NVIDIA RTX 4080 rumored specs updated with faster memory
NVIDIA is working around the clock to prepare their next-gen GPU series for launch later this year. This also means that the company is cycling through multiple cards specs while they evaluate their cost and performance of each model. Kopite7kimi, better known for his accurate RTX 30 “Ampere” leaks, is providing frequent updates on how the RTX 40 series specs change, and yes, they change a lot.
According to the latest update, NVIDIA GeForce RTX 4080 SKU might get faster 23 Gbps GDDR6X memory. This is an upgrade over previous specs listing 21 Gbps, and possibly an indicator that the flagship RTX 4090 will get similar speed as well. Furthermore, Kopite now claims that the TBP has changed from 420W to 340W. Note this is the total board power, not TDP. The remaining specs did not change, and this model is still expected to get AD103 GPU with 9728 CUDA cores and 16GB of memory across a 256-bit bus. Those unfamiliar with NVIDIA production cycle might think that Kopite is just making those specs up, but rest assured NVIDIA really does change the prelaunch specs very often. Some of these cards appear on the leaked roadmaps (RTX 3070 Ti 16GB, RTX 3080 20GB etc.) but are later canceled or postponed (RTX 3090 Ti, GTX 1630). So yes, it is very likely that those updates are based on real information, some we can even confirm ourselves, but Kopite’s updates are much faster. However, one shouldn’t automatically assume those specs are final, after all we are still 1.5 – 2 months away from official launch.