Nvidia’s RTX 4060 Ti is the most recent Lovelace GPU to be the topic of an data dump relating to its purported spec, as rumors concerning the RTX 4000 vary proceed to return thick and quick.

This hypothesis is from Kopite7kimi, an everyday in terms of Twitter {hardware} leaks, and it supplies a good few particulars (hat tip to VideoCardz (opens in new tab) for recognizing the tweet).

RTX 4060 Ti has a really quick reference board. The PG190 nonetheless makes use of CEM5 connector. AD106-350-A14352FP328G 18Gbps GDDR632M L2220WDecember 13, 2022

See extra

Go into this with plentiful seasoning in hand, however as you’ll be able to see, the GPU used is AD106 (supposedly AD106-350-A1, to be exact), and we have now a CUDA Core rely of 4,352, alongside 8GB of GDDR6 video RAM (at 18Gbps). The TGP is pegged at 220W.

As VideoCardz factors out, earlier spillage has instructed that the AD106 GPU will run with a 128-bit reminiscence bus, which might give the RTX 4060 Ti a bandwidth of 288GB/s in concept.

We’re additionally informed that the RTX 4060 Ti has a “very quick” reference board.

Evaluation: Nonetheless a protracted street forward for this GPU

The purported spec proven right here has prompted disappointment as a normal response on-line.

Naturally, we must be skeptical about any rumored specs – and we must always keep in mind that that is nonetheless early improvement for the RTX 4060 Ti, so far as we’re conscious, so even when true, the specs floated right here might nonetheless be modified by Nvidia down the road.

The CUDA Core rely itself has drawn some groans, however we are able to’t choose the facility of a card purely off that spec. Sure, it truly has fewer cores than the RTX 3060 Ti (which has 4,864) in keeping with this spillage, however there are generational architectural enhancements to take into consideration, and all the opposite GPU specs too.

What’s extra of a sticking level for a lot of is the usage of 8GB of VRAM, and that quite meagre wanting theoretical reminiscence bandwidth, which might be an unlucky mixture – although the much-increased L2 cache, 32MB, might be a useful increase in comparison with the 3060 Ti which had 4MB. Moreover, we must always add additional salt right here, because the bandwidth calculation is predicated on a earlier rumored spec, which makes issues shakier.

At any price, avid gamers on the market have been already sad with the 3060 Ti utilizing 8GB, not to mention a extra highly effective 4060 Ti. However regardless of the case, broad expectations are for a graphics card that’s the equal of the RTX 3070 Ti – and rather a lot will then rely on the pricing. Which given Nvidia’s present type, doesn’t make us optimistic.

Above all, we must always keep in mind that all these specs might change based mostly on the response to the RTX 4070 Ti and 4070, and the way issues pan out with mid-range RDNA three GPUs, after all. There’s seemingly a very long time to go, too, earlier than the RTX 4060 Ti emerges – actually the rumor mill doesn’t imagine the RTX 4060 will likely be out till mid-2023 – and Nvidia nonetheless has a good whack of RTX 3000 inventory to clear, as everyone knows, little question a great deal of it within the mid-range.

So, on condition that prolonged timeframe for the RTX 4060 Ti and 4060’s seemingly look, even when these are the specs on the board proper now, we’d be shocked in the event that they didn’t change at throughout time based mostly on the elements talked about and altering GPU panorama.

At the moment’s greatest graphics card offersPresent Extra OffersWe test over 250 million merchandise daily for the very best costs