Wednesday, November 20th 2019
NVIDIA Readying GeForce RTX 2080 Ti SUPER After All?
NVIDIA could launch a "GeForce RTX 2080 Ti Super" after all, if a tweet from kopite7kimi, an enthusiast with a fairly high hit-rate with NVIDIA rumors is to be believed. The purported SKU could be faster than the RTX 2080 Ti, and yet be somehow differentiated from the TITAN RTX. For starters, NVIDIA could enable all 4,608 CUDA cores, 576 tensor cores, and 72 RT cores, along with 288 TMUs and 96 ROPs. Compared to the current RTX 2080 Ti, the Super could get faster 16 Gbps GDDR6 memory.
It's possible that NVIDIA won't change the 352-bit memory bus width or 11 GB memory amount, as those would be the only things stopping the card from cannibalizing the TITAN RTX, which has the chip's full 384-bit memory bus width, and 24 GB of memory. Interestingly, at 16 Gbps with a 352-bit memory bus width, the RTX 2080 Ti Super would have 704 GB/s of memory bandwidth, which is higher than the 672 GB/s of the TITAN RTX, with its 14 Gbps memory clock. These design choices would ensure NVIDIA has a sufficiently faster product than the RTX 2080 Ti, without an increase in BOM, provided it has enough perfectly-functional "TU102" inventory to go around. There's no word on availability, although WCCFTech predicts a CES 2020 unveiling.
Sources:
kopite7kimi (Twitter), WCCFTech
It's possible that NVIDIA won't change the 352-bit memory bus width or 11 GB memory amount, as those would be the only things stopping the card from cannibalizing the TITAN RTX, which has the chip's full 384-bit memory bus width, and 24 GB of memory. Interestingly, at 16 Gbps with a 352-bit memory bus width, the RTX 2080 Ti Super would have 704 GB/s of memory bandwidth, which is higher than the 672 GB/s of the TITAN RTX, with its 14 Gbps memory clock. These design choices would ensure NVIDIA has a sufficiently faster product than the RTX 2080 Ti, without an increase in BOM, provided it has enough perfectly-functional "TU102" inventory to go around. There's no word on availability, although WCCFTech predicts a CES 2020 unveiling.
139 Comments on NVIDIA Readying GeForce RTX 2080 Ti SUPER After All?
Anyway, this is pretty ridiculous... I'm done trying to clarify the written word for people. .. JFC. :(
Also, HD 5800/6900 weren't very much slower than Nvidia's GTX 400/500 counterparts, again the aggressive pricing made AMD the winner IMO in those times too. I liked when they released their new cards in about similar times, it was interesting to wait which one had the faster cards. Now we see that Nvidia has the lead (in both performance and releasing a hella fast card), and AMD just can't compete in the enthusiast range like in the old days.
Its called marketing and you're still looking at GCN with another round of tweaks.Headlines. But if they do this, it does kinda makes you wonder how far Ampere is out still. It could be another 780ti... but then the 780 was also cost effective and 2080ti is far from it. Its going to take major price drops to make one or both products even remotely interesting, this late in a gen.
The upper midrange is way too crowded, and the top end is 100% priced out of the market, while the new top end is too late to matter what with next gen's RT and all. And the value of ALL these cards is going to drop unusually sharply with next gen, because we've slowed down on the current gen.
They need to get +30% to even remotely stay competitive. With NV dropping to 7mm, they should be able to get a healthy boost. They already have the power advantage so likely most of the efficiency/clock improvement s are going to perf/clocks.
I imagine Big Navi's coming in June after being announced in January. Probably in tandem with the launch of Zen3-based CPU's. nVidia announces/limited releases a SUPER version of 2080 Ti in January. Then they announce Ampere in March with a datacenter/not-gaming product to get people used to the idea. Then a gaming version is announced in June/July preceded by steady May leaks to rain on Big Navi's launch. I imagine those with a limited release in Sept/Oct with Founder's Editions.
Truth is, that if you have the money do it, never mind anything else.
But you said that you don't care what we think about your posts. Because "that's how forums work".
So, while some of your posts may be true, I have no reason to assume that. You could be making everything up. Or just some of it. Or having fun. Or whatever.Just slightly more "unaffordable" than regular 2080Ti and Nvidia has no problem selling those.
Top gaming components are getting more and more expensive. That's it.
This means gamers with big budgets can finally buy sensible products that actually give them some gaming potential. Today you can spend $5000 on a consumer PC buying just gaming-oriented parts.
Few years age gamers with this kind of budget were buying HEDT parts and often ended up unhappy (because there was hardly any gain or it turned out 7700K gives more fps).
On the other hand, low-level parts are just as affordable as ever. So we're not losing anything.
Nothing has changed about hedt either as gaming should never have been the focus of an hedt build. You may get lucky that an hedt build did well in gaming but clocks were/are (when comparing the same arch) king.
I am not fond of the pricing but it is what it is. Take it or leave it. I think we'll find that AMD is not going to save anything except their bank account. They are going to slot in with the new pricing structure NV established. That was already on display with 5700 series.
Few years ago you just couldn't go that far.
In mid-2015 you got top fps with a 6700K + 980Ti. That's ~$1000 total (today top CPU and GPU are ~$1000 each).
RAM was cheaper. High-end Z170 mobo was $350. It was hard to justify spending over $3000 on a gaming PC (fps-wise).
So gamers who wanted to spend more went for HEDT (that made hardly any difference) or SLI (that seldom worked, but usually just generated problems).
Basically, gaming component makers decided that high-paying client segment is large enough to make these expensive parts sellable.