Alleged NVIDIA GeForce RTX AD106 GPU Benchmarks Leak Out, On Par With RTX 3070 Ti
Alleged NVIDIA GeForce RTX AD106 GPU Benchmarks Leak Out, On Par With RTX 3070 Ti

Alleged performance benchmarks of an NVIDIA GeForce RT 40 graphics card based on the AD106 GPU have been leaked over at Chiphell forums.
According to Chiphell, the name of this particular GeForce RTX 40 GPU has not been finalized yet but it does feature the full AD106 GPU die with 4608 CUDA cores and packs 8 GB memory which means that it should offer a 128-bit bus interface. The GPU also packs 32 MB of L2 cache, a TMU count of 144, and 48 ROPs.
The card was tested within AIDA64 and various 3DMark benchmarks. In FP64, the card delivers up to 26 TFLOPs of compute power which is 13% higher than the RTX 3070 Ti. This gives us a boost of around 2.8 GHz which is higher than what we have seen on the other Ada cards. In 3DMark, AD106 scores anywhere from 10% slower to 3% faster than the RTX 3070 Ti in standard benchmarks and a 25% increase in the DXR Feature Test which shows its raytracing prowess.
The most important number is the power consumption which sees the card running with a peak board power draw of just 180W which is 27% lower than that of the RTX 3070 Ti and 23% lower than the older TU104 GPUs. The NVIDIA AD106 GPU is expected to land in the RTX 4060 graphics card but with different specifications as far as rumors are concerned. The interesting part is that the RTX 4070 Laptop GPU has similar specifications but it doesn't feature a boost clock this high nor does it feature a board power that exceeds 115W.
Compared to the RTX 3060 Ti which had a TGP of 200W, the RTX 4060 Ti seems to feature a 20% lower TGP which sounds in line with what we are hearing about the card itself. The GeForce RTX 4060 Ti is expected to utilize the AD106-350-A1 GPU core, a cut-down version of the full AD106 graphics chip, and based on rumors, it should pack 34 SMs or 4352 CUDA cores, an 8 GB GDDR6 memory running at 18 Gbps across a 128-bit bus interface, providing the card with 288 GB/s of bandwidth. There's also 32 MB of L2 cache on board the GPU which is an 8x increase over the GeForce RTX 3060 Ti.
This particular AD106 GPU from NVIDIA would most likely target 1080p gamers who want high refresh rates while 1440p gaming would also be possible using DLSS techniques. NVIDIA has a lot of new mainstream products expected to hit shelves by the second half of 2023.
News Sources: Harukaze5719, Olrak
What's Your Reaction?






