Nvidia might skip 12nm GPUs and go straight to 7nm, regardless of contemporary GTX 1180 / 1170 rumours

Nvidia might skip 12nm GPUs and go straight to 7nm, regardless of contemporary GTX 1180 / 1170 rumours

Fresh hypothesis in regards to the next-generation of Nvidia graphics playing cards has popped up over the weekend, with new rumours of a GTX 1170 doing the rounds. I imply, in fact it has, that’s the PC gaming tech world’s favorite pastime proper now: the life and instances of graphics playing cards. Which GPUs can be found, how a lot they price, who they’re sleeping with, what’s going to the next-gen appear like, and when are they going to reveal all in a tabloid exposé?

If you want a brand new GPU proper now, choose one of many best graphics cards accessible at present.

The new GTX 1170 specs have come from the ever-reliable WCCFTech, and appears to be fashioned from as a lot hypothesis and guesswork because the GTX 1180 ‘leak’ that appeared to get everybody all excited final week. What the WCCFTech piece is actually saying is that the GTX 1170 goes to be… a graphics card. And they then typed out some numbers plucked from the air above their keyboard.

They don’t appear to have any actual concrete data, however are nonetheless betting on the Nvidia Turing hearsay delivering a 12nm GT104 GPU that measures round 400mm2 and packs in 2,688 CUDA cores. 

All we know for sure about the next-gen GPUs is they'll have green writing

It’ll even have someplace between 8GB and 16GB of 12Gbps GDDR6 reminiscence (actually, at these costs?!), some form of free guess at a imprecise base frequency, which has principally cribbed from the present Pascal playing cards, and the identical for increase clock pace. Oh, and a TDP of between 140 and 160W. My, there’s loads of margin for error constructed into these leaks…

WCCFTech, in fact, can attribute no supply to those numbers, simply that they have been supplied particularly to them. Someone as irked by this baseless hypothesis as I is a brand new poster on the Overclockers UK forum with the name Nv Gfx Prgmr (appears 100% legit).

They rail in opposition to such hearsay mill websites, and say that the supposedly leaked specs are fully false. Good for him.

He does then go and damage all of it together with his personal baseless hypothesis, claiming to know that there shall be no new 12nm gaming playing cards launched this yr and that “Nvidia’s new gaming GPUs will be on 7nm and their current release window is set for late November.” 

As a lot as that’s both blatant under-the-bridge trollism, or rumour-mongering of the very best order, it’s nonetheless not essentially past the realms of risk for Nvidia to skip 12nm in favour of 7nm for his or her subsequent era of gaming graphics playing cards. Especially as, by the tip of the yr, they may have basically skipped a whole era of gaming GPUs by holding onto Pascal for therefore very lengthy.

It might sound implausible, however hear me out.

Nvidia Titan X Pascal

TSMC, the seemingly producers of Nvidia’s graphics silicon, have already began volume production of 7nm processors, as C. C. Wei, TSMC President instructed monetary analysts final month, “More than 50 products tape-outs has [sic] been planned by end of this year,” he says, “from applications across mobile, server CPU, network processor, gaming, GPU, PGA, cryptocurrency, automotive and AI. Our 7nm is already in volume production.”

Now, we all know that AMD have already mentioned they’re utilizing TSMC’s 7nm design for his or her upcoming product strains, splitting them up with GlobalFoundries, and it appears seemingly that their 7nm Vega GPUs shall be coming from TSMC, with this yr’s 7nm Zen 2 samples anticipated to be constructed by GloFo. 

So 7nm is already very a lot ‘a thing’ and with a lot time spent on the forefront of graphics expertise Nvidia aren’t going to need to sit too far behind the competitors. To be truthful, AMD have already mentioned their 7nm GPU isn’t going to be became a high-volume gaming card, and potentially never will, however that’s extra due to its inextricable ties to the ultra-expensive HBM2. As younger, and as costly, as GDDR6 is as an rising graphics reminiscence tech it’s unlikely to have fairly the prohibitive prices of HBM2.  

It’s not going to be the super-exciting EUV-toting 7nm silicon that Intel, Samsung, GlobalFoundries, and TSMC are introducing within the subsequent few years, however it’ll nonetheless supply a 60% drop in energy consumption, and a possible 70% drop in die dimension.

But if AMD are displaying off any 7nm graphics silicon this yr, for machine studying or in any other case, then Jen-Hsun may not be too completely happy releasing a model new gaming vary on a lithography that’s very clearly last-gen by that time. 

Hubris could be a harmful factor, nonetheless, as can pairing a brand new graphics structure with a model new manufacturing lithography – simply take a look at the new mess that was the GTX 480. And Nvidia are unlikely to need to lead with a brand new structure on a brand new course of node with a gaming GPU as an alternative of a lower-volume, professional-class graphics card.

However, if the following era of GeForce GPU is closely primarily based on the present Nvidia Volta structure – as is broadly anticipated, no matter title it will get given – then it gained’t actually be a brand new design. By that point Volta shall be a much more mature GPU design, and the engineers might nicely have a bead on simply how nicely it might translate all the way down to a smaller, high-volume 7nm node.

As a lot because it’s all baseless hypothesis proper now (sorry, Nv Gfx Prgmr), skipping a short-lived 12nm GPU, in favour of a 7nm design that might final one other two years or extra, does make loads of sense. Will they do it? Right now, who is aware of. Realistically, it is unlikely, the timing is probably too tight, and it is in all probability extra of a raffle than Nvidia must make proper now. But sadly we’re in all probability not going to search out out for positive till August/September time on the earliest. Gamescom perhaps?


 
Source

Read also