It is no secret to anyone that the RTX 4070 Ti that will be launched on the market in a few days is none other than the 12 GB RTX 4080 that was originally announced, but under another name. All because of the fact that luckily they realized that it did not leave them in a very good place to use the same trade name for two graphics cards with such different specifications from each other. Well, it seems that technical specifications have been leaked of said card, and this time it was NVIDIA itself.
NVIDIA seems to have planned the launch of its entire RTX 40 range in the opposite direction to the power that you are offering, they started with the RTX 4090, they continued with the RTX 4080 and now it is the turn of the RTX 4070 Ti. The big difference is that while the two previous graphics cards did not use the most powerful version of their respective GPUs, in the present case they did. That is, if the RTX 4090 uses the AD102 chip that has 144 physical cores, the 4090 uses 128 of them. In the case of the 4070 Ti, the maximum possible configuration of its chip is used, the AD104.
These are the official specifications of the RTX 4070 Ti
Apparently, the NVIDIA website had a small “lapse” and published the specifications of its RTX 4070 Ti ahead of time, which we had already known for months, but the difference is that it is no longer a slide on paper , but a real product that we should be able to enjoy in a few days. At least among the lucky ones who can have one. For our part, our interest is to see how it compares face to face with the RX 7900 XT, which will be sold for a similarly priced space and with the advantage of having 8 GB more memory than this NVIDIA model and higher bandwidth.
The advantages of this graphics card over its rival? Support for DLSS 3, higher Ray Tracing performance and higher clock speed. However, in things like the number of cores the AMD card beats, since it has 84 compared to the 60 cores of its rival from NVIDIA. In any case, comparing figures from different architectures does not make sense and in the end it all comes down to the performance of the games. Although we will not have to wait long to find out which of the two graphics cards is better under that price segment.
Although it is not the top model of the RTX 40, it is high-end.
And it is that NVIDIA sells it as a 4K graphics card, despite its 192-bit bus, which is what gives it its biggest disadvantage compared to its older sisters and against the RTX 30 of similar power. What’s more, NVIDIA places it as a rival to the RTX 3080, which had a 320-bit bus, but with a slower GDDR6X. In any case, we have to assume that the comparative graph in the image above is with the RTX 4070 Ti having all the standard aids active, especially the DLSS 3 frame interpolation. Not surprisingly, Flight Simulator, Cyberpunk and Warhammer 40,000 have their respective patches with support for the new technologies of the Lovelace architecture.
As for the price of the card, at the moment we do not know it, although when it was the RTX 4080 of 12 GB it had a price already set, it is possible that NVIDIA after seeing its AMD rivals has decided to vary the price slightly. Although we need a little more than a day to clear up doubts and have the final official price.
