Nvidia RTX 4070
Put that foreign holiday on hold, shouts Dave James, it’s time to invest in a shiny new graphics card instead!
SPECS
Architecture: Ada
GPU: AD104
Process: TSMC 4nm
Trans: 35.8B
Die size: 294mm2
CUDA: 5,888
SMs: 46
Tensor: 184
Ray: 46
ROPs: 64
Clock: 1.92GHz (2.48GHz boost)
Memory: 12GB GDDR6, 192-bit
Cache: 36MB
TBP: 200W
API: CUDA 8.1, Vulkan RT, OpenGL 4.6, AV1 codec
The Nvidia RTX 4070 is a £80-cheaper RTX 3080. That’s the easiest way to describe the green team’s new graphics card. This is the fourth entry in the notoriously expensive Ada generation of GPUs, and in standard metrics it performs as well as the fourth-tier card from the Ampere line-up. On the face of it, then, it’s just a cheaper chip.
The RTX 4070 is like a proper graphics card. It’s not some monstrous hulk of PCIe socketrending GPU, it’s a modest card the size of its RTX 3070 forebear. That makes it a cute-looking thing. Well, in terms of scale anyway; that brushed aluminium Nvidia frame still looks pleasingly serious.
And that’s more than aesthetics, too. The size of the card hints at the efficiency of the 4nm Ada GPU inside. If you want a powerful but low-power card, the RTX 4070 fits the bill. Which will no doubt make it the darling of the small form factor PC brigade, and deservedly so.
That’s certainly one of the benefits we were alluding to earlier in the RTX 4070 versus RTX 3080 debate. A less clear-cut one is that this Ada card has access to DLSS 3.0 and its Frame Generation. A game-changer when supported, but if you don’t own any (of the few) games that use it, it’s left an effectively useless feature.