TechPowerUp GPU-Z 0.5.8 Released

TechPowerUp today released the latest version of GPU-Z, our popular video subsystem information and diagnostic utility that provides you with accurate information about the graphics hardware installed, and lets you monitor their clock speeds, fan speeds, voltages, VRAM consumption, etc., in real-time. Version 0.5.8 introduces two new features. The first one is a render test that applies sufficient load (not stress) on the GPU to pull it out of PCI-Express link-state power-management, to ensure the Bus information is accurate. If you find the PCI-Express bus link speed or PCIe version displayed incorrectly, simply click on the "?" button next to the field to launch the load test.

The next new feature is ASIC quality, designed for NVIDIA Fermi (GF10x and GF11x GPUs) and AMD Southern Islands (HD 7800 series and above), aimed at advanced users, hardware manufacturers, and the likes. We've found the ways in which AMD and NVIDIA segregate their freshly-made GPU ASICs based on the electrical leakages the chips produce (to increase yield by allotting them in different SKUs and performance bins), and we've found ways in which ASIC quality can be quantified and displayed. Find this feature in the context menu of GPU-Z. We're working on implementing this feature on older AMD Radeon GPUs.

DOWNLOAD: TechPowerUp GPU-Z 0.5.8

The full change-log follows.

  • Added explanation about PCI-Express power savings and 3D render test to accurately measure bus config under load
  • Added function to display ASIC quality for Fermi and Southern Islands. (Located in the GPU-Z system menu)
  • Fixed crash on older ATI cards
  • Added voltage monitoring for HD 7970
  • Improved real-time clock monitoring for HD 7970
  • Fixed OpenCL detection for AMD Antilles, Whistler, Seymour, Blackcomb
  • Improved default clock reading for AMD HD 7970 and Fusion
  • Added support for AMD FirePro V7900, HD 6930, HD 7690M, HD 6410D
  • Fixed Intel Sandy Bridge IGP to be DirectX 10.1, 32 nm
  • Added support for NVIDIA Tesla C2075, GeForce GT 630M


21

United States sin0822 says:

Anyone see the GPu quality feature? You can see the quality of the silicon of your GPU if you right clock the menu and select some ASIC quality. now this could be the quality of the silicon, so maybe lower %=higher leakage or higher percent = lower leakage.

This is one of my GTX 570s which is a Giga reference card:



I tried my other same exact GPU but it got 71% i believe. Guys at TPu were comparing, let's compare b/c i know we got a ton of GPUs!

Who is going to take up my offer?

Christian Ney says:

[QUOTE=Haldi]hey, can yo do me a big favour?

If you have a HD7970 and MSI Afterburner installed try to figure out GPU Quality and post it (4 digits Hex Code, yes with the 0 at the end!) here you can find the result's of this poll here[/QUOTE]


So higher is the %, better is the gpu or ?

TaPaKaH says:

higher % => longer e-peen

United States sin0822 says:

yea i actually think lower percent could lead to higher OCes under better cooling. If it in fact has something to do with transistor leakage, then higher leakage transistors can run faster, but with higher temps. haha i have a better GPU than all those dude. one way to tell if this has anything do with with transistor leakage, if people with the same GPUs take them and compare how hot they run at different speeds and voltage. I bet maybe this is the way manufactures bin, through numbers like these.

Norway knopflerbruce says:

If enough extreme OC people submitted their scores and the GPU quality thing, it'll be easy to see if there is a connection.

Greece Stelaras says:

I knew from the start that this reference card was good ! It must be from outer space ! LOL !!! :D



United States sin0822 says:

nice yea there was one guy at TPU who had like 105% or something, but that is the second highest i have seen!

Greece Stelaras says:

I checked the comments in TPU .
I saw some screenshots with 115% and higher.

TaPaKaH says:

"Magical Percentage" sounds like a plan for a new hwbot benchmark, makes binning cards much easier :D

Finland Mean Machine says:

I don't think those values have that much to do with anything, really.

United States sin0822 says:

Stelaras said: I checked the comments in TPU .
I saw some screenshots with 115% and higher.


ahh maybe i missed that.

Well maybe it can help with binning if we find out what the percent has to d, its probably different in every uarch, as in some uarch is better to have higher than lower, who knows. We can try and find out, just try it out!

Indonesia Lucky_n00b says:

The 'ASIC Quality' feature is really interesting.
* I wonder if there's people who got more than 140% :p :p *

EDIT :
After a 10-min quick test, it looks like bigger 'asic quality' number doesn't always mean good OCability :

- My 1400Mhz-ish capable GTX 580Lightning : 88.6% ASIC Quality
- 'Banana' lightning which won't get past 1250Mhz no matter what : 92.7% ASIC Quality
- Another lightning, 945Mhz max. on air, 1350-ish LN2 result : 99.7 %
- Cherry-picked DA GTX 580 3GB, 1000Mhz on FurMark with default stock cooling : 89.1 %

More testing needed to see if there's indeed a connection..

United States sin0822 says:

great how about everyone post their cards and their qualities and most importantly how about Stock VID and max OC and required voltage. maybe we can try and see if there is some connection between any of those factors.

Germany QAI says:

Tehe my 470 only gets 52,3% :D VID is 1,0V and it gets to stable 920MHz on air at around 1,2V.

Canada Vinster says:

My 580 has 90.6% and so far OC's pretty good. I have yet to really put this card

I wish the render test had an artifact alarm... it would be a good replacement for ATI tool.

Vin

Germany QAI says:

Why not GPU-tool? Its rendertest is at least as good as the one in Ati-tool and it has better artifact detection.

Finland Mean Machine says:

The render test is just to place literally "some" load on the GPU so it can detect your PCI-E bus width properly, nothing else.

Canada Vinster says:

Mean Machine said: The render test is just to place literally "some" load on the GPU so it can detect your PCI-E bus width properly, nothing else.


Gotcha, when I saw it, it was the first thing that came to mind.

I'll check out GPU-Tool. I never heard of that one.

Vin

Christian Ney says:


Finland Mean Machine says:

My 1560MHz 580 Lightning apparently scores 75.1% in 0.5.9, while it was lower in 0.5.8.

United States sin0822 says:

Christian Ney said:


Interesting they added that in, so lower %=higher leakage, and higher percent = lower leakage, that is what I thought to begin with. Odd though, that it changed your score, maybe it is getting more accurate?

I can see this being of value when binning GPUs, if infact it is accurate.

Please log in or register to comment.