Guest GoriLLakoS Posted January 16, 2008 Posted January 16, 2008 (edited) Aloha guys, Is there on your future plans to add a new category with Max frequency on GPUs like the CPU-Z hall of fame? Now with the release of the GPU-Z , it should be easier and more accurate than before.. Regards Bill ________ Blowjob swallow Edited April 2, 2011 by GoriLLakoS Quote
knopflerbruce Posted January 16, 2008 Posted January 16, 2008 Aloha guys, Is there on your future plans to add a new category with Max frequency on GPUs like the CPU-Z hall of fame? Now with the release of the GPU-Z , it should be easier and more accurate than before.. Regards Bill I don't OC high-end video cards, but this sounds a great idea to my ears anyway:D But are you talking about CORE frequency, MEMORY frequency, or the total sum of those? Quote
Monstru Posted January 16, 2008 Posted January 16, 2008 I think that CORE and MEM as separate categories would be super I would also like to propose a CPU-Z Memory ranking. DDR, DDR2, DDR3 categories. Quote
MaSell Posted January 16, 2008 Posted January 16, 2008 Overclocking gpus for 2d? Stupid idea for me Quote
giorgos th. Posted January 18, 2008 Posted January 18, 2008 I think that CORE and MEM as separate categories would be super I would also like to propose a CPU-Z Memory ranking. DDR, DDR2, DDR3 categories. very very nice M8.. Quote
K404 Posted April 27, 2008 Posted April 27, 2008 What about having 3DM01 nature or AM3 as a minimum requirement for a GPU clock speed "verification" SPi1M and CPU-Z are pointless for any indication of stability, but are so popular for people breaking GHz barriers etc Im working on some GPU stuff at the mo and hope to get either of those benches to complete to show off some MHz. Dont see why it should be less of an achivement than using SPi 1M to show "OMG I broke 5.7GHz on an E84" or whatever Quote
TaPaKaH Posted April 27, 2008 Posted April 27, 2008 Suicide runs (no stability tests) 'd be pointless as most of the modern cards have separate 2D and 3D clocks - so you can set 2000/2000 for 3D mode on whatever card you own - run GPUz in 2D mode - GPUz 'll read 2000/2000 - you have #1 spot stupid if we're talking about "benchable" rankings - best way to combine GPU/MEM clocks without separating the categories is a bench that is 100% GFX dependant Sadly, 01 nature isn't such anymore - with 9800GX2 cards you have a heavy CPU bottleneck there Quote
K404 Posted April 27, 2008 Posted April 27, 2008 so you'd suggest two rankings?stable & unstable:) Nope For me personally- I dont see the point of CPU-Z benching because I dont see the point of "having" MHz that cant do anything, so a GPU-Z picture of MHz wouldnt interest me at all, only a bench or a test ran at xxx MHz. Also- seeing as its purely a function of MHz, it loses a bit more of the point because it doesnt recognise the differences between cards or clock efficiency. Just my opinion though. Quote
TaPaKaH Posted April 28, 2008 Posted April 28, 2008 but we need similar conditions to all the cards and some just won't run that high settings Quote
Eldonko Posted April 28, 2008 Posted April 28, 2008 I would rather see CPUZ for mem (ddr1/ddr2/ddr3) than GPU. Quote
TheKarmakazi Posted May 27, 2008 Posted May 27, 2008 I would rather see CPUZ for mem (ddr1/ddr2/ddr3) than GPU. Yeah CPU-Z for DDR/DDR2/DDR3 would be an AWESOME addition to hwbot. +100000000 to that idea I agree that the GPU-Z HOF is a bad idea, too many variables to standardize it. Quote
knopflerbruce Posted May 27, 2008 Posted May 27, 2008 A ranking of DDR1/DDR2/DDR3 frequencies sounds great! Tweaking memory is a skill, just as getting the highest CPUZ validation. GPUZ valid's would also be great. It's nearly impossible to get an "ever-lasting" hardwawre record for GPUs atm. Old records will be beaten once faster CPUs are released. It's not fun to get a really good result, and see it get owned because AMD/Intel released a faster CPU. Quote
Massman Posted May 28, 2008 Posted May 28, 2008 Pure frequencies? So using 6-12-12-48 as timings on ddr2? Quote
Monstru Posted May 28, 2008 Posted May 28, 2008 No pure frequencies, but with timings...best for 3-3-3-x, best for 4-4-4-x, best for 5-5-5-x Quote
TheKarmakazi Posted May 28, 2008 Posted May 28, 2008 maybe some basic rules could be established for this category also... i.e. DDR1 no looser timings than 4-4-4-x.... DDR2 no looser than 5-5-5-x.... DDR3 no looser than 7-7-7-x Or something along those lines. Im not an expert memory tweaker so the fine points should be fleshed out by those who know whats up. Quote
knopflerbruce Posted May 29, 2008 Posted May 29, 2008 maybe some basic rules could be established for this category also... i.e. DDR1 no looser timings than 4-4-4-x.... DDR2 no looser than 5-5-5-x.... DDR3 no looser than 7-7-7-x Or something along those lines. Im not an expert memory tweaker so the fine points should be fleshed out by those who know whats up. CAS is the best way to sort the scores. IMO. Quote
George_o/c Posted July 31, 2008 Posted July 31, 2008 So what happened with that ? No CPU-Z DDR/DDR2/DDR3 rankings, no GPU-Z higest freq ? Quote
K404 Posted August 4, 2008 Posted August 4, 2008 Id like to add 2 1/2 reasons why GPU-Z HOF is not a good idea 1/ screeny of freq is pointless- nVidia and ATI dont care about raw MHz the same way Intel did. theres just no point to the bench except as another way to award silverware points 2/ It will give another skewed value to certain hardware 3/ This is a legitimate score, yet 100% BS at the same time: Yup...thats a 6600GT at 1GHz core, 3DM03 stable, with a TEC. These are my opinions, but its how I feel about the situation. IMHO raw MHz are meaningless Quote
knopflerbruce Posted August 4, 2008 Posted August 4, 2008 Id like to add 2 1/2 reasons why GPU-Z HOF is not a good idea 1/ screeny of freq is pointless- nVidia and ATI dont care about raw MHz the same way Intel did. theres just no point to the bench except as another way to award silverware points 2/ It will give another skewed value to certain hardware 3/ This is a legitimate score, yet 100% BS at the same time: Yup...thats a 6600GT at 1GHz core, 3DM03 stable, with a TEC. These are my opinions, but its how I feel about the situation. IMHO raw MHz are meaningless The problem with today's GPU benchmarks is that they're all very CPU dependent. When benching a CPU you can get a nice pi score by using an FX5200 PCI card:D But benching 3dmark05 with a celeron isn't much fun, is it? I have tons of AGP cards here. Someday I'll buy an ASRock board and an X6800, just to get a lot of points - which will be quite easy (I think) when I fight agains Bartons etc. It would make things alot more interesting if there was a bench that wouldn't benefit from greater CPU speeds - max frequencies don't. Quote
K404 Posted August 4, 2008 Posted August 4, 2008 I agree completely- a 100% GPU bench would be very nice, but theres a strong chance if there was one it would be along the same lines as Sandra- just a GFLOP chart.... Quote
George_o/c Posted August 16, 2008 Posted August 16, 2008 I have tons of AGP cards here. Someday I'll buy an ASRock board and an X6800, just to get a lot of points - which will be quite easy (I think) when I fight agains Bartons etc. Will do same thing too Quote
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.