Saturday, February 22nd 2025

NVIDIA GeForce RTX 5090 Spotted with Missing ROPs, NVIDIA Confirms the Issue, Multiple Vendors Affected, RTX 5070 Ti, Too

TechPowerUp has discovered that there are NVIDIA GeForce RTX 5090 graphics cards in retail circulation that come with too few render units, which lowers performance. Zotac's GeForce RTX 5090 Solid comes with fewer ROPs than it should—168 are enabled, instead of the 176 that are part of the RTX 5090 specifications. This loss of 8 ROPs has a small, but noticeable impact on performance. During recent testing, we noticed our Zotac RTX 5090 Solid sample underperformed slightly, falling behind even the NVIDIA RTX 5090 Founders Edition card. At the time we didn't pay attention to the ROP count that TechPowerUp GPU-Z was reporting, and instead spent time looking for other reasons, like clocks, power, cooling, etc.

Two days ago, one of our readers who goes by "Wuxi Gamer," posted this thread on the TechPowerUp Forums, reporting that his retail Zotac RTX 5090 Solid was showing fewer ROPs in GPU-Z than the RTX 5090 should have. The user tried everything from driver to software re-installs, to switching between the two video BIOSes the card comes with, all to no avail. What a coincidence that we had this card in our labs already, so we then dug out our sample. Lo and behold—our sample is missing ROPs, too! GPU-Z is able to read and report these units counts, in this case through NVIDIA's NVAPI driver interface. The 8 missing ROPs constitute a 4.54% loss in the GPU's raster hardware capability, and to illustrate what this means for performance, we've run a couple of tests.

In the first test, "Elden Ring" at 4K UHD with maxed out settings and native resolution (no DLSS), you can see how the Zotac RTX 5090 Solid falls behind every other RTX 5090 we tested, including the NVIDIA Founders Edition, a de facto reference-design that establishes a performance baseline for the RTX 5090. The Zotac card is 5.6% slower than the FE, and 8.4% slower than the ASUS ROG Astral RTX 5090 OC, the fastest custom design card for this test. Officially, the Solid is clocked at 2407 MHz rated boost frequency, which matches the Founders Edition clocks—it shouldn't be significantly slower in real-life. The interesting thing is that the loss of performance is not visible when monitoring the clock frequencies, because they are as high as expected—there's just fewer units available to take care of the rendering workload.

A ROP (Raster Operations Pipeline) unit in the GPU processes pixel data, handling tasks like blending, antialiasing, render-to-texture, and writing final pixel values to the frame buffer. In contrast, a shading unit, aka "GPU core" is responsible for computing the color, lighting, and material properties of pixels or vertices during the rendering process, without directly interacting with the frame buffer, so the performance hit of the eight missing ROPs depends on how ROP-intensive a game is.
For example, in Starfield, the performance loss is much smaller, and in DOOM Eternal with ray tracing, the card actually ends up close to its expected performance levels.

We've also put the card through a quick 3DMark Time Spy Extreme graphics score run.
  • NVIDIA Founders Edition: 25439
  • Zotac Solid: 22621
  • Gigabyte Gaming OC: 26220
This should be a number that you can test easily for yourself, if you're one of the lucky RTX 5090 owners. The quickest way is definitely to just fire up GPU-Z and look at the ROP count number, it should be "176."

So far, we know only of Zotac 5090 Solid cards that are affected, none of our review samples from ASUS, Gigabyte, MSI, Palit, and NVIDIA exhibit this issue, all 5090 owners should definitely check their cards and report back.

This is an issue with quality assurance at both NVIDIA and Zotac. NVIDIA's add-in card partners (AICs) do not have the ability to configure ROP counts, either physically on the silicon, or in the video BIOS, and yet the GPU, its video BIOS, and the final product, cleared QA testing at both NVIDIA and Zotac.

We are working with Zotac to return the affected card, so they can forward it to NVIDIA for investigation. At this time Zotac was unable to provide a statement, citing the fluidity of the situation. As for possible fixes. We hope the issue is localized to a bug with the driver or the video BIOS, so NVIDIA could release a user-friendly BIOS update tool that can run from within Windows and update the BIOS of the affected cards. If, however, the ROPs were disabled at the hardware-level, then there's little that end-users or even AIC partners can do, except initiating a limited product recall for replacements or refunds. If the ROPs really are disabled through fuses, it seems unlikely that NVIDIA has a way to re-enable those units in the field, because that would potentially provide details to how such units can be reactivated on other cards and SKUs from the company.

Update 14:22 UTC:
Apparently the issue isn't specific to Zotac, HXL posted a screenshot of an MSI RTX 5090D, the China-specific variant of the RTX 5090 with nerfed compute performance, but which is supposed to have 176 ROPs. Much like the Zotac RTX 5090 Solid, it has 8 missing ROPs.

Update 16:38 UTC:
Another card has been found, this time from Manli.

Update 17:30 UTC:
ComputerBase reports that their Zotac RTX 5090 Solid sample is not affected and shows the correct ROP count of 176. This confirms that the issue isn't affecting all cards of this SKU and probably not even all cards in a batch/production run.

Update 17:36 UTC:
Just to clarify, because it has been asked a couple of times. When no driver is installed, GPU-Z will use an internal database as fallback, to show a hardcoded ROP count of 176, instead of "Unknown." This is a reasonable approximation, because all previous cards had a fixed, immutable ROP count. As soon as the driver is installed, GPU-Z will report the "live" ROP counts active on the GPU—this data is read via the NVIDIA drivers.

Update 19:18 UTC:
A card from Gigabyte is affected, too.

Update Feb 22nd, 6:00 UTC:
Palit, Inno3D and MSI found to be affected as well

Update Feb 22nd, 6:30 UTC:
NVIDIA's global PR director Ben Berraondo confirmed this issue. He told The Verge:
NVIDIAWe have identified a rare issue affecting less than 0.5% (half a percent) of GeForce RTX 5090 / 5090D and 5070 Ti GPUs which have one fewer ROP than specified. The average graphical performance impact is 4%, with no impact on AI and Compute workloads. Affected consumers can contact the board manufacturer for a replacement. The production anomaly has been corrected.
Very interesting—NVIDIA confirms that RTX 5070 Ti is affected, too.

While NVIDIA talks about "one ROP unit," this really means "8 ROPs" in our context. Many years ago, marketing decided that higher numbers = better, so they started to report the number of pixels that can be processed per unit, instead of the actual unit counts. So in this case, one hardware unit is disabled, which mean eight pixels per clock less can be processed, resulting in a loss of "8 ROPs".
Add your own comment

329 Comments on NVIDIA GeForce RTX 5090 Spotted with Missing ROPs, NVIDIA Confirms the Issue, Multiple Vendors Affected, RTX 5070 Ti, Too

#176
JustBenching
But would a defective 5090 just report as missing 8 rops? How? Doesn't nvidia have to fuse them off on a hardware level?
Posted on Reply
#177
sethmatrix7
Was considering the 50xx series before launch. After all this... I'm good. AMD will come out with a compellingly priced performance uplift or I'll stick with the 3080.

Posted on Reply
#178
Sir Beregond
Hecate91easy editors choice award.
Glad I am not the only one to notice that.

So...Nvidia is selling the AIBs defective chips, and the AIBs are taking them and not sending them back to Nvidia as defective. Scummy all around.

Nvidia really is handing AMD a prime opportunity to really make Nvidia look more than just greedy, but stupid obviously greedy. Hope AMD doesn't squander it with dumb launch prices and no volume.
GasarakiIt might not beNVIDIA's fault. NVIDIA sells them chips meant for 5080s or 5070Tis and then Zotac probably just used these defective chips in the 5090 instead.

Companies use "defective" chips for lower speced models. It is up to the AIBs to put proper firmware on their own cards.
Wouldn't make sense. The 5090 uses GB202. The 5080 and 5070 Ti use GB203. You don't just slot a GB202 die into a standard PCB for a 5080. Now down the line, it's certainly possible. There were AD102 based 4070 Ti Supers out there. But I don't think that's what's happening here. -8 ROPS isn't a 5080.
Posted on Reply
#179
mxthunder
Anyone that has a defective 5090 with the missing ROPs and had to send it in to exchange for a new one, would you? I would have a hard time letting it out of my possession. Make them cross ship at least. Hold my credit card. not getting it out of my possession until i have replacement.
Ugh. what a pickle.
Posted on Reply
#180
kapone32
Vayra86Vega was faster? What parallel universe are you in? Polaris 'caught' what exactly? Miners? Or the better half of the gaming market that had already bought a 970 about 2 years earlier? The RX480 wasn't faster or better than it, only slightly cheaper and way too late.

Vega released in august 2017, 1 year and 3 months later, and was slower than the GTX 1080, not cheaper, and guzzled about 1,4x the power to get to that performance. Vega 56 was eclipsed by the 1070ti which released 3 months later.



Seriously wondering where you get the factoids to construe this weird reality you are posting.
I was not comparing it to Nvidia but Polaris.
Posted on Reply
#181
Gooigi's Ex
BigMack70I don't know if you are an AMD fan or not, but this kind of reasoning - which relies on making excuses and hand-waving away AMD's problems - is exactly the problem.

No, AMD did not "catch them" with Polaris or Vega. Neither was competitive at the high end with Nvidia's products, neither was polished enough to compete with Nvidia's feature set and end user experience, and neither was priced aggressively enough to make Nvidia's offerings look like overpriced nonsense. Hence, no mind share and no market share for AMD.

"AMD is just as good except for {xyz}" has been required copium for AMD fans for 13 years, and it's just that - copium. The last time AMD had a product that caught Nvidia and was just as good was the HD 7970 - 13 years ago. They have failed time and time and time again to capitalize on Nvidia's missteps, and I fully expect they will do the same here with RDNA 4.
Holy shit this reeks of fanboyism it hurts lol. It seems like Nvidia fanboys wanna point out failures of AMD GPUs or how they are not catching up…but forgot that the AMD RX 6000 series existed and was by far the best GPU architecture AMD has ever made point blank. It DEFINITELY caught up to Nvidia but then a new narrative came about(RT AND DLSS) and made that so important and now AMD behind again. Rinse and repeat.

However, as saying goes, AMD never misses an opportunity to miss an opportunity so even though Nvidia is fucking up, I would be shocked if AMD manage to take advantage of it. History tends to repeat itself.
Posted on Reply
#182
Legacy-ZA
Just going to leave this here too: (Never have we been asked so much, for so little)

Posted on Reply
#183
JohH
Quality $2000 hardware. Did I say $2000? I mean $2600 or more.
Posted on Reply
#184
MxPhenom 216
ASIC Engineer
AssimilatorOMFG NVIDIA, what are you DOING?
Posted on Reply
#186
Sir Beregond
5090 melting connectors...we can bring back the Fermi era Nvidia memes:



As for the -8 ROPs..."Nvidia - The way the customer is meant to be played"
Posted on Reply
#187
theouto
Gambling is a feature is the 5000 series, this is all intended behaviour.
Posted on Reply
#188
LabRat 891
mxthunderAnyone that has a defective 5090 with the missing ROPs and had to send it in to exchange for a new one, would you? I would have a hard time letting it out of my possession. Make them cross ship at least. Hold my credit card. not getting it out of my possession until i have replacement.
Ugh. what a pickle.
I'd be tempted to keep it. These days, I could see a Class Action stipulate that those already 'made whole' need not apply.
Posted on Reply
#189
BigMack70
Gooigi's ExHoly shit this reeks of fanboyism it hurts lol. It seems like Nvidia fanboys wanna point out failures of AMD GPUs or how they are not catching up…but forgot that the AMD RX 6000 series existed and was by far the best GPU architecture AMD has ever made point blank. It DEFINITELY caught up to Nvidia but then a new narrative came about(RT AND DLSS) and made that so important and now AMD behind again.
As I said..
BigMack70"AMD is just as good except for {xyz}" has been required copium for AMD fans for 13 years, and it's just that - copium. The last time AMD had a product that caught Nvidia and was just as good was the HD 7970 - 13 years ago. They have failed time and time and time again to capitalize on Nvidia's missteps, and I fully expect they will do the same here with RDNA 4.
Last time AMD matched or beat Nvidia without the need for hilarious amounts of copium was HD 7970. No, the 6000 series was not as good as Ampere.
Posted on Reply
#190
sbacc
I see it now, the 5090 is revolutionary, it is the first "quantum" GPU, it has 8 ROP that might be here or not all at the same time, this is a Schrödinger's cat level of genius computing. Until you open the "GPUz box" you don't know in what state it is.
Posted on Reply
#191
Bomby569
nvidia drivers are also going to shit. Not sure what is going on, this seems like Intel before the crash, they just don't have competition
Posted on Reply
#192
dyonoctis
mb194dcThe 5 series is indeed cursed ? Just like the old 5900 ultra series woes ...
A new architecutral paradigm coming with the RTX 8090 ultra ? (very unlikely, but one can hope. AMD already couldn't deliver a RX 9700 that would have emulated the 9700 pro)
Posted on Reply
#193
R0H1T
BigMack70Last time AMD matched or beat Nvidia without the need for hilarious amounts of copium was HD 7970. No, the 6000 series was not as good as Ampere.
And your point is? Nvidia fanboys would pay 2x more money for 1.2x more performance in their "favorite" benches :rolleyes:
Bomby569this seems like Intel before the crash, they just don't have competition
This is actually their biggest market in retail ~
www.notebookcheck.net/Asus-ROG-Flow-Z13-GZ302EA-Convertible-Review-AMD-s-Strix-Halo-GPU-is-neck-and-neck-with-the-RTX-4070-Laptop.963266.0.html

Although the halo (desktop) cards probably do have much better margins!
Posted on Reply
#194
JustBenching
R0H1TAnd your point is? Nvidia fanboys would pay 2x more money for 1.2x more performance in their "favorite" benches :rolleyes:
They don't need to pay 2x more, the 4080 already smokes the best amd has to offer even in raster (latest TPU gpu benchmark). That's a 3 year old card, smoking AMD's best, at an area AMD is best at (raster). Nough said, move on.

People really need to get a grip, nvidia has the 6 fastest cards as of right now in raster. RASTER, an area they are not even trying to be competitive at...That's without fake frames, without DLSS, just pure raster performance. But yeah, those nvidia fanboys.

Posted on Reply
#195
azrael
The 5000 series launch is turning more and more into a complete shit show. I guess, Jensen's take on it would probably be "the more faulty dice you use, the more you earn".
Posted on Reply
#196
soulphie
JustBenchingBut would a defective 5090 just report as missing 8 rops? How? Doesn't nvidia have to fuse them off on a hardware level?
Yes, it seems like Nvidia fucked this up and mixed up 5090 and 5090D dies
Posted on Reply
#197
efikkan
So I guess the revised specs will be: "Up to 176 ROPs*****" then…
Darmok N JaladHaha, I collected baseball cards as a kid, and the “Error cards” were indeed worth more. So there we have it, the ZOTAC 5090E.
Shhhh! Don't give them any ideas! :scream:
Posted on Reply
#198
sethmatrix7
JustBenchingThey don't need to pay 2x more, the 4080 already smokes the best amd has to offer even in raster (latest TPU gpu benchmark). That's a 3 year old card, smoking AMD's best, at an area AMD is best at (raster). Nough said, move on.

People really need to get a grip, nvidia has the 6 fastest cards as of right now in raster. RASTER, an area they are not even trying to be competitive at...That's without fake frames, without DLSS, just pure raster performance. But yeah, those nvidia fanboys.

And half those cards could catch on fire, black screen of death off PCIe issues, or possibly even come with neutered performance like these 5090s.

Keep eating up that slop at minimum 25% over MSRP!
Posted on Reply
#199
dyonoctis
Sound_CardNvidia is literally the Apple of PC hardware. They can do whatever they want, because their consumer base allows them to. It does not matter if Apple breaks things and does slimy tactics, the sheep will continue to buy the products because it's a status statement. You are either in the 'in' or you are in the 'out'. Nvidia enjoys the exact same with their consumer base. It does not matter if they screw you over, the ecosystem is worth it, being in the club is worth it
Slimy tactics with the storage sure, but apple dislike when things aren't working on a hardware level (the bumpgate made them banish nvidia from the Mac, wich really hurt them for over a decade in the professional space.) And so far they've always fixed issues that became a real hindrance, like the butterfly keyboard, the antennagate etc...

Now Nvidia is in a different position where for a subset of their customers, there's no real 1:1 equivalent. You need top performance at 4K, the most popular GPGPU API ? There's only one brand that provide that right now.

I've said that before, but the company that manage to become a trillion dollar company by selling defectives products on the regular is a myth, small hindrance can get a pass, a fuck up once in while as well. But people never shut up about stuff that's actually getting in the way of using said product especially if it becomes reccurent. When you start to collect those kind of headlines your image takes a hit. The status symbol is credible because there's some truth to the premium aspect of the product. Ignore the issues for too long, and your credibility will be lost. Apple had to discontinue/ bend the knee and make changes when a product was actually so mediocre, the sales figures where actually not that good.

Nvidia hasn't released a product that had sooooo many contrersy about it for a long time, it's not as if every release was like that, and they've managed to climb to 90% marketshare regardless.
Posted on Reply
#200
R0H1T
JustBenchingThey don't need to pay 2x more, the 4080 already smokes the best amd has to offer even in raster (latest TPU gpu benchmark).
And how much does this out of production(?) card cost now?
Posted on Reply
Add your own comment
Feb 22nd, 2025 09:45 EST change timezone

New Forum Posts

Popular Reviews

Controversial News Posts