After all leaks, it’s time for something official. Details on the specifications of the GeForce RTX 2080 Ti and GeForce RTX 2080 have now been revealed by PNY, Nvidia’s partner, who rushed to publish the information card.
This one presented his non-reference GeForce RTX 2080 Ti and GeForce RTX 2080 models from the XLR8 Gaming series. It will offer higher clocks than Nvidia’s reference proposals, which will also translate into higher TDP. But let’s start with the specification itself.
The GeForce RTX 2080 Ti XLR8 boasts 4352 CUDA cores and 11 GB of GDDR6 memory with a 352-bit bus with a bandwidth of 616 GB / s. In turn, its core clock is the base 1350 MHz and 1545 in the boost mode. As this is a turned up model, we can postpone leaks with frequencies in the order of 1800 MHz between fairy tales. Its TDP is to be 285W, and all amateurs of multi-GPU configurations will have to settle for a bi-directional NVLink connector.
Also read: Nvidia claims that the cryptocurrency demand for the GPU is a thing of the past
On the other hand, the GeForce RTX 2080 XLR8 has a slightly higher clock speed, because it reaches 1710 MHz, but it is of course more truncated in terms of the rest. We are talking about 2944 CUDA cores and 8 GB of GDDR6 memory with a 256-bit bus and a bandwidth of 448 GB / s. Surprisingly, the TDP in this case also reaches 285W. So it is possible that this specification will change in the future, but rather in terms of TDP and clocks.
As you can see, both the GeForce RTX 2080 Ti and GeForce RTX 2080 have not addressed the issue of Tensor cores. However, these should be there, as indicated by some information about ray-tracing support. The price itself is also likely fake and will change over time.