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

RTX 3080 Ti - Curiosity - Has anyone noticed a 10W difference in power draw between pin #1 and pin #2.

Olias

New Member
Joined
Apr 15, 2023
Messages
3 (0.01/day)
The Tl;DR version of this is that I'm not having any issues other than a mystery I'd love to solve. If anyone owns one of these cards, please post your results.

For the record the card in question is a GIGABYTE RTX 3080 Ti EAGLE OC

This is a fairly new card that I purchase about 8 months ago. I put it through it's paces at the beginning by stress testing it in multiple motherboards and power supplies. Presently I have it hosted in a Z690 based MSI Carbon motherboard powered by a Seasonic SSR-850FX. CPU is an Intel I7-12700K. The tests I ran are with the system at defaults and also overclock. However, there is no difference in the 10W-12W power differential I've observed between overclocked or default settings. NVIA drivers are current.

My stress testing uses the usual suspects. i.e. Superposition, Kombustor, Furmark, 3Dmark etc.

I also utilize a standing connection to a Kill-o-Watt meter which is mounted next to my rig so that I can see it while running bench marks.

There is nothing new significant to report other than I have a consistent 10W power difference between Pin#1 and Pin#2 with Pin#1 always drawing 10W more no matter what the load is on the system. The image on the left with yellow highlight is the system at relative rest and the one at the right is being exorcized with Kombustor. It really doesn't matter what I use to stress the GPU, the difference is always between 10W-12W average.


GPU Power usage Screenshot 2023-06-07 040023.png
GPU Power usage Screenshot 2023-06-07 040006.png
RTX 3080 TI - Screenshot 2023-06-07 041312.png


Cross checked with HWiNFO64. Similar results.
Power differential Screenshot 2023-06-07 042202.png


Variations that I've already tried but produced no change:
  • Swapped power cables at the board
  • Swapped power cables at the PSU
  • Hosted the board in an older Z97 ASUS mother board and an older ASROCK motherboard.
  • Also tested using a Thermaltake 850W, Thermaltake 750W, Corsair 850W PSUs.
  • Also tested the card both in the system and using an external riser card as well as powering the GPU itself using one of the above PSU's externally.
None of these variations produced different results. I'm kind of kicking myself for not noticing this earlier because I had a ZOTAC card in my possession for two weeks but returned it because it was too large for my case by about 4mm. Had I noticed this last fall, I would have done a side by side between the two cards.
 
Top