NVIDIA has teased the launch event of its next-gen GeForce RTX graphics cards set to take place during GTC 2022. Although this event is known for AI, machine learning, and other scientific implementations of NVIDIA GPGPUs, Jensen has decided to announce the RTX 40 “Lovelace” series GPUs during the same. The below tweet from the GeForce Twitter account points to a 20th September launch/reveal.
It is believed that the high-end and upper midrange SKUs including the RTX 4090, RTX 4080, and the RTX 4070 will be shown off to the public. The GeForce RTX 4060, RTX 4060 Ti, 4070 Ti, and possibly even the RTX 4090 Ti are reportedly going to hit retail in the first half of 2023.
GPU | GA102 | AD102 | RTX 4090 | AD103 | RTX 4080 | RTX 4070 Ti (AD104) | RTX 4070 |
---|---|---|---|---|---|---|---|
Arch | Ampere | Ada Lovelace | Ada Lovelace | Ada Lovelace | |||
Process | Sam 8nm LPP | TSMC 5nm | TSMC 5nm | TSMC 5nm | |||
GPC | 7 | 12 | 11 | 7 | 7 | 5 | 5 |
TPC | 42 | 72 | 64 | 42 | 40 | 30 | 30 |
SMs | 84 | 144 | 128 | 84 | 80 | 60 | 60 |
Shaders | 10,752 | 18,432 | 16,384 | 10,752 | 9,728 | 7,680 | 7,680 |
TP | 37.6 | ~100 TFLOPs? | 83 TFLOPs | ~50 TFLOPs | 47 TFLOPs? | ~35 TFLOPs | 35 TFLOPs? |
Memory | 24GB GDDR6X | 48GB GDDR6X | 24GB GDDR6X | 16GB GDDR6X | 12GB GDDR6X | ||
L2 Cache | 6MB | 96MB | 72MB | 64MB | 48MB | ||
Bus Width | 384-bit | 384-bit | 256-bit | 192-bit | |||
TGP | 350W | 600W | 450W | 450W | 285-340W | 300W | 285W |
Launch | Sep 2020 | Sept 22? | Sept 22? | Q1 2023? |
From what we’ve seen the RTX 40 series cards will offer nearly a 2x performance gain over their 30 series predecessors with a substantial boost in ray-tracing render times. Leveraging TSMC’s 5nm N5 process, NVIDIA has packed a whopping 18,432 cores onto a single die, designing the first consumer GPU to potentially break the 100 TFLOPs barrier.