Bofferbrauer2 said:
Well, RDNA (Navi) was already rather competitive with NVidia's counterparts apart from the lack of Hardware Raytracing. RDNA2 is supposed to both add the Raytracing and drop the power consumption by about 50% for the same performance. If AMD can manage this (and seeing how it went with Zen and Zen 2, I expect this to be very possible), then NVidia really needs to act, otherwise AMD would blow right past them like they did with Intel. That said, I don't really believe those leaks either, not only because those specs look more like a geek's dream, even more since Ampere is supposedly the Tesla V100/Titan V successor with full double precision calculation capacity, not any GTX/RTX GPU (yet). |
The key difference here is that, unlike Intel, Nvidia hasn't been resting on their laurels. They have kept pushing their architectures and products for the enterprise side of their business with enhancements and features that ended up coming down to their gaming oriented products, only with fewer parts.
In any case, we'll know more in a few days when Nvidia does its conference. Hopefully we'll know the max specs for their enterprise products and that will give us an idea of what we can expect for their gaming products (Turing's top enterprise card has 72 SMs and 4600 shaders while the 2080Ti "only" has 68 SMs and 4352 shaders).
Please excuse my bad English.
Former gaming PC: i5-4670k@stock (for now), 16Gb RAM 1600 MHz and a GTX 1070
Current gaming PC: R5-7600, 32GB RAM 6000MT/s (CL30) and a RX 9060XT 16GB
Steam / Live / NNID : jonxiquet Add me if you want, but I'm a single player gamer.







