With 2700MHz as top clock I bet you can go down to 1040~1050mV at least
Mine at 2850~2900MHz with demanding games needs 1070mV and less demanding games down to 1050mV.
Of course not all GPUs are exactly the same, even on same model as we both have.
Also dont look power maximum at non gaming...
Not all time exactly but quite a lot
Here is a 3 hours gameplay session
Limit of the board is 375W
Max value is ~110W over the limit and average value about 80W over the average board power.
Potentially all those 3 sensor can reach the specified Tjmax depending the task load, but its more like to happen on all core loads (at 150-165W) if the cooling cant keep it below that 95C threshold.
You may see also a couple or more of individual core temps also around there. Those would may be...
Most likely you didnt read the description (tooltip) of the sensor.
Enable tooltips from settings and hover your pointer over the "Power Reporting Deviation" sensor.
This sensor only has value under 100% CPU load. Any other time is useless and should be ignored.
What does the PSU 12V rail says?
Is it single or multi 12V rail?
Maybe GPU sensors are a bit off.
And to be honest the standards about 3.3, 5 and 12V rails are +/- 5%
So for 12V it’s 11.4~12.6V.
I would rather have 12.5V than 11.5V
I have these settings...
...and when I click the "X" it goes to tray
...and when I click the "-" it goes to taskbar (alternatively if you click HWiNFO on the task bar it does the same)
If you want to close it then right click tray icon and click quit
simple
RX 560 XT had less Shaders and TMUs activated compared to a RX 470.
Some of them could be flashed and unlocked to a 470 though.
https://www.techpowerup.com/gpu-specs/radeon-rx-470.c2861
https://www.techpowerup.com/gpu-specs/radeon-rx-560-xt.c3388
It’s been a while (months) since I’ve seen the double misreported values.
But occasionally I have the plus ~6% mostly on FCLK, UCLK, MCLK. Very rarely on SoC voltage too.
It must be a bug like the double ones. It’s not from BCLK for sure.
R9 5900X
Not sure if this is a mistake from Gigabyte or they do it deliberately. When PRD is <100% then the CPU is forced over its default performance. Not over its silicon health though.
Maybe they are trying to make the board look good in CPU performance comparisons?
No one can really tell.
I have a...
What are you talking about?
I have been downloading HWiNFO64 for many years now the same way
Hovering pointer over any green button opens the blue dropdown menu
Not once had any trouble (using Firefox, IE, and latest years Chrome) downloading it. The installer or the portable (win) version...
I had the R5 3600 for 3 years before I switch to 5900X
Yeah, your PRD reading during the CB-R24 is way off.
Meaning that the board is reporting back (feedback) to the CPU only half (49.7%) the true power consumption.
CPU PPT reads ~53W but you have to divide that with 0.497 (from PRD 49.7%)...
I will consider the 2 on the left to be a little more accurate. At least the one on the CPU Enhanced section is usually polled straight from CPU.
The one on the right is from board sensor and as @Martin said its unclear how ASUS is doing it.
You can find a lot of discussions below
https://www.techpowerup.com/forums/forums/intel-arc-gpus.94/
...and some reviews and benchmarks
https://www.techpowerup.com/review/?category=Graphics+Cards&manufacturer=Intel&pp=25&order=date
https://www.youtube.com/results?search_query=Intel+Arc
By default, AMD uses the "CPU Tctl/Tdie" sensor for CPU cooler temp control.
Tctl = T_control
This is what the board will use in BIOS to control CPU cooler through CPU_FAN header or CPU_OPT header, if no third party software is installed.