The NVIDIA AD102 chip could double the performance of the RTX 3090
July 19, 2022
0
graphics processor AD102 aims to be the next technological reference NVIDIA. In addition to the RTX 4090 and its Ti variant, it could also be used in the
graphics processor AD102 aims to be the next technological reference NVIDIA. In addition to the RTX 4090 and its Ti variant, it could also be used in the next generation TITAN, a series more oriented towards professional sectors and research, although this does not mean that it cannot run video games and other applications used primarily by users. common
We recently learned about the existence of a graphics card that uses a variant of the processor AD102, which, however, equips 18,432 CUDA coreswhich is approximately 12.5% more than theoretically introduced in the standard model RTX 4090. This number of CUDA cores could correspond to the Ti variant or the next generation of the TITAN series, but for now it is more likely to be the latter.
With an AD102 processor and 18,432 CUDA cores, this mysterious graphics card doubled the performance offered by the RTX 3090 in some tests performed with the video game. Management. The conditions were to run the game in 4K and with ray tracing and DLSS enabled. The RTX 3090 recorded an average of 72 frames per second, while an unknown GPU AD102 achieved more than 160 frames in a second. Unfortunately, there are many things missing in this data for a complete vision of the panorama and especially a comparison of the two graphs.
The 3090 in Control with the same settings does 72 FPS on Avg, so a 2.2x gain pic.twitter.com/UdK2kCEXhO
An improvement by an unknown chart may have been caused Ada Lovelacenext generation and NVIDIA architecture, would use a a new generation of Tensor cores that is said to bring noticeable improvements in DLSS performancethe green giant’s supersampling technology created to minimize the performance loss caused by ray tracing, which opened the door to standardizing the use of 4K and 8K resolutions on PCs.
From what we see, Ada Lovelace could represent a significant improvement over Ampere, but for now it’s better to take all the data we know with tweezers, because it wouldn’t be the first time that a generation of graphics promises a revolution and then be a less qualitative leap. At the end of the summer, we could have the first touch of Ada Lovelace with a possible announcement of the RTX 4090 and/or its Ti variant. When that time comes, we’ll have a fuller and clearer picture of what the next generation of NVIDIA graphics will offer.
The complete AD102 chip aims to consume up to 800 wattsan amount that wouldn’t be too surprising if we consider that it will be included in the high-end products (RTX 4090 and probably TITAN).
Alice Smith is a seasoned journalist and writer for Div Bracket. She has a keen sense of what’s important and is always on top of the latest trends. Alice provides in-depth coverage of the most talked-about news stories, delivering insightful and thought-provoking articles that keep her readers informed and engaged.