NVIDIA GeForce RTX 4060 Ti Reportedly Features 8GB GDDR6, 220W TDP
This is the lowest model of the RTX 40 Series leaked so far – once again courtesy of prominent NVIDIA leaker @kopite7kimi.
While this specs are not necessarily final (NVIDIA may change them as development moves forward), RTX 4060 Ti is expected to pack 4,352 CUDA cores – about one-quarter of that from the flagship RTX 4090 (which contains 16,384 CUDAs). It is expected to use a 128-bit memory bus according to earlier leaks, and with 8GB GDDR6 clocking 18Gbps, should calculate to theoretical max bandwidth of 288GB/s.
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
This card is rated for 220W TDP, which is 30W less than the specs rumored for its bigger sibling, RTX 4070. Another interesting point from this leak is the size of the reference board, which claimed to be “very short” – supposedly be even smaller than RTX 4070 models. It is likely a 12VHPWR connector will be used to reduce part of that footprint – and that also means an adapter (most likely a 2x 8-pin) is pretty much warranted for this card.
Just a few days earlier, the RTX 4070 was leaked prior to this leak: which, reportedly, is coming with a cut-down AD104 silicon containing 5,888 CUDAs. That GPU will also be running on a 192-bit, 12GB GDDR6X memory and amounting to a TDP of 250W. So far, both models mentioned in this article has no release dates known.
Source: Videocardz
Pokdepinion: Interestingly they dropped GDDR6X on the successor of the 3060 Ti…
Spread the PokdeLove
Related
NVIDIA GeForce RTX 4060 Ti Reportedly Features 8GB GDDR6, 220W TDP
This is the lowest model of the RTX 40 Series leaked so far – once again courtesy of prominent NVIDIA leaker @kopite7kimi.
While this specs are not necessarily final (NVIDIA may change them as development moves forward), RTX 4060 Ti is expected to pack 4,352 CUDA cores – about one-quarter of that from the flagship RTX 4090 (which contains 16,384 CUDAs). It is expected to use a 128-bit memory bus according to earlier leaks, and with 8GB GDDR6 clocking 18Gbps, should calculate to theoretical max bandwidth of 288GB/s.
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
This card is rated for 220W TDP, which is 30W less than the specs rumored for its bigger sibling, RTX 4070. Another interesting point from this leak is the size of the reference board, which claimed to be “very short” – supposedly be even smaller than RTX 4070 models. It is likely a 12VHPWR connector will be used to reduce part of that footprint – and that also means an adapter (most likely a 2x 8-pin) is pretty much warranted for this card.
Just a few days earlier, the RTX 4070 was leaked prior to this leak: which, reportedly, is coming with a cut-down AD104 silicon containing 5,888 CUDAs. That GPU will also be running on a 192-bit, 12GB GDDR6X memory and amounting to a TDP of 250W. So far, both models mentioned in this article has no release dates known.
Source: Videocardz
Pokdepinion: Interestingly they dropped GDDR6X on the successor of the 3060 Ti…
