Friday, January 24th 2025

GeForce RTX 5090 Power Excursions Tested: Can Spike to 901W Under 1ms

Igor's Lab conducted an in-depth analysis of the power management system of the new NVIDIA GeForce RTX 5090 graphics card, including the way the card draws peak power within the tolerances of the ATX 3.1 specification. This analysis should prove particularly useful for those still on older ATX v2.51 PSUs, and plan to use the included power adapter that converts four 8-pin PCIe power connectors to a 12V2x6. Igor's peak power analysis shows that the RTX 5090 is capable of excursions as high as 627.5 W for 10 ms to 20 ms durations; as high as 738.2 W in 5 ms to 10 ms durations, as high as 823.6 W in the 1 ms to 5 ms category, and as high as 901.1 W in spikes under 1 ms in duration.

An excursion is a brief increase in power draw beyond the continuous power delivery limit of the connector (600 W in case of the RTX 5090's single 12V2x6 input and adapter that converts four 150 W 8-pin PCIe inputs). There is nothing particularly alarming about these numbers, and the excursions part of Igor's analysis fall within the specification of the ATX 3.1 standard, which calls for excursions of up to 200% (1200 W) up to 1 ms. Any PSU meeting the ATX 3.1 specs that even has a continuous power output of less than 1200 W will be capable of handling these spikes. It's only with the much older generations of PSUs, such as ATX v2.51 (mid-2010s) that excursions can trigger OCP. Find other great insights in the Igor's Lab review linked below.
Source: Igor's Lab
Add your own comment

43 Comments on GeForce RTX 5090 Power Excursions Tested: Can Spike to 901W Under 1ms

#26
phanbuey
It's physically uncomfortable gaming next to the rig at full blast. Like the right side of my body being 5 degrees warmer than the left is a bit immersion breaking.
Posted on Reply
#27
Gasaraki
You guys think a 1ms spike will cause the cable to spontaneously combust? Get real.
LabRat 891If the cable/conductors aren't losing heat faster than they're gaining, they'll eventually runaway under sustained load. -Regardless of peaks.
Not to mention, physical migration/movement from thermal expansion/contraction...
The spec is designed for 600W continuous load. No cable is melting. If the spec is designed for 600W continuous, that means there's probably a 20% buffer built in, so the cable can support 720W.
Posted on Reply
#28
gridracedriver
But do you remember the RX480 that melted the mobos for peak consumption? good times of terrible misinformation against AMD, Nvidia is the Intel of the 2000s with the press in favor and everyone silent.
Posted on Reply
#29
Ruru
S.T.A.R.S.
So basically you need a 1.2kW PSU to be on the absolutely safe side. I still remember when those were mostly for HEDT rigs with SLI/CF or something similar.
gridracedriverBut do you remember the RX480 that melted the mobos for peak consumption? good times of terrible misinformation against AMD, Nvidia is the Intel of the 2000s with the press in favor and everyone silent.
Remember that issue but did it even cause anything else than negative PR? I guess that PCIe has some headroom over its 75W spec.
Posted on Reply
#30
InVasMani
Does this GPU identify as a microwave yes or no Jensen answer the question.
Posted on Reply
#31
claster17
Guys...

Stop panicking. This is all well within ATX 3.0 spec.

Posted on Reply
#32
evernessince
JustBenchingA good 1k or 1.2k psu should be able to without triggering ocp without the atx specification required.
1,000 watts is the bare minimum with an efficient CPu I believe. TechYesCity was getting full system shutdowns with a 850w unit and a 7800X3D. With an Intel CPU 1.2K is minimum.
GasarakiYou guys think a 1ms spike will cause the cable to spontaneously combust? Get real.


The spec is designed for 600W continuous load. No cable is melting. If the spec is designed for 600W continuous, that means there's probably a 20% buffer built in, so the cable can support 720W.
Max rated wattage is 684w: en.wikipedia.org/wiki/16-pin_12VHPWR_connector

That said the cable is designed to carry 600w. Anything above that is iffy depending on manufacturing variance. Ironically the 5090 astral does exceed 600w...

Look at the margin compared to the 8 pin, world of difference.
Posted on Reply
#33
JustBenching
evernessince1,000 watts is the bare minimum with an efficient CPu I believe. TechYesCity was getting full system shutdowns with a 850w unit and a 7800X3D. With an Intel CPU 1.2K is minimum.
Lol, ok man. My 9800x 3d draws more power in gaming than my 12900k, but sure, you do you.
Posted on Reply
#34
Gasaraki
motov8600w is too much for a room without air conditioning. This is an oven. I don't want to know how someone doesn't have air conditioning and has 28'C in the room in the holidays playing on a 200-250 watt graphics card like me.
The card is not putting out 600W of heat. This is not a heater. The energy is used to do work. A 600W resistive heater doesn't even put out 600W of heat, it has losses.
Posted on Reply
#35
londiste
GasarakiThe card is not putting out 600W of heat. This is not a heater. The energy is used to do work. A 600W resistive heater doesn't even put out 600W of heat, it has losses.
With electronics, the energy used to do work is effectively theoretical. For practical purposes all the consumed energy goes into heat output.
Posted on Reply
#36
chrcoluk
LabRat 891Let the 12V-2x6 burninating, begin! (again)
12V-3x4 already off the drawing board and ready to go.
Posted on Reply
#37
Gasaraki
evernessince1,000 watts is the bare minimum with an efficient CPu I believe. TechYesCity was getting full system shutdowns with a 850w unit and a 7800X3D. With an Intel CPU 1.2K is minimum.



Max rated wattage is 684w: en.wikipedia.org/wiki/16-pin_12VHPWR_connector

That said the cable is designed to carry 600w. Anything above that is iffy depending on manufacturing variance. Ironically the 5090 astral does exceed 600w...

Look at the margin compared to the 8 pin, world of difference.
Surprised it only has a 1.1 safety factor instead of 1.2 like in home electrical wires.
Posted on Reply
#38
DemonicRyzen666
claster17Guys...

Stop panicking. This is all well within ATX 3.0 spec.

No technically it's not it's 901 watts, it's 1 watt over it. lol
Posted on Reply
#39
trsttte
TimbalooThis thread is about spikes though, and these really don’t matter from a thermal point of view.
They do if they happen frequently enough, but it's off topic yes
Gasarakithere's probably a 20% buffer built in
Not with this connector there isn't
Posted on Reply
#40
katzi
The RMA's are going to be *Wild*

Posted on Reply
#41
jnv11
Lots of pre-ATX 3.0 power supply units are likely to prematurely fail due to not being designed to handle these high magnitude power spikes if they are connected to an RTX 5090.

How will power supply reviewers have to modify their methodologies to account for new video cards which routinely generate ridiculously large power spikes like RTX 5090 video cards are prone to doing?

When these power spikes eventually destroy the power supply units that were not designed to ATX 3.x standards, how would those power supply units fail, and what effects should we expect when they do fail? Will they just be unable to start and be dead, or would they spark, smoke, or catch fire? I am not an expert in how power supply units fail.
Posted on Reply
#42
JustBenching
jnv11Lots of pre-ATX 3.0 power supply units are likely to prematurely fail due to not being designed to handle these high magnitude power spikes if they are connected to an RTX 5090.

How will power supply reviewers have to modify their methodologies to account for new video cards which routinely generate ridiculously large power spikes like RTX 5090 video cards are prone to doing?

When these power spikes eventually destroy the power supply units that were not designed to ATX 3.x standards, how would those power supply units fail, and what effects should we expect when they do fail? Will they just be unable to start and be dead, or would they spark, smoke, or catch fire? I am not an expert in how power supply units fail.
Why would they fail? There is 0 reason for a well built PSU to fail, especially not because of power spikes. It will trip the over current protection and shut down. As long you are not running your PSU at 100% capacity all day long (so probably you should avoid an 850w PSU with a 5090) youll be fine.
Posted on Reply
Add your own comment
Feb 26th, 2025 02:10 EST change timezone

New Forum Posts

Popular Reviews

Controversial News Posts