NVIDIA's new GeForce RTX 5090 and GeForce RTX 5070 Ti graphics cards are shipping with missing ROPs, seeing the flagship RTX 5090 experiencing 4.5% reduced performance.

The news started with TechPowerUp user "Wuxi Gamer" discovering that some review samples and even retail graphics cards with missing ROPs, with TPU noting that its ZOTAC GeForce RTX 5090 SOLID graphics card sample was underperforming by a few percentage points, falling behind NVIDIA's in-house GeForce RTX 5090 Founders Edition.
The missing ROPs are a major issue, with ROPs being responsible for pixel data, taking care of tasks like blending and anti-aliasing. NVIDIA usually reduces ROPs count to keep separation between SKUs, but this is a far bigger issue.
It's also not an isolated case with a ZOTAC-branded GeForce RTX 5090, but the Manli GeForce RTX 5090D is also experiencing issues, and leaker MEGAsizeGPU mentioned on X that "the root cause is the chip. A small batch of GB202 is defective, and the BIOS can not do anything with this issue".
NVIDIA's global PR director Ben Berraondo confirmed the issue, telling The Verge: "We 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".


TPU explained: "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".