Nvidia’s RTX 4070 is probably not among the many first batch of next-gen GPUs launched, and if the rumor mill is true, we’ll get two completely different spins on the RTX 4080 arriving in addition to the flagship RTX 4090.

The newest chatter from the grapevine on Lovelace comes from a number of sources who VideoCardz (opens in new tab) spoke to, mixed with information drawn from common {hardware} leaker Kopite7kimi on Twitter (who just lately summarized (opens in new tab) all their newest findings), and Chiphell (opens in new tab) (Chinese language boards), all of which corroborates VideoCardz’s findings in some locations.

There’s some recent information on clock speeds and energy utilization, however the greatest revelation right here is that Nvidia appears to be planning to launch two variations of the RTX 4080, one with 16GB of VRAM, and one with 12GB – however there will probably be different main variations with the spec (take all this with massive heaps of seasoning, naturally, as with every rumor).

The RTX 4080 16GB is the one stated to be constructed on the AD103 GPU with 9,728 CUDA Cores, with the 12GB spin on the 4080 supposedly dropping down to make use of the lesser AD104 GPU, and it’ll sport 7,680 CUDA Cores – fairly a distinction.

As VideoCardz observes, what seems to be the case is that the RTX 4080 12GB is what was beforehand recognized on the rumor mill because the RTX 4070 (or 4070 Ti) – which type of explains why it appeared so beefy, one thing we’ve commented on in latest occasions.

VideoCardz informs us that Nvidia has made a “last-minute title change” for this card to show it right into a lesser spin of the RTX 4080, and for the second, there isn’t a RTX 4070 – with graphics card producers not anticipating this to alter (with the Lovelace launch now so shut: it’s anticipated on September 20, at GTC).

Different spec nuggets shared on this recent rumor dump embrace that the RTX 4080 12GB may have an influence utilization of 285W (TGP or Whole Graphics Energy), which might be pushed as much as 366W. The increase clock will purportedly be 2610MHz.

The stronger model of the RTX 4080 with 16GB of VRAM will supposedly increase to 2505MHz, with a TGP of 340W (in a position to be pushed as much as 516W most). The GDDR6X reminiscence is claimed to run at 23Gbps, notably, and this would be the solely GPU to have that speedy VRAM.

As for the RTX 4090, that enhances to 2520MHz, and the rumor mill once more factors to that long-held TGP of 450W, however at max it may be pushed to 660W. The 24GB of video RAM will probably be 21Gbps modules, the identical because the lesser RTX 4080 12GB.

The tech web site tells us that the RTX 4090 ought to be out within the first half of October, and the RTX 4080 fashions are apparently set to debut within the preliminary two weeks of November.

Evaluation: RTX 4070 ETA?

So, what concerning the RTX 4070 then? Properly, we had heard point out a few weeks again – once more from Kopite7kimi on Twitter – that there have been two attainable specs for the 4070 graphics card. The talked about higher-end chance is now the RTX 4080 12GB, in idea, in order that would go away the opposite configuration of an AD104 chip with 10GB of VRAM and seven,168 CUDA Cores to be the RTX 4070. Perhaps. (Let’s face it, the Lovelace rumor area is changing into more and more complicated).

If that is what’s taking place, it’ll be disappointing for a few causes. Firstly, the RTX 4070 solely having 10GB of video reminiscence gained’t go down properly, and secondly, the prospect of it being late to the desk is disheartening, too. When will it flip up? If  there’s no sight of it but, in any respect, is that this a GPU that Nvidia is planning to delay till early 2023, as some rumors have instructed?

Earlier than we get too downhearted, although, what we don’t but have is any indication of pricing for Lovelace exterior of the RTX 4090 (we’ve simply heard some hints as to the flagship’s price ticket, and sadly, the information isn’t what we wished to listen to).

If we aren’t getting an RTX 4070 initially, however we’re getting an RTX 4080 12GB which is priced comparatively okay for its efficiency – contemplating it does drop down a good method from the 4080 16GB when it comes to a lesser GPU and significantly fewer CUDA Cores – then possibly it gained’t matter. Which is to say that it’ll all be concerning the worth/efficiency ratio of those GPUs, no matter they find yourself being known as.

That Nvidia has purportedly determined to notch up the title to an RTX 4080 from the 4070, nevertheless, leaves a worrying feeling about pricing in our intestine. Notably given the above talked about RTX 4090 pricing leak. Time will inform, as ever, and with lower than every week to go earlier than the Lovelace launch, we gained’t have to attend lengthy to find extra about Workforce Inexperienced’s next-gen graphics playing cards.

All this does go away us questioning, nevertheless, whether or not the tactic is likely to be to cost a premium for the brand new merchandise – just about as anticipated for high-end GPUs recent off the meeting strains – and additional knock down pricing on that extra RTX 3000 inventory as the choice possibility for individuals who don’t wish to splash out an excessive amount of.

Right this moment’s finest graphics card offersPresent Extra OffersWe verify over 250 million merchandise day-after-day for one of the best costs