If the low stock for the RTX 50 series GPU wasn’t enough of a reason to criticize Nvidia, some users are now reporting less than advertised Render Output Units (ROPs) across their GPUs. First spotted by techpowerup, the RTX 5090 was spotted with 168 ROPs instead of the advertised 176. Some RTX 5070 Ti also displayed 88 ROPs instead of 96.
This was revealed during a check through the GPU-Z application. Soon, many others also started reporting similar issues. Nvidia has addressed the issue and asked affected users to contact board manufacturers for a replacement.
“We have identified a rare issue affecting less than 0.5% (half a percent) of GeForce RTX 5090 / 5090D, RTX 5080, and 5070 Ti GPUs which have one fewer ROP than specified. The average graphical performance impact is 4%, with no impact on AI (Artand Compute workloads. Affected consumers can contact the board manufacturer for a replacement. The production anomaly has been corrected,” the statement said in a statement to The Verge.
Also Read: Nvidia RTX 5090 and 5080 sold out in under a minute after launch
Some RTX 5080 GPUs were also affected
Days after the 5090 and 5070 Ti with lower ROPs surfaced, the RTX 5080 also joined the fray to complete the holy trinity of another Nvidia launch. It is also important to note that Nvidia mentioned one fewer ROP in its statement which translates to 8 ROPs. This is because one hardware unit disabled means eight pixels per clock less, hence a loss of 8 ROPs as per techpowerup.
Although fewer ROPs remain a smaller concern compared to the fact that people are unable to get their hands on the RTX 50 series GPUs. Those that are available are reported to be up to 30% above the MSRP thanks to third-party board partners.
More RTX cards to be announced in March
With enthusiast and mid-range cards revealed, the entry-level and budget GPUs are expected to be revealed sometime in mid-March. These include the RTX 5060/RTX 5060 Ti. There are also mentions of an RTX 5050 which is another interesting addition to this generation.