in contrast?
Seems a bit, Milk.
My score.
http://www.3dmark.com/3dm/44226
Score 5605 with AMD Radeon HD 7870(1x) and AMD Phenom II X6 1090T
CPU: 4.1Ghz | GPU: 1200 / 1500
Graphics Score 6258
Physics Score 8731
Combined Score 2417
Seems a bit, Milk.
My score.
http://www.3dmark.com/3dm/44226
Score 5605 with AMD Radeon HD 7870(1x) and AMD Phenom II X6 1090T
CPU: 4.1Ghz | GPU: 1200 / 1500
Graphics Score 6258
Physics Score 8731
Combined Score 2417
- Joined
- Oct 14, 2007
In Cloud Gate, it absolutely lacks contrast. It's like you're in the clouds, which seems like their intent....but it's so cloudy & sunny, you can barely see anything.
Yeah... There needs to be a higher definition between objects and the background, it almost all merges together.
- Joined
- May 30, 2012
- Location
- California
Man, the firestorm demo was almost unwatchable on my rig and during the benchmarks I was getting like 7fps... my 3mark score was 4769marks. It also used 1021mb of my vram I need newer cards for this.
I need newer cards for this.Ambient Occlusion overkill?
3DMark for Steam is not working for me, complaining about Steam not being on. It obviously is. I think it's something on Valve's side, as 3DM11 isn't working either.
PS: I finally finished my yearly watercooling rebuild (Of all the things, I ran out of TIM. I usually buy 2x30g MX-2 syringes. I was not expecting this
3D Mark can also be downloaded from Guru3d and Techpowerup without Steam:
http://www.guru3d.com/files_details/3dmark_download.html (links at the way bottom)
http://www.techpowerup.com/downloads/2201/3DMark_v1.00.html
You might need to login to get it activated, then try running it in Offline mode, unless it has to run online so it can upload the results to Futuremark. My drivers are beta, so I didn't bother.
Similar threads
