Nvidia’s RTX 4080 graphics card might be rather less beefy than has been beforehand rumored, which follows swiftly on the heels of reports that the RTX 4070 could be a peppier GPU than anticipated.

This contemporary Lovelace hypothesis once more comes from Kopite7kimi, a daily {hardware} leaker on Twitter who has been supplying rumors at fairly a tempo of late (although that’s to be anticipated because the RTX 4000 vary nears its launch timeframe, with an anticipated debut in or round October, assuming no delays).

I am not a chatterbox, however I’ve to make some updates. I hope you do not thoughts. a potential RTX 4080, PG136/139-SKU360AD103-300-A19728FP32256bit 16G 21Gbps GDDR6Xtotal energy ~420WTSE ~15000Now I’ve accomplished the most recent replace for 4090, 4080 and 4070.August 5, 2022

See extra

The speculation is that the ‘potential’ RTX 4080 – a reminder that nothing is essentially set in stone, even when that is Nvidia’s present considering with the spec of the GPU – may run with 9,728 CUDA Cores, which is fewer than the beforehand rumored 10,240.

The leaker believes that the RTX 4080 will come geared up with 16GB of GDDR6X (at 21Gbps), and an influence utilization of round 420W, all of which stays in keeping with what was floated beforehand – the one change is that trimming down of the core depend at this level.

Evaluation: Nvidia’s positioning of RTX 4000 fashions

The very first thing to notice right here is that this isn’t a significant change. Dropping down 512 CUDA Cores – which suggests decreasing the streaming multiprocessors on-board by 4, from 80 to 76 – shouldn’t be an enormous transfer, but it surely does nonetheless symbolize a slight discount in uncooked grunt for the GPU.

Let’s keep in mind that that is only a rumor, and the leak panorama round Lovelace is seemingly shifting fairly often nowadays – which can be a mirrored image of Nvidia tinkering with and adjusting relative specs because the agency strikes in the direction of a finalized bunch of merchandise.

As talked about on the outset, Kopite7kimi additionally not too long ago introduced us information on the RTX 4070, with the obvious concept for that GPU being to up its efficiency stage at this stage – to the spec that was beforehand rumored for the RTX 4070 Ti, in truth. A fairly large step up, for positive. So, assuming that is all right, what may Nvidia’s reasoning be right here?

Bringing down the CUDA Core depend of the RTX 4080 a bit, whereas severely beefing up the RTX 4070 – not only for CUDA Cores, however the loadout of VRAM as properly – means, clearly sufficient, that the relative efficiency of the RTX 4070 and RTX 4080 is theoretically being introduced extra carefully in line.

With much less differentiation between the 2, and Nvidia’s xx80 mannequin usually being a great chunk costlier than the extra mainstream xx70 card – the comparatively inexpensive high-end GPU, if you’ll – with Lovelace, are we going to be taking a look at an RTX 4070 which is much more within the candy spot for many patrons?

Or to place it one other means, if the RTX 4080 isn’t all that a lot quicker, who will fork out for one relatively than a 4070? Except inventory is a matter for the latter, in fact; that might come into play, significantly if it’s actually common.

Additionally, we have to do not forget that CUDA Cores aren’t the total story for efficiency, with many different elements coming into play, like clock speeds. It’s right here that there might be a clue within the RTX 4070 supposedly sticking with round 300W for energy consumption, whereas the RTX 4080 will purportedly hit 420W or thereabouts – this might level to Nvidia severely beefing up clock speeds (and subsequently efficiency) with the higher-end GPU in comparison with the 4070.

Moreover, Kopite7kimi has maintained the earlier guestimate for a 3DMark TimeSpy Excessive rating for the RTX 4080 (round 15,000), regardless of these CUDA Cores being shaved off, so curiously sufficient, that might additionally recommend Nvidia could be pushing tougher with clocks right here for the 4080. Though the leaker doesn’t point out clock speeds particularly.

Absolutely it doesn’t make sense to have the RTX 4080 as a comparatively unattractive proposition in comparison with the 4070 – a a lot greater energy utilization gained’t assist both, by way of forcing PSU upgrades, and power prices normally – so for now, we’ll assume that if this rumor is true, we are able to’t but see the total image of what Nvidia is making an attempt to do with the positioning of Lovelace GPUs. Or that there’ll be one other rejigging of specs but to return…

Relative pricing might be an element right here, too – we haven’t heard a lot on that entrance from the rumor mill but, however we are able to’t think about the RTX 4000 collection shall be pitched with affordability in thoughts. Significantly not if Nvidia has loads of extra RTX 3000 inventory to take care of, because the grapevine contends, and shifting all that overlaps with the Lovelace launch (which is probably going solely a few months away now).

Regardless of the case, you’d think about Staff Inexperienced would’ve discovered from its errors by way of the RTX 3080 getting off on the improper foot, coming with solely 10GB of VRAM in its preliminary incarnation – an unpopular transfer compounded by the later launch of a 12GB variant. The latter was beefier not simply by way of the extra video reminiscence, leaving some patrons of the unique RTX 3080 with a critical case of regret.

By way of VideoCardz (opens in new tab)