Moore’s Law Is Dead has recently shared a leaked prototype design for the RTX 4070. NVIDIA announced the Ada Lovelace series last month unveiling the RTX 4090 and the RTX 4080. There was no talk about the 4070 but we expect it to launch by December or early Q1 2023.
The RTX 4070 from what we can see features a 2-slot design which is much smaller than other Lovelace GPUs. The GPU appears to be a slight alteration of the RTX 3070 FE. It’s almost identical to the RTX 4080, which has an FE design exclusive to the 16GB variant.
Other than that, this shows that NVIDIA is indeed making FE models for GPUs below the 4080.
The RTX 4070
The RTX 4070 is probably one of the most controversial GPUs ever. Initially, this GPU was supposed to feature 12GB of memory over a 192-bit bus. That was all fine and dandy until NVIDIA revealed the RTX 4080 12GB. This GPU as highlighted by many enthusiasts is actually an RTX 4070 in disguise. The pricing also became a major concern since you’re paying $900 for a once -70 series GPU.
This really does beg the question. What should we expect from the actual RTX 4070? If we follow the leaks, the 4070 ships with 10GB of G6X memory over a thin 160-bit bus. NVIDIA may opt for a cut-down AD104 GPU which for the sake of comparison is the same chip powering the 4080 12GB and possibly the 4060 as well.
In terms of pricing, NVIDIA has placed major traps for itself. Even at $700, which is $200 less than the RTX 4080 12GB, the RTX 4070 will be a whopping 40% more expensive than last-gen. On the other hand, a price of $600 is still reasonable but leaves too much market share for AMD in the $600–$900 category.
|SKU||Chip||FP32/CUDA||Max Clock||Cache||Memory Bus||VRAM||Memory Spec||Speed (Gbps)||TDP|
|RTX 4000 Titan?||AD102-450||18176||3.0GHz+?||96MB?||384/382-bit||48GB||GDDR6X||24||~800W|
|RTX 4090 Ti||AD102-350||18432||3.0 GHz?||96MB||382-bit||24GB||GDDR6X||24||600W?|
|RTX 4090||AD102-300-A1||16384||2.52GHz||96MB||384-bit||24GB||GDDR6X||21||450W+ (TGP) / 660W (Max TGP)|
|RTX 4080 Ti||AD102||14848?||2.7 GHz?||80MB?||320-bit||20GB||GDDR6X||23||420W|
|RTX 4080 (Variant 1)||AD103-300-A1||9728||2.505GHz||64MB||256-bit||16GB||GDDR6X||22.5||320W(TGP)/ 516W (Max TGP)|
|RTX 4080 (Variant 2)||AD104-400-A1||7680||2.61GHz||48MB||192-bit||12GB||GDDR6X||21||285W (TGP) /366W (Max TGP)|
|RTX 4070 Ti||AD104-300?||7680||3.0GHz?||?||192-bit||12GB||GDDR6X||21||300W?|