I still find that cpu score very suspicious, why would 8800gts hider cpu score that much, doesnt make any sense at all.. With IvyBirdge & 8800gts you had 50 points but now with SandyBridge & 680gtx a whopping 1100+ points? heh xD Do you have a 500 series gpu too? If yes could you please test it with your SandyBridge cpu? Edit: And could you test 310.70 driver with current SandyBridge + 680gtx to see if it makes any difference by each test? I see you used 306.97, thanks
Hey there, welcome to Guru3D First of all, i love the work you guys are doing.. Im a huge Fan of the Demo Scene since i had my c64, hrhr.. I was allways impressed with the demos you guys did, Copy Paste from 2002 or Lady of Dark Autumn just to name some of my favs.. Well, i loved 195/95 and into the Pink..I really hope we see more from you guys in the Future :banana: anyway, enough dreaming of the good old days.. The Benchmark indeed is inspired by the Demo Scene, im very shure and is something refreshing different. Love the Music also.. I do have to admit, that theres something off with the CPU - Physics score.. The homepage is great, i think much nicer then the one frm futuremark, aldo its hard to compare and navigate trough the pages. That could be better.. I do love the Group thing..i would love to see that the best bench runs would display in teh group news automaticly, if a member submitts a new run, etc.. but i guess its enough time left.. Keep it up and good to hear ur on a new Beta version..Capturing CPU speed under load was the first thing i though off, and look at here, you guys are allready on it..Its also good to see you guys care about feedback, since u regged here Greetz -BetA-
Great Post BetA, I miss the Amiga demo days so much ; (btw first post in 7 years, sorry guys I've been reading) Also the other reason I post here again is regarding the benchmark, I think I have discovered a small exploit. I have submitted a report, if you did not receive it please let me know and I will resubmit.
All I'd like to know is, does this provide specific/nice indications of an unstable overclock? Like, counting errors or something?
Something wierd is goin' on here, I just disabled SLi and CF for both my rigs, here's the score for my HD7970 @1100/1450, 3960X @4.25ghz: And here's the MSI GTX670 OC Ed + FX8120 @3.7ghz: What I find to be weird is, look at the CPU + GPU score, the FX8120 + GTX670 did better in the first test (GPU + CPU), while the GTX670 + FX8120 combo is beaten in the 1x CPU + 3x GPU tests. I really don't know what to make of it.:3eyes: I ran them at the same default Tiger setting, only diff was the HD7970 run was done offline.....
It's a BD, so I had more or less expected bad score, but my gripe is, despite the horrendous cpu score, it still comes within ~300 points of my HD7970 + 3960X score.
Phenom II X4 stock 3.4GHz Sapphire Reference HD6870 stock 8GB (2x4GB) 1333 DDR3 Vertex 4 (slow for an SSD)
i see, so not only does it dislike AMD cards it also gives you a bad cpu score if you have one too, just for good measure.
loading times are highly affected by ssd/hdd and cpu ipc, latter is underestimated by many.. your system is bottleneck'd by your cpu just like mine in this particular test..
Hi, What we know for sure is that Crossfire has some problems in scalling and SLI does not. The problem is that our 4 cards rig has stuck at the border because of Christmas, and we can't make any tests . Currently when we unlock the full scaling speed at ATI - we got visual corruption. But it's a matter of getting the hardware. On the other hand on single card configuration AMD GPU's are performing quite good and I don't worry about that.
To explain CPU test. CPU test is not artificial. We could just calculate PI on all cores/threads. And that would be perfect. Our CPU test is combined with GPU, because of rendering. In order to compare different CPUs, you need to stay on the same GPU. Notice that on high end cards there's not too much difference in CPU test in different profiles. This scene is highly CPU bound. CPU is used to handle a lot of rendering batches, which is heavily dependent of drivers and also does physics on all cores. This test should resemble the style of programming that is used on current consoles. You use maximum CPU to save time of GPU. That's why we call it a CPU test. The best GPU test is raymarch test. It tests GPUs ALU power. The fluid test is depends on GPU memory speed. The fur tests geometry shader performance.
BetA: Thanks man! As for features. We don't want to release it and forget about it. We started the beta like google did with gmail and minecraft did for a longer period. We want to keep it that way as long as we will be polishing it. This is our first benchmark experience and we lear each day new things. The hardest thing is retrieving system info, since every component for each vendor needs to be handled separately. That's why there are still incorrect infos about clocks SLI etc.
Hello Nicolas. We don't do it since we would need a reference to compare. ANd since it's a visual benchmark then it's really hard. For example when you calculate PI, you can store the actual PI value on a disk and make a check with computed one. Then you can see if there's a computational error or not. In our case we would need to compare frame by frame with reference frames. I's quite hard to do .
Will do...Update the driver and i can test with a 448 ti later also. * updated drivers...same about So did anyone try to run in 32bit? or it cant be done? I think its the only way to have msi osd.
Try again on a fresh install of windows on spare drive, I'm not making it up. I have turned up my security on Comodo, that may be part of why I caught them?? I doubt once installed that it will remove itself from the trusted installers when you uninstall, so it may not try and make those entries again when you install again. I will post screenshots of what happens but I'm away for for about a week and dont have time to get the proof until after. I didnt let it make any of those entries in my registry so it should be virgin for those entries. If not, I'll restore a backup to another drive and get the screenshots from that when installing. Please do give me a nudge after next Thursday if I havent reported back, I dont take it lightly what they are doing. Right now I'm packing my car, then I'm off, Merry Christmas everyone
