5080 NVVDD output power sensor value issue

M

mindsfinest

Member
Hi,

Apologies if this is in the wrong section.

Does HWInfo64 currently support the 50 series rail powers?

I am seeing a value on the GPU core NVVDD output power that seems unrealistic. My 5080 hits 901W max in a heavy workload (cyberpunk2077 pathracing). All input powers don't exceed 320w and the other NVVDD Output Power does not exceed 317w.

My Corsair RM850x is 850W and I am using the three 8 pin adapters to power the card (which I believe totals 450w)

Could this be a true transient spike of 901w, or could HWinfo be misreporting the sensor?

After resetting hwinfo, so far the max I have seen in the same field is just over 600w.

1000029007.jpg

I have seen another thread where it was mentioned that these NVVDD power value meanings are not shared by Nvidia. So I do understand if the answers to my questions are unknown: https://www.hwinfo.com/forum/threads/msi-4090-suprim-x-nvvdd-output-power-reading.8473/
 
RTX 50 series should be well supported, this is rather an issue of NVIDIA drivers reporting such spikes.
 
Thank you for such a speedy reply and for creating HWinfo64 - your work is really appreciated!

It's an interesting article from Igor. Do you think a 5080 would also hit these transient spikes in the same way as a 5090? Or do you think the Nvidia driver could be bugged? It seems odd to me that all the input power values do not exceed 320W. So unless these sensors are not "sensitive" enough to capture the transient spikes, surely the internal power cannot exceed that which is input...
 
It would seem from these two reviews that the power draw spikes should not be more than 600w:

www.igorslab.de

NVIDIA GeForce RTX 5080 Founders Edition Review - GeForce RTX 4080 Ti Super with Blackwell genes | Page 11 | igor´sLAB

Today’s article offers a comprehensive review of the GeForce RTX 5080 and builds on the launch article for the RTX 5090, which I will of course also link to below. But I’ll save myself redundant…
www.igorslab.de www.igorslab.de
www.igorslab.de

MSI GeForce RTX 5080 Vanguard SOC review - The better FE with the SUPRIM inner workings and the slightly cheaper outer skin | Page 6 | igor´sLAB

With the MSI RTX 5080 Vanguard SOC, MSI has slightly reduced the price of the rather serious-looking SUPRIM and is looking for a different target group (purely visually) for an otherwise pretty much…
www.igorslab.de www.igorslab.de
 
A fellow Reddit user shared their NVVDD for the Gainward phoenix model of the 5080. I just wanted to record it on this thread as the mods removed my thread from Reddit, presumably as they thought it was a debug thread...
1738854048605.png
 
chipping in:
5080 msi vanguard, 111% power, +375core, no v changes.
1739119385986.png
After CP2077 (spikes usually in the 1st minute of the game to 600-800, then rare spike up to -highest i noted was ~1,15kw)
Buuut, after burning up for some time in furmark:
1739119684798.png
wondering if it has something to do with the way it boosts, power draw/boost spikes in fm/occt are stable, but in cp they jump like monkey on coke.
 
Ich habe mit meiner rtx 5080 ein ähnliches Problem, jedoch toppe ich wohl alles mit 1,9kW... da stellt sich mir die frage ob es nicht vllt doch ein Software fehler ist. Hmm
 

Attachments

  • Screenshot_20250212_161243_WhatsApp.jpg
    Screenshot_20250212_161243_WhatsApp.jpg
    108.6 KB · Views: 17
Adding my data here. I also have a thermistor sitting next to the cables, only reached 45 C.
 

Attachments

  • 5090 power draw.jpg
    5090 power draw.jpg
    231.4 KB · Views: 26
LuftManu said:
Hi! After some time playing Cyberpunk

View attachment 12218


Cable was hot to the touch :(
Click to expand...
Interesting, I put together a build for a friend last night and I was getting weird readings with the RM850x (2024) using a 12vhpwr. On the verge of returning it later this weak for a seasonic. Averaging 11.85 and dropping to 11.73 on load. I know it's within spec, but just makes me uneasy after touching it (kinda hot too)
 
You must log in or register to reply here.

Similar threads

M
MSI 4090 Suprim X (NVVDD) Output Power reading?
Replies
3
Views
2K
mickey_padge
M
T
power reporting deviation %1100
Replies
5
Views
626
They not like us Wop wop
T
J
[SOLVED] BUG: VDDCR_SOC (SVI3 TFN) sensor with (maybe) wrong values
Replies
2
Views
1K
jhowts
J
Dave1001
Calculating Power Draw with HWInfo Confusion?
Replies
6
Views
12K
LeDoyen
L
A
Need Help Understanding Sensors & Diagnosing High Temperature Readings!
Replies
0
Views
2K
Adeptivity
A
Back
Top