At the point when we originally discovered information about NVIDIA’s forthcoming GeForce RTX 4080 “Ada” coming in 12 GB and 16 GB variations, we realized there was more separating the two than just memory size and memory transport width. Turns out there’s much more. As indicated by itemized particulars spilled to the web, while the 16 GB variation of the RTX 4080 depends on a similar AD103, the second biggest chip after the AD102; the 12 GB RTX 4080 depends on the more modest AD104 chip which has a truly smaller memory transport.

It appears as though NVIDIA is appearing with the RTX 40-series with no less than three models — RTX 4090 24 GB, RTX 4080 16 GB, and RTX 4080 12 GB. The RTX 4090 is the head honcho part, with the ASIC code “AD102-300-xx.” It’s enriched with 16,384 CUDA centers, a lift recurrence of up to 2.52 GHz, 24 GB of 21 Gbps GDDR6X memory, and a run of mill designs influence (TGP) of 450 W, which is “configurable” up to 600 W. The RTX 4080 16 GB depends on the AD103-300-xx” accompanies 9,728 CUDA centers, a lift recurrence of 2.50 GHz, and 16 GB of 23 Gbps GDDR6X memory across a smaller memory transport than the one the RTX 4090 accompanies. This card supposedly has a 340 W TGP configurable up to 516 W.

The GeForce RTX 4090 12 GB is situated an indent underneath its 16 GB namesake, yet depends on the more modest AD104 chip, with 7,680 CUDA centers running at paces of up to 2.61 GHz, 12 GB of 21 Gbps GDDR6X memory, and a TGP of 285 W that is configurable up to 366 W. It’s fascinating the way that the hole incorporates TGP, yet additionally most extreme configurable TGP. The different load-up accomplices will use the last option as their power cutoff points to accomplish overclocked speeds. Indeed, even the NVIDIA Originators Release board is actually “custom design,” thus it could highlight higher-than-stock TGP.