NVIDIA’s Lovelace mega-flagship is allegedly going to be a whopping 2.6x faster than the RTX 3090, courtesy of the fully enabled AD102 die. According to Raichu, the GeForce RTX 4090 Ti can hit the 26,000 mark in 3DMark TimeSpy Extreme (possibly with a mild overclock). This is a notable 25% improvement over the RTX 4090 which manages just over 10K points.
Update: A 3DMark TimeSpy Extreme benchmark of either the RTX 4090 or the 4090 Ti has been spotted:
Compared to the RTX 3090, we’re talking about a massive upgrade of 2.6x or 160% and a substantial 2.3x or 130% over the RTX 3090 Ti.
The RTX 4090 Ti manages to overcome the Radeon RX 6950 XT by 2.4x or 140%.
| 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? |
