[ad_1]
Kopite7kimi, one of the most renowned figures on the web when it comes to revealing Nvidia information, has struck again. The latter thus reveals the technical characteristics of the next GeForce RTX 5080 and RTX 5090 graphics cards. We can thus discover above all that they will carry GDDR7 type video memory, which in reality corresponds with the recent launch of the production of mass launched by Hynix and Micron. This memory allows superior performance compared to previous generation models, equipped with GDDR6X.
Only 16 GB of VRAM for the GeForce RTX 5080
We will therefore be entitled to 32 GB of VRAM for the GeForce RTX 5090 and 16 GB of VRAM for the GeForce RTX 5080. And this is where the problem lies: in fact, 16 GB of video memory for a high-end graphics card , it’s not much. Not to mention that the GeForce RTX 5080 is also much more energy consuming, with a consumption of 400W, compared to 320W for the GeForc RTX 4080. Note also in this regard that the GeForce RTX 4090 could consume up to 600W (!) .
Moreover, the GeForce RTX 5090 will have a much higher number of computing units: 21,760. This is twice as many as the GeForce RTX 5080 (10,752), which nevertheless remains close to the GeForce RTX 4080 with its 9,728 computing units, and the reason for this quasi status quo is simple: the GeForce RTX 5080 uses the same 256-bit memory bus as the GeForce RTX 4080, where the RTX 5090 uses a bus width 512 bit. So, if the GeForce RTX 5090 graphics card does indeed present a real boost in power and graphics capabilities, the result could be less glorious for the GeForce RTX 5080, especially if the latter remains in the same price range as the GeForce RTX 4080 during of its marketing.
We now have to wait for an official announcement from Nvidia, hoping that some of this information turns out to be incorrect (notably the 16GB of VRAM for the RTX 5080…).
[ad_2]
Source link -98