Wednesday, June 12th 2019
NVIDIA's SUPER Tease Rumored to Translate Into an Entire Lineup Shift Upwards for Turing
NVIDIA's SUPER teaser hasn't crystallized into something physical as of now, but we know it's coming - NVIDIA themselves saw to it that our (singularly) collective minds would be buzzing about what that teaser meant, looking to steal some thunder from AMD's E3 showing. Now, that teaser seems to be coalescing into something amongst the industry: an entire lineup upgrade for Turing products, with NVIDIA pulling their chips up one rung of the performance chair across their entire lineup.
Apparently, NVIDIA will be looking to increase performance across the board, by shuffling their chips in a downward manner whilst keeping the current pricing structure. This means that NVIDIA's TU106 chip, which powered their RTX 2070 graphics card, will now be powering the RTX 2060 SUPER (with a reported core count of 2176 CUDA cores). The TU104 chip, which power the current RTX 2080, will in the meantime be powering the SUPER version of the RTX 2070 (a reported 2560 CUDA cores are expected to be onboard), and the TU102 chip which powered their top-of-the-line RTX 2080 Ti will be brought down to the RTX 2080 SUPER (specs place this at 8 GB GDDR6 VRAM and 3072 CUDA cores). This carves the way for an even more powerful SKU in the RTX 2080 Ti SUPER, which should be launched at a later date. Salty waters say the RTX 2080 Ti SUPER will feature and unlocked chip which could be allowed to convert up to 300 W into graphics horsepower, so that's something to keep an eye - and a power meter on - for sure. Less defined talks suggest that NVIDIA will be introducing an RTX 2070 Ti SUPER equivalent with a new chip as well.
This means that NVIDIA will be increasing performance by an entire tier across their Turing lineup, thus bringing improved RTX performance to lower pricing brackets than could be achieved with their original 20-series lineup. Industry sources (independently verified) have put it forward that NVIDIA plans to announce - and perhaps introduce - some of its SUPER GPUs as soon as next week.
Should these new SKUs dethrone NVIDIA's current Turing series from their current pricing positions, and increase performance across the board, AMD's Navi may find themselves thrown into a chaotic market that they were never meant to be in - the RT 5700 XT for $449 features performance that's on par or slightly higher than NVIDIA's current RTX 2070 chip, but the SUPER version seems to pack in just enough more cores to offset that performance difference and then some, whilst also offering raytracing.
Granted, NVIDIA's TU104 chip powering the RTX 2080 does feature a grand 545 mm² area, whilst AMD's RT 5700 XT makes do with less than half that at 251 mm² - barring different wafer pricing for the newer 7 nm technology employed by AMD's Navi, this means that AMD's dies are cheaper to produce than NVIDIA's, and a price correction for AMD's lineup should be pretty straightforward whilst allowing AMD to keep healthy margins.
Sources:
WCCFTech, Videocardz
Apparently, NVIDIA will be looking to increase performance across the board, by shuffling their chips in a downward manner whilst keeping the current pricing structure. This means that NVIDIA's TU106 chip, which powered their RTX 2070 graphics card, will now be powering the RTX 2060 SUPER (with a reported core count of 2176 CUDA cores). The TU104 chip, which power the current RTX 2080, will in the meantime be powering the SUPER version of the RTX 2070 (a reported 2560 CUDA cores are expected to be onboard), and the TU102 chip which powered their top-of-the-line RTX 2080 Ti will be brought down to the RTX 2080 SUPER (specs place this at 8 GB GDDR6 VRAM and 3072 CUDA cores). This carves the way for an even more powerful SKU in the RTX 2080 Ti SUPER, which should be launched at a later date. Salty waters say the RTX 2080 Ti SUPER will feature and unlocked chip which could be allowed to convert up to 300 W into graphics horsepower, so that's something to keep an eye - and a power meter on - for sure. Less defined talks suggest that NVIDIA will be introducing an RTX 2070 Ti SUPER equivalent with a new chip as well.
Should these new SKUs dethrone NVIDIA's current Turing series from their current pricing positions, and increase performance across the board, AMD's Navi may find themselves thrown into a chaotic market that they were never meant to be in - the RT 5700 XT for $449 features performance that's on par or slightly higher than NVIDIA's current RTX 2070 chip, but the SUPER version seems to pack in just enough more cores to offset that performance difference and then some, whilst also offering raytracing.
126 Comments on NVIDIA's SUPER Tease Rumored to Translate Into an Entire Lineup Shift Upwards for Turing
Videocardz is also based on Wccftech and Vonguru which they are using each other as sources.
When Videocardz publishes something it's no longer a rumor :laugh: but i agree we will probably know more when Nvidia is ready to talk .
RTX 2060: 1920 CUDA Cores (TU106)
RTX 2060 SUPER: 2176 CUDA cores (TU106), +13%
RTX 2070: 2304 CUDA cores (TU106), +6%
RTX 2070 SUPER: 2560 CUDA cores (TU104), +11%
RTX 2080: 2944 CUDA cores (TU104), +15%
RTX 2080 SUPER: 3072 CUDA cores (TU102), +4%
RTX 2080 TI: 4352 CUDA cores (TU102), +41%
From the looks of this, the 2060 SUPER will come close to matching 2070 performance (at least when OC'd), but the rest are nowhere near their higher-tier counterparts. If the reported 2080 SUPER specs are correct, there's still a 41% gap between it and the 2080 Ti. If this was a "$500 price cut" there would need to be equivalent performance between the 2080 SUPER and Ti. There definitely won't be, particularly if it still has a 256-bit RAM bus.Partly, sure, but remember how gargantuan Turing dice are. The 2070 SUPER will move from a 445mm2 die to a 545mm2 die, and the 2080 SUPER from 545mm2 to 754mm2. Even if the 12nm process is relatively cheap, that's not an insignificant increase, and will force them to sell cut-down chips rather than fully enabled ones - which I can't believe is an actual problem given the maturity of the process. Sure, they probably have stockpiles of dud chips that they haven't been able to use for fully enabled SKUs up until now, but at some point this move is going to start costing them significantly. If they need to sell a TU104 off the line as a 2080 SUPER rather than a 2080 Ti, that's $300-400 less money per sale. Even subtracting a bit from the lower BOM for the card, this is significant. I doubt Nvidia cares, but it's a move that shows that they're feeling some pressure.
I better not see you moan about price then eh.
If the 2070 Super comes in at 2560 cores then that is still 15% less cores than the present 2080. Assuming it will have a higher clock speed that will help but anyone can OC a 2080 also. How much the 2070 Super can OC remains to be seen.
My guess when all is said and done a 2070 Super OC will still be at least around 10% slower than a 2080 FE OC.
www.guru3d.com/news-story/rumor-nvidia-super-is-a-refreshed-geforce-rtx-2060-(249)2070-(399)2080-(599).html
I sorta of feel cheated out of the extra 256 shader cores, and 2 GB GDDR6 (which now is likely a 256-bit bus.)
Many claim to want competition, yet they bash Nvidia (or Intel) every time they refresh their lineup.
It used to be regarded as a good thing when a company refreshed their lineup, it's a natural thing as yields improve and production costs go down.
Did AMD screw their customers when they refreshed RX 480 with RX 580 or R9 290X with R9 390X?
E.g: look at how everybody reacted to Turing pricing and then look at the reactions then Navi comes in the same performance ballpark at the same exact prices. One year later.
Just dont eat anything Nvidia throws at you, like "our cards cost too much, that's why they are expensive" or the reason why they didn't move to 7nm. They are just dishonest and not only not telling the truth but they are obviously and intentionally lying. (Yes there is a huge difference between keeping silent and not saying anything and between lying) Nvidia didn't move to 7nm because there were no NEED for it (yet). That is the only and the only reason for that.
It's amazing how we got to the point of criticizing them for supposedly going to release better and faster cards!
Basically if they could deliver the concept as Ryzen chips.
Manufacturing smaller GPUs and stacking them together it always gonna be cheaper.
Putting two Navi chips with an interconnect would be no different from connecting Nvidia GPUs with NVLink, and would have to be operated in the same way we do multi-GPU today.
Videocardz has finally released some specs and it turns out 2080 Super won't use TU102 chip just full TU104 and so on for the rest wich is a bit disapointing this is really the bare minimum Nvidia could do but it's not like they really need to do more anyways . Hopefully vanilla 20 series will get at least a good pricecut .
Nothing to see here for 1080Ti owners , next year is when the exciting stuff comes out .
GPUs are supposed to devalue fast, if they do not, it means progress is stalling. That is what we've been looking at post-Pascal to present... 1080ti performance still sells for about MSRP.
This SUPER feels a bit like a Kepler refresh to me. Not a bad thing IMO, more cards = lower prices.
I bought a X1800XT and low, the X1900XT displaced it and devalued it so it lost a couple hundred bucks in value for resale.
Did it still work? Yes.
Was it worth the cost? No.
It's about consumer value, if I buy a premium product I would expect it to be valued as a premium product.
Just my opinion.
You may have had to wait 1-2 years to see a "real" refresh or 3 years for a new thing from Nvidia. While AMD were refreshing actually because they weren't able to make new cards because they had to focus on CPU segment and on console graphics AND this mining thing just saved AMDs ass in recent years or they'd had very tough period.