After months of extortionate graphics card prices and a general feeling of malaise toward today’s current crop of GPUs, many of us have been looking forward to seeing what Nvidia have in store for their next set of GeForce GTX graphics cards, which are currently going by the code name ‘Turing’. Set to replace many of today’s best graphics cards like the GTX 1080 and GTX 1070, the current thinking is that Nvidia’s Turing GPUs will either become a new 11-series of graphics cards, taking on names like the GTX 1180, or even a new 20-series under monikers like the GTX 2070.
However, while all previous information pointed toward a vague autumn release date, Nvidia’s CEO Jensen Huang has now thrown everything into doubt, saying it will be “a long time” before we see the next GeForce GPU. What does that mean for Turing? Here’s everything we know so far.
Nvidia Turing: GTX 1180 release date
Ever since the GTX 1180 rumour mill whirred to life earlier this year and started churning through the daily grind of rumour and speculation, the release date for Nvidia’s Turing family has been in a constant state of flux. Originally, TweakTown reported that we’d have them by April, then we heard from Tom’s Hardware that Turing’s release had been delayed until mid-June, before DigiTimes shot back about a week and a half later saying they’d been delayed again due to revised production schedules and that we now won’t see them until the autumn.
All that seemed quite plausible for a time, but events of recent days have put the kibosh on almost everything. It started on June 1, when eagle-eyed conference program checkers spotted that Nvidia was due to give a talk at the upcoming Hot Chips conference about their “next mainstream GPU” this August.
Even if Nvidia didn’t actually launch Turing at that point, at least it looked like we’d get some proper concrete details on what to expect, and it would have fitted in with the general idea of them releasing the cards sometime this autumn.
On June 5, however, Nvidia’s own CEO Jensen Huang threw everything up in the air when he said we wouldn’t see the next GeForce GPU for “a long time” at this year’s Computex. He wouldn’t be drawn on exactly how long “a long time” actually meant, but it certainly doesn’t sound like Turing will be here in just a couple of months time.
Immediately after he answered that question, Huang was then asked about whether he had anything to say about that aforementioned Hot Chips talk, which in the course of four days has since been changed to just ‘TBD’, with all mention of Nvidia scrubbed from existence despite the company remaining one of the event’s platinum sponsors – to which Huang simply replied we should all “live in the present,” and that if they ever had anything to announce or talk about, journalists would be invited to a dedicated event to hear the news.
So we don’t really know what’s going on with Turing’s release date at the moment. When it’s come from the CEO’s own mouth, I’d say our previous estimate of a fuzzy autumn release is looking incredibly unlikely, and I’d even hazard a guess that we may not even see them until next year now – despite previous rumours stating that the new Turing cards were set to go into mass production by September at the very latest and a supposed Founders Edition of the GTX 1180 was due to arrive this July.
I will, of course, update this article accordingly if I hear anything to the contrary, but for now, anyone who was thinking of waiting for Turing before they bought a new graphics card may want to reconsider their upgrade plans…
Nvidia Turing: GTX 1180 specs
According to information obtained by completely unknown means by Wccftech, the Nvidia GeForce GTX 1180 may well bear a closer resemblance to today’s GTX 1080Ti than the GTX 1080. Powered by a GPU currently known as the GT104, the GTX 1180 will allegedly have a whopping 3584 CUDA cores, a 256-bit GDDR6 memory interface and between 8-16GB of GDDR6 memory, clocked at 16Gbps.
If true, this would mean the GTX 1180’s memory is significantly faster than either the 8GB of GDDR5X memory currently in the GTX 1080 or any known form of HBM2 memory (2nd Gen high-bandwidth memory). This in turn would give the GTX 1180 a huge memory bandwidth of 512GB/s. For comparison’s sake, the GTX 1080 only offered 320GB/s, while the GTX 1080Ti provides 484GB/s.
The graphics card itself, on the other hand, would be clocked somewhere in the region of 1.6GHz, with a max boost clock of around 1.8GHz. The thermal design power (TDP), meanwhile, is expected to be somewhere between 170-200W.
That’s what Wccftech says, but until Nvidia confirm anything themselves, all this should be taken with a huge pinch of salt. Still, if these specs prove accurate, it would mean the GTX 1180 would offer around 1.5x the performance of the current GTX 1080, which seems like a sensible target to aim for considering it’s been two years since the GTX 1080 first came out.
Nvidia Turing: GTX 1180 price
Those very same leaked specs also contain a potential price for the GTX 1180 as well: a rather eye-watering $699. That’s $100 more than the GTX 1080’s launch price, but given recent fluctuations in graphics card prices (even if stock levels are finally returning to normal), that may well actually be a good thing rather than bad.
The cost of having GDDR6 memory will no doubt be a large part of that purported price increase, especially if Nvidia end up stretching to the reported full 16GB, but it could just be Nvidia wanting a larger piece of their respective GPU pie.
Nvidia Turing: GTX 1180 or GTX 2080, and what about Volta?
Hold on a second, I hear you cry. Wasn’t Volta meant to be Nvidia’s next GPU architecture? Yes, for a long time it looked as though Volta would indeed be the successor to Nvidia’s consumer line-up of 10-series Pascal cards. It’s already been deployed in high-end cards like the ludicrous Titan V, but that theory seems to have gone out of the window now in favour of the current code name, Turing.
For a while, it looked as though they might have been called Ampere as well, with Volta being reserved for Nvidia’s top-end cards alone. Now it looks like Turing will be the name of Nvidia’s new consumer cards, while Ampere will be confined to new business-orientated cards.
That still doesn’t help in trying to figure out what model names they’ll actually use, though. While the leaked specs above appear to suggest Nvidia’s Turing cards will be known as the 11-series, there have also been rumours flying around that Nvidia will jump straight to 20 instead. That’s according to TweakTown, which would suggest they’ll be known as the GTX 2070 and GTX 2080, to give just two examples, rather than the GTX 1170 or GTX 1180.
This rumour started with the so-called April release date speculation, as it was hotly tipped that Nvidia would reveal the so-called GTX 2070 and GTX 2080 at their GTC conference at the end of March. Well, no such announcement materialised at GTC, so it may well be the 20-series name is a load of bobbins as well.
If Nvidia did jump straight to 20, presumably then jumping to 30 afterwards and then 40 etc, then it wouldn’t leave them with nearly as many numbers for subsequent generations as it would going up in single digits. Of course, tech companies are renowned for changing their naming conventions as and when it suits them, so may Nvidia will adopt 20 after all. No one really knows at this point, but 11 probably/maybe is a bit more sensible right now.