• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

TechPowerUp GPU-Z v2.37.0 Released

btarunr

Editor & Senior Moderator
Staff member
Joined
Oct 9, 2007
Messages
47,684 (7.42/day)
Location
Dublin, Ireland
System Name RBMK-1000
Processor AMD Ryzen 7 5700G
Motherboard Gigabyte B550 AORUS Elite V2
Cooling DeepCool Gammax L240 V2
Memory 2x 16GB DDR4-3200
Video Card(s) Galax RTX 4070 Ti EX
Storage Samsung 990 1TB
Display(s) BenQ 1440p 60 Hz 27-inch
Case Corsair Carbide 100R
Audio Device(s) ASUS SupremeFX S1220A
Power Supply Cooler Master MWE Gold 650W
Mouse ASUS ROG Strix Impact
Keyboard Gamdias Hermes E2
Software Windows 11 Pro
TechPowerUp today released the latest version of TechPowerUp GPU-Z, the popular video sub-system information, monitoring, and diagnostic utility, for PC enthusiasts and gamers. The latest version, 2.37.0 adds support for new GPUs, and improves the user interface in many ways. To begin with, supported is added for the upcoming GeForce RTX 3060, and recently launched GPUs, such as the RTX 3080 Mobile, RTX 3070 Mobile, RTX 3060 Mobile, compute accelerators such as the A6000, A40, A100-SXM4-40 GB, Drive PX2, P106M, and professional-visualization cards such as the Quadro K510M, and Tesla K20M modded to Quadro K6000. On the AMD side, we have preliminary support added for the upcoming Radeon RX 6700 series and RX 6600 series.

Among the many feature updates, fixes, and improvements we've made with GPU-Z 2.37.0 are detection for memory vendors on graphics cards based on RDNA and RDNA2 graphics architectures. A workaround has been added for a hardware bug in NVIDIA Ampere's PCIe controller. Filtering has been added for EVGA iCX sensor misreading. Fake GPU detection for "GT218" silicon has been improved. Lastly, we've brushed up our Russian translation. Grab GPU-Z from the link below.

DOWNLOAD: TechPowerUp GPU-Z 2.37.0



The change-log follows.
  • Added memory vendor detection on Navi 1x and Navi 2x
  • Added workaround for NVIDIA Ampere PCIe hardware bug
  • Added filter to avoid misreadings on EVGA iCX
  • Fixed fake detection for some GT218 variants
  • Improved Russian translation
  • Added preliminary support for Radeon RX 6700 and RX 6600 Series
  • Added support for NVIDIA GeForce RTX 3060, RTX 3080 Mobile, RTX 3070 Mobile, RTX 3060 Mobile, RTX A6000, A40, A100-SXM4-40GB, Drive PX2, P106M, Quadro K510M, modded Quadro K6000
  • Added support for additional variants of NVIDIA GTX 1650 Max-Q, Quadro P1000, GTX 650, GT 430
  • Added support for AMD Cezanne, Radeon Pro V520, R9 290X ES, Barco MXRT 2600
  • Added support for Intel Comet Lake Graphics (Celeron 5205U and i7-10810U)
  • Added vendor detection for Yeston

View at TechPowerUp Main Site
 
Thanks for the hotspot and memory sensors being included @W1zzard :)

Question: are they supposed to be the same temperature? They seem to be within 0.1 from each other, at most.
 
Thanks for the hotspot and memory sensors being included @W1zzard :)

Question: are they supposed to be the same temperature? They seem to be within 0.1 from each other, at most.
Hmm maybe I should have tested this on older cards, you're still using the RTX 2070 from your System Specs?
 
Hmm maybe I should have tested this on older cards, you're still using the RTX 2070 from your System Specs?
Yep, thats the one.
 
Tested on a mobile GTX 1050 4GB, 8 degree deviation between hotspot and GPU temperature at idle, somewhat concerning, but I suppose it is working as intended.
 
QwCqVqP.png


Oof, I guess this is what happens when you have idle fan stop lol
 
Hello W1zzard

Is there a reason why GPUZ reads VDDCI (mV) - the I/O bus voltage, instead of mvdd of the actual gddr6 on navi1?

For ex. the micron gddr6 on nav1 runs at 1.35V (mvdd) - this is actual gddr6 chip voltage.

GPUZ instead shows VDDCI (mV) - the I/O bus voltage - which runs at 850mv as "Memory Voltage"
 
Tested on a mobile GTX 1050 4GB, 8 degree deviation between hotspot and GPU temperature at idle, somewhat concerning, but I suppose it is working as intended.
An 8 degrees delta is nothing to be concern about, as long as it doesn't widen substantially under load.
The 3090FE for example has an up to 20 degrees delta, and that is consider in spec.

Generally large dies like the GA102 has higher delta, because it is harder to maintain an even mounting pressure across a larger surface.
My 2080ti with a Bykski waterblock has around 14C delta, about the same as my friend's Seahawk EK that has a factory installed block.
 
Tested on a mobile GTX 1050 4GB, 8 degree deviation between hotspot and GPU temperature at idle, somewhat concerning, but I suppose it is working as intended.
QwCqVqP.png


Oof, I guess this is what happens when you have idle fan stop lol
It's a hotspot, so it will be always above the GPU Temperature reading. Beside, 50-60C is nothing to these chips.
 
Really cool seeing even Turing got GPU hotspot temp and GDDR6 junction temp. My niece's 2080Ti now reads in those temp as well. At least it shows I did a OK job for the AIO installation for the 2080Ti.

2080ti_junctioj_temp.jpg
 
Back
Top