edgeofsanitySame problem with power draw on my card. Right around 460w board power is the maximum before power throttling starts. Running an undervolt at 1075mV for PR runs right now to prevent power throttling but its hurting my score. Hit a wall at 14800, cannot go further without getting the portable ac unit out or opening a window next to the PC for sub ambient. No coil whine issues thankfully, card is very quiet. This feels A LOT like the bugged 500w bios on my FTW3. I run the XC3 bios on that card as a workaround. I'm using a 1200w P2 PSU with custom 16awg cables that I built.
Stickboy46I don't see 500+ on PR or anything in 3dMark. Moment I fire up Furmark and run the short little 4k test, it pulls 500-520 consistently. I'm not seeing any downclocking though .. it's a rock solid clock throughout PR or even when gaming. https://www.3dmark.com/pr/653446
Basgolferagree with stickboy. Run Furmark and it'll use 500-520 watts. 3d mark benchesalmost Always sub 500 watts
kris1111Basgolferagree with stickboy. Run Furmark and it'll use 500-520 watts. 3d mark benchesalmost Always sub 500 wattsThe problem here is not a 520W power draw in 3DMark but constantly hitting PL at 450W on the LN2 bios.
kris1111Just now - 82,7W
dante`afksounds like you guys got duped by EVGA again
Cadillac94pimpinkris1111Just now - 82,7W Now, i really don't know if this has anything to do with it, I'm just relaying what I have read here and elsewhere. This is from Sajin's thread mentioned above when someone asked what the PCIE slot power draw was. You would need to read into it and draw your own conclusion and maybe reach out to support to see if they can help. "Pulled like 56w at stock clocks on the LN2 vbios"
Cool GTXNot sure LN2 BIOS helps on standard cooling (above 0 C)
TheHyperMatrixCool GTXNot sure LN2 BIOS helps on standard cooling (above 0 C)The “LN2” bios isn’t really an LN2 bios. It’s not the same bios they used when they had Steve and Jay actually overclocking them on LN2. It’s 520W on a card with a 360mm AIO radiator. Strix is 480W on fully air cooled card.
MagreciteAgain, the problem is that we're not getting anywhere near the 520W limit. It peaks around 450W and doesn't come close to the 120% power target allowed by the LN2 BIOS.
TheHyperMatrixMagreciteAgain, the problem is that we're not getting anywhere near the 520W limit. It peaks around 450W and doesn't come close to the 120% power target allowed by the LN2 BIOS.I know. He seemed to be saying the reason people aren’t able to get the full 520W is that it you need to be below 0C to take advantage of the LN2 bios. I was explaining to him that the LN2 is only LN2 in name and has a power limit that is barely any higher draw than a fully air cooled card.
ChamidorixMake sure you people are running timespy extreme to test power draw. It is unlikely, but possible, that a your kingpin won't ever draw 520W on port royal at stock voltage (1.1 or lower), it is only a 1440p bench. Furmark artificially annihilates memory and cache power (fbvdd and msvddd) which, if you care to destroy your sanity keeping up to date on the FTW3 XOC thread, does not seem to be tied with pci-e slot power like the core/nvvdd does. So getting higher power in Furmark isn't really practically relevant, while timespy extreme (4k dx12 gaming rasterization) is. All these apologists always ignore the simple fact that a Asus Strix or MSI trio will EASILY coast to and cap at 520W (after flashing kpe bios) on Timespy extreme even at sub 1.1V. When shunting to remove power limits we know 3090s tend to use around 600W+ on 1.1V on TSE at ~40-50c (golden bins excluded). As someone with a shinny new kpe that finally just arrived this morning, I'm now personally invested in finding out if we can get some confirmation that someone has a kingpin that won't hit 520W on the 520W bios in Timespy Extreme, because it absolutely should. Otherwise it is staying in the box for a free return.
USMC1419ChamidorixMake sure you people are running timespy extreme to test power draw. It is unlikely, but possible, that a your kingpin won't ever draw 520W on port royal at stock voltage (1.1 or lower), it is only a 1440p bench. Furmark artificially annihilates memory and cache power (fbvdd and msvddd) which, if you care to destroy your sanity keeping up to date on the FTW3 XOC thread, does not seem to be tied with pci-e slot power like the core/nvvdd does. So getting higher power in Furmark isn't really practically relevant, while timespy extreme (4k dx12 gaming rasterization) is. All these apologists always ignore the simple fact that a Asus Strix or MSI trio will EASILY coast to and cap at 520W (after flashing kpe bios) on Timespy extreme even at sub 1.1V. When shunting to remove power limits we know 3090s tend to use around 600W+ on 1.1V on TSE at ~40-50c (golden bins excluded). As someone with a shinny new kpe that finally just arrived this morning, I'm now personally invested in finding out if we can get some confirmation that someone has a kingpin that won't hit 520W on the 520W bios in Timespy Extreme, because it absolutely should. Otherwise it is staying in the box for a free return. As I've said above, I'm following all of this closely. I hope to hear more about this because I'll be doing the same with my KP that is coming Monday. I'll just stick with my FTW3 Hybrid if this is the case.
MagreciteJust spoke with EVGA Support over the phone, card is unfortunately going to have to be RMA'd (both for the insanely loud buzz and the lack of power draw). Sad day.