Ticker

6/recent/ticker-posts

NVIDIA GeForce RTX 4090/4080/4070 get new rumored specs, RTX 4070 with 7168 cores and 10GB 160-bit memory?

 Yet another rumored GeForce RTX 40 specs

RTX 4090 with more cores
According to Kopite, the RTX 4090 would get more cores than previously reported. Instead of 16128 CUDAs, the new specs list 16384 cores, so an upgrade from 126 SM (Streaming Multiprocessors) to 128. The full configuration of AD102 GPU supposedly has 144 cores, but such configuration is more than likely to be reserved for future TITAN/RTX4090Ti SKU.
Other parts of the specs are unchanged. This card is the flagship RTX 40 model with TDP expected to reach 450W. It should launch with 24GB GDDR6X memory clocked at 21 Gbps.

RTX 4080 with 10240 CUDA cores
Furthermore, we now have first rumored GPU configurations of RTX 4080 and RTX 4070 GPUs. The former is the only known AD103 based graphics card, which would supposedly debut a month after RTX 4090. According to Kopite, the RTX 4080 would feature 10240 CUDA cores and 16GB memory.

Interestingly, this card might not rely on GDDR6X like RTX 4090 supposedly does, but GDDR6 instead. This part of the specs is most likely to change though, because 18 Gbps GDDR6 memory would actually be slower than RTX 30 predecessors (RTX 3080 has 19 Gbps memory). However, one should underline the fact that AD102 GPU has larger L2 cache, which may work just as efficiently as Radeon RX 6000 series. In other words, the memory speed does not need to be that much faster.

RTX 4070 with 10GB and 160-bit memory
Finally, RTX 4070 sees a rumored spec update. The card is no longer expected with 12GB memory but 10 GB instead. This would also mean a narrower memory bus from 192-bit to 160-bit. It is said that RTX 4070 would feature 7168 CUDA cores, which would mean 512 disabled CUDA cores on AD104 GPU.

The RTX 40 series are now rumored to launch in late Q3/early Q4. The initial lineup would feature all three SKUs mentioned above, however their rollout would span across the fourth quarter.

Post a Comment

0 Comments