The GeForce RTX 4060 Ti specifications have surfaced for the first time. Based on the AD106 die, this budget Ada GPU will feature 4,352 FP32 CUDA cores and 8GB GDDR6 memory across a 256-bit bus. The memory will be clocked at 18Gbps, backed by 32MB of L2 cache. The graphics card has a total board power of 220W.
RTX 4060 Ti has a very short reference board. The PG190 still uses CEM5 connector.
AD106-350-A1
4352FP32
8G 18Gbps GDDR6
32M L2
220W— kopite7kimi (@kopite7kimi) December 13, 2022
The GeForce RTX 4060 Ti is the first 40 series GPU we know of that features fewer cores than its predecessor (4,352 vs. 4,864). NVIDIA has offset this by pairing the card with faster 18Gbps memory (vs. 14Gbps on the RTX 3060 Ti), backed by an additional 32MB of L2 cache.
| 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? |
The AD106 is the third Ada Lovelace die after the AD102 (RTX 4090) and the AD104 (RTX 4080). We expect the RTX 4060 Ti to cost between $400 and $500. The GPU should hit retail shelves in Q3 2023.
