Nvidia GeForce RTX 5090 Founders Edition review: Blackwell commences its reign with a few stumbles

More performance, power, and features — plus some driver teething pains.

Nvidia GeForce RTX 5090 Founders Edition card photos and unboxing
Editor's Choice
(Image: © Tom's Hardware)

Why you can trust Tom's Hardware Our expert reviewers spend hours testing and comparing products and services so you can choose the best for you. Find out more about how we test.

(Image credit: Tom's Hardware)

All our gaming tests are conducted using an Nvidia PCAT v2 device, which allows us to capture total graphics card power, GPU clocks, GPU temperatures, and some other data as we run each gaming benchmark. We have separate 1080p, 1440p, and 4K results for each area.

Despite the 575W TGP (Total Graphics Power, the power used by the entire graphics card) rating, the RTX 5090 Founders Edition doesn't often come anywhere near that limit. Technically, a few games did exceed (slightly) 575W, but most were far below that mark. Check the table at the bottom of this page for more details.

Clock speeds among the different GPUs and architectures aren't particularly important, but it's interesting to see where things land. The 5090 clocks lower at 4K, where power use is higher and so it has to keep things in check. At 1080p, though, it ends up with higher clocks than the 4090.

Like the clock speeds, comparing GPU temperatures without considering other aspects of the cards doesn't make much sense. One card could run its fans at higher RPMs, generating more noise while being "cooler." So these graphs should be used alongside the noise and performance results.

Considering the reduction in size combined with the increased power draw, it's impressive that the 5090 Founders Edition does this well. It doesn't beat the 4090 on thermals, but it's not too far behind. The problem is, this chart uses the geometric mean of all 22 gaming tests, and that can hide potential concerns.

Our benchmarks only run for a limited amount of time, and it can take a bit for a graphics card to fully heat up. Combine that with a bunch of games that run at much lower power levels and the dual-slot cooler may seem perfectly adequate. So we took a closer look at thermals using a worst-case scenario: Cyberpunk 2077.

If you look below, you'll see average power use at 4K with RT-ultra settings was 572W. Several other games hit similar levels (Avatar, Control, and A Plague Tale: Requiem — with Star Wars Outlaws and Warhammer 40,000: Space Marine 2 not too far behind). Leaving the RTX 5090 Founders Edition running Cyberpunk 2077 for over 15 minutes, temperatures eventually stablized at ~80C on the GPU core but 96C on the GDDR7 memory.

That's within spec, but it makes for a very hot to the touch graphics card... and the game crashed after about 20 minutes. That's not good, at all. Maybe the crash was just "bad luck," but the temperature was pretty much as expected. Here's a shot of A Quiet Place: The Road Ahead chugging along at 560W of power draw on the RTX 5090 FE. After running for 20 minutes or so, the GPU core temperature is at 81C, the GDDR7 at 96C, and the exterior of the card has hot spots of 75C.

Basically, any game that can make the GPU draw over 525W is going to end up with the card running pretty hot. I'm not sure it's quite as bad as the RTX 3080 Ti Founders Edition, but I definitely need to let the card cool off a bit before powering down and swapping if it's been running an intense workload.

(Image credit: Tom's Hardware)

We check noise levels using an SPL (sound pressure level) meter placed 10cm from the card, with the mic aimed right at the center of one fan: the center fan if there are three fans, or the right fan for two fans. This helps minimize the impact of other noise sources like the fans on the CPU cooler. The noise floor of our test environment and equipment is around 31–32 dB(A).

Our new noise testing uses Cyberpunk 2077 to hit close to maximum power use, which should also give us worst-case noise levels. With the RTX 5090 Founders Edition, after about 15 minutes we measured 50.0 dB(A) from the graphics card. Given the thermal issues noted above, the fans should perhaps spin even faster to keep thermals in check. Alternatively, the old triple-slot cooler from the 4090 FE might have been able to cope better.

[Charts to come, sorry! Still testing...]

Here's the full table of testing results, with FPS/$ calculated using MSRP for both the 5090 and 4090 cards. (Neither is available at MSRP right now.) Note again that most of the power and temperature data is for more limited duration testing. We have to reload and restart the benchmark sequences in most games, which gives the cards time to cool down in between runs and thus doesn't provide a full picture of what thermals might look like during an extended gaming session.

Jarred Walton

Jarred Walton is a senior editor at Tom's Hardware focusing on everything GPU. He has been working as a tech journalist since 2004, writing for AnandTech, Maximum PC, and PC Gamer. From the first S3 Virge '3D decelerators' to today's GPUs, Jarred keeps up with all the latest graphics trends and is the one to ask about game performance.

  • Crazyy8
    Quick look, raster performance seems a bit underwhelming. Weird that the RTX 5090 can be slower than the 4090(in niche cases). Wasn't going to buy the 5090 anyway, too expensive for a plebe like me. Looking forward to DLSS 4 and how amazing(or not)it'll be.
    Reply
  • Gururu
    Amazing but expected as far as I am concerned. I don't think there is a lot of need to compare against anything including AMDs new cards. It was brilliant to pair with the 13900 with crazy interesting results. Will read a few more times to glean more details. (y)
    Reply
  • Admin said:
    We also tested the RTX 5090 on our old 13900K test bed, with some at times interesting results. Some games still seem to run faster on Raptor Lake, though overall the 9800X3D delivers higher performance. The margins are of course quite small at 4K ultra.

    For me, as a 13900K owner, that's a consolation :cool:
    Reply
  • Gaidax
    Okay, that IS a sick cooler that actually manages to do the job.

    I bet aftermarket 4 slot monstrocities will do better, but for 2 slots 600w that's insane.
    Reply
  • m3city
    Products like this should receive 3stars max. Great performance but at what cost? Is it the right direction that power draw increases at each iteration? Is it worth to chase max perf each time? For me it would be perfect if 5000 series stayed at same TDP as previous ones - meaning better design, better gpu - with understandably lower increase of perf compared to 4000. And then, 6000 series to have even reversed direction: higher perf with drop of TDP.

    And secondary, how come 500W gpu can be air cooled, but nerds on forums will claim you absolutely NEED water cooling for 125W ryzen, cause "highend"?. Yeah, i know 125W means more actual draw.
    Reply
  • redgarl
    Okay Jarred... you are shilling at this point.

    4.5 / 5 for a 2000$ GPU that barely get 27% more performances?

    While consuming 100W more than a 4090?

    And offering the same cost per frame value as a 4090 from 2 years ago?

    Flagship or not, this is horrible.

    Not to mention the worst uplift from an Nvidia GPU ever achieved... 27%...

    https://i.ibb.co/4fks6Gt/reality.jpg
    Reply
  • redgarl
    Did you bench into an open bench or a PC case? I am asking because there is some major concerns of overheating because the CPU coolers is choked by the 575W of heat dissipation inside a closed PC Case. If you have an HSF for your CPU, then you are screwed.

    https://uploads.disquscdn.com/images/7a5bf4d586b20ffe0aa6281c57d419012a32cbdabd43b3e8050d2aa9a00d6cc1.png
    Reply
  • oofdragon
    20% better at 4K and 15% better at 2K, all that having 30% more cores and etc............... got it. Oh boy the 5080 and 5070 are sure going to disappoint a lot of people.

    The good news is the RX 9070 will bring 4070 Ti Super performance to the table around $500 including the VRAM, ray tracing and dlss image quality. AMD will prolly counter the multi frame gen nonsense with something like the LSFG 3.0 is doing and smart buyers will finally have a good GPU to replace their 3080 or 6700 XT.
    Reply
  • vanadiel007
    They should have given it code name sasquatch, because that's the chance you will be seeing these sell for $2,000 in the coming months.
    More like $3,000 and a lot of luck needed to find one in stock.

    I pass on it.
    Reply
  • YSCCC
    a real space heater inside the case, and extremely expensive with not great raw performance increase... sounds like built for those with more money than brain or logic..
    Reply