After the GeForce RTX 5090, the RTX 5070 Ti seems to follow a similar pattern of bringing fewer ROPs than NVIDIA specified.
NVIDIA RTX 5070 Ti Spotted With 88 ROPs Instead of 96, Theoretically Reducing Performance Noticeably
This is messy! NVIDIA did confirm that 0.5% of their produced RTX 5090/5090D and RTX 5070 Ti GPUs feature fewer ROPs than specified. Even though this number can't be verified, it's already misleading from the very start. Not informing the customers about the inferior specifications is inexcusable since it directly affects the promised results.
If you think the reports of fewer ROPs on the RTX 5090 have stopped surfacing, then you should check another one, but this time, it's the GeForce RTX 5070 Ti. It should be obvious now, as NVIDIA itself confirmed the RTX 5070 Ti to be one of those affected GPUs, but it doesn't seem to be as small as NVIDIA wants to show.
A user on Facebook posted a screenshot of the GPU-Z (via @GawroskiT), revealing the reduced ROP count on an RTX 5070 Ti edition. While we don't know the exact GPU edition and to which vendor it belongs, this confirms the presence of such RTX 5070 Ti GPUs in the market. Compared to the 96 ROPs specified by NVIDIA in the official spec sheet for the GeForce RTX 5070 Ti, this one brings only 88 ROPs. This is an 8.4% reduction, which is noticeable and will definitely impact the gaming performance.
While NVIDIA says that the impact is 4% on average, it could be higher or lower depending on the games being played. Nonetheless, this figure is still high and cannot be excused since customers are paying the price in full. In most cases, they are paying even more than what the GPU actually costs, given the inventory situation.
Now, if you take a quick look at some other specs of this particular RTX 5070 Ti edition, you can clearly see that the reduction in ROPs has impacted the Pixel Fillrate, which is directly related to ROPs. Compared to 287.7 GPixel/s on the RTX 5070 Ti with 96 ROPs, the nerfed RTX 5070 Ti delivers only 223.7 GPixel/s of Pixel Fillrate. However, do keep in mind that the boost clock on the latter is much lower. Keeping the boost clock at 2.99 GHz on the affected GPU still delivers 263.12 GPixel/s of Pixel Fillrate, which is roughly 9% lower.
NVIDIA and its board partners are misleading its customers by not notifying them of inferior specs even though these cards pass QA testing before reaching the shelves. Despite claiming that these account for up to only 0.5% of all the manufactured units, the number can quickly go into the thousands.
About the author: Sarfraz Khan is a hardware reporter with a focus on PC components and the builder community. With years of experience writing about PC hardware and laptops, his work has been featured on several reputable technology publications. Sarfraz's hands-on experience is demonstrated through his first-person accounts of using and comparing different hardware configurations, providing practical and relatable insights for everyday users. His technical analysis is respected by peers in the enthusiast community and has been cited by specialized hardware sites such as Germany's Igor's Lab.
Follow Wccftech on Google to get more of our news coverage in your feeds.
