NVIDIA GeForce RTX 2080 Ti Super

Someone posted that in the 3xxx series thread. A 2080ti super is a very boring card... maybe 6% faster than a 2080ti.

If that happens it makes you wonder if nVidia misstepped with Ampere or if it’s an aggressive design.

Personally I’d be happy with RT add in cards. My 2080ti has more than enough rasterized performance.
 
Unless nVidia drops the 2080 Ti non-Super down I only see the true enthusiasts wanting the maximum performance possible going for the 2080 Ti Super, if it is real.
 
wonder what the pricing will be for that 6% optimal gain
 
After a solid year of production I would expect that the 2080 TI Super is just a 2080TI but their processes have cleaned up so they are able to squeese out an extra 100 MHZ on the clock and have paired it with slightly faster memory, they will keep the price the same and just discontinue the 2080 TI. There is too little competition up there to fill the space with 2 of your own products.
 
Only if they reduce the price as well. The article claims it is targeted at people who still have 1080Tis and don't feel the need to upgrade. That would be me, the 2080Ti was too much money for the improvement offered so I kept the 1080Ti. If the new super card drops down to $800, or maybe $900 then ya I'll probably pick one up. I would have gotten at 2080Ti at $800. However if it stays over $1k? No thanks, I'll keep what I have. New high end GPUs are a fun luxury I like to waste money on, but there are limits to how much I'll waste :p
 
Only if they reduce the price as well. The article claims it is targeted at people who still have 1080Tis and don't feel the need to upgrade. That would be me, the 2080Ti was too much money for the improvement offered so I kept the 1080Ti. If the new super card drops down to $800, or maybe $900 then ya I'll probably pick one up. I would have gotten at 2080Ti at $800. However if it stays over $1k? No thanks, I'll keep what I have. New high end GPUs are a fun luxury I like to waste money on, but there are limits to how much I'll waste :p
Eh I will still hang on to my 1080ti. It is no slouch. Since I would need to buy a new water block I will hold out til a 3xxx maybe even 4xxx ti cards.
 
Eh I will still hang on to my 1080ti. It is no slouch. Since I would need to buy a new water block I will hold out til a 3xxx maybe even 4xxx ti cards.

For sure, I have no need to upgrade at all, I just like new toys. I was all ready to buy a 2080Ti when it launched, thinking it would launch for the same price as the 1080Ti. When I saw the price I said "nah".
 
Why are you sorry? Wcc is legit. They repost 90% of other sites.

Wtf is wrong with [H] and the hatred if wcc?

It's like calling Drudgereport fake news when Drudge doesnt post news. He post links to other news sites.

In that case the Entire Hard Tech news section is fake news because it's all links to other news sites articles.
Except this is editorial, not reposting anything. There is no where in the article posted that points to an original source for the speculation, just recursive links to their own articles to back up the author's assertions. WCCFTech is not simply a news aggregator.

The one source I see near the end of the article is arguing for some reason that Ampere will be on Samsung's 5nm, when we all know already that it will be on 7nm. Samsung's 7nm taped out at the end of 2018 when they started risk production. The length of time between then and the release window opening for Ampere seems spot on with putting the products into production and getting the scale to have a product launch.

NVIDIA will team up with Samsung for 7nm GPU technology
Samsung undercuts TSMC, to fab NVIDIA Ampere on its 7nm process
Big win for Samsung as NVIDIA picks its 7nm process for next-gen GPU
7nm NVIDIA GPUs could come in 2020 through partnership with Samsung... will sample the new silicon to NVIDIA and other partners by end of [2019].

Nothing has come out in the ensuing months but from non-descript tweets from people claiming to be industry insiders, so there is no reason at this time to believe that anything has changed unless NVIDIA was not satisfied from the 7nm tapeout.
 
Except this is editorial, not reposting anything. There is no where in the article posted that points to an original source for the speculation, just recursive links to their own articles to back up the author's assertions. WCCFTech is not simply a news aggregator.

The one source I see near the end of the article is arguing for some reason that Ampere will be on Samsung's 5nm, when we all know already that it will be on 7nm. Samsung's 7nm taped out at the end of 2018 when they started risk production. The length of time between then and the release window opening for Ampere seems spot on with putting the products into production and getting the scale to have a product launch.

NVIDIA will team up with Samsung for 7nm GPU technology
Samsung undercuts TSMC, to fab NVIDIA Ampere on its 7nm process
Big win for Samsung as NVIDIA picks its 7nm process for next-gen GPU
7nm NVIDIA GPUs could come in 2020 through partnership with Samsung... will sample the new silicon to NVIDIA and other partners by end of [2019].

Nothing has come out in the ensuing months but from non-descript tweets from people claiming to be industry insiders, so there is no reason at this time to believe that anything has changed unless NVIDIA was not satisfied from the 7nm tapeout.

Ok fair argument
 
2080 Ti Super this late in the cycle wouldn't make much sense. AMD is getting ready to release big Navi so nVidia can't afford to screw around and delay ampere.
 
2080ti Super this late in the game eh. Jensen really knows what his customers like.
 
This smells really bad to me. I would be very surprised to see this happen at this stage in the game. The TAM at that price point is so small there is no reason to launch something like this. It would only telegraph internal engineering issues if NV did, and I don't think it would want to do that. I just do not see a win in terms of marketing or business. Anyway, my quick 2 cents, you may need change.
 
This smells really bad to me. I would be very surprised to see this happen at this stage in the game. The TAM at that price point is so small there is no reason to launch something like this. It would only telegraph internal engineering issues if NV did, and I don't think it would want to do that. I just do not see a win in terms of marketing or business. Anyway, my quick 2 cents, you may need change.


You may be spot on sir. No supers are doing the space invaders.
 
the only way I see 2080Ti Super happening is if Big Navi matches or beats the regular 2080TI and drops more than 3 months ahead of Ampere. Then NV releases a very slightly cut-down Titan RTX with 12GB GDDR6 as 2080Ti Super as a stopgap to spoil AMDs launch. If Big Navi can't match 2080Ti and/or launches within a couple months of Ampere, then no Ti Super.
 
I think my previous estimates of 'big navi' were that if it happens, would be May 2020 timeframe, maybe June. Nvidia doesn't have to release anything until then. Or beat AMD to market by 2 weeks with something faster, followed by AMD price cuts 2 days before their product launches..
 
Last edited:
What, in case AMD bucks their eternal trend of releasing something slower, less efficient, and two years late?

#waitforbignavi

Normally I'd agree with that sentiment but big Navi seems ready to roll out since the DXR portion of the architecture is already in PS5 and Xbox. I think AMD is finally beginning to execute a little more aggressively which should hopefully bring some much needed competition at the high end. I won't go near a $1000 gpu again, Titan XP was my last one.
 
wonder what the pricing will be for that 6% optimal gain

The Titan RTX has 20% more gpu power than the 2080ti due to more cores and higher clocks. Bandwidth is only 10% higher, but if the new Ti Super uses faster 2080S memory, it could be 20% faster as well to ensure a full 20% boost in performance which is nothing to sneeze at.
 
The Titan RTX has 20% more gpu power than the 2080ti due to more cores and higher clocks. Bandwidth is only 10% higher, but if the new Ti Super uses faster 2080S memory, it could be 20% faster as well to ensure a full 20% boost in performance which is nothing to sneeze at.
Where are you getting that the Titan RTX is 20% faster than the 2080 TI though? The big RTX cards have abysmal scaling compared to the previous 1080 TI and 980ti over the 1080 and 980 respectively. Even at 4k, the 2080 TI is only 18% faster than the 2080 Super despite having over 40% more cores and a massive bandwidth advantage.

https://www.techpowerup.com/review/nvidia-geforce-rtx-2080-super-founders-edition/27.html
 
  • Like
Reactions: N4CR
like this
Meh, I'll wait for the 2080 Ti Super Duper or maybe even the 2080 Ti Super-Deduper.

They should just launch a 2085 since they did a GTX 280 refreshed 285 back in the day, but Nvidia can't have a consistent naming scheme like that. :p
 
Where are you getting that the Titan RTX is 20% faster than the 2080 TI though? The big RTX cards have abysmal scaling compared to the previous 1080 TI and 980ti over the 1080 and 980 respectively. Even at 4k, the 2080 TI is only 18% faster than the 2080 Super despite having over 40% more cores and a massive bandwidth advantage.

https://www.techpowerup.com/review/nvidia-geforce-rtx-2080-super-founders-edition/27.html

I only used the Titan to compare the theoretical gflops for now.

The ti might have 40% more cores but gpu power is only 20% higher due to the lower clocks. The bandwidth difference isn't massive as well, again due to lower clocks. Its 20% as well which is line with 18% performance boost.

A full tu102 with a 384 bit bus running similiar memory and gpu clocks as the 2080s would be 50% higher in both gpu power and bandwidth compared to the 2080s which is significant. Scaling won't be perfect, but it would see a nice boost over the 2080ti.
 
I only used the Titan to compare the theoretical gflops for now.

The ti might have 40% more cores but gpu power is only 20% higher due to the lower clocks. The bandwidth difference isn't massive as well, again due to lower clocks. Its 20% as well which is line with 18% performance boost.

A full tu102 with a 384 bit bus running similiar memory and gpu clocks as the 2080s would be 50% higher in both gpu power and bandwidth compared to the 2080s which is significant. Scaling won't be perfect, but it would see a nice boost over the 2080ti.
Sorry but I think you are still way off. If you look at the actual real world boost clock not the advertised boost clock there is only just a few percentage difference. The 2080 super in that review has an average boost clock of 1919 compared to the 2080 TI at 1824 if you look at its review. That is only a 5% difference real world boost difference tested by the same site. Even if I give you perfect scaling on that, that would still only make the 2080 TI 23% faster than the 2080 super even at 4K. That is pathetic scaling at 4k for a card with over 40% more cores, nearly 40% more ROPs and higher bandwidth.
 
Last edited:
  • Like
Reactions: N4CR
like this
Normally I'd agree with that sentiment but big Navi seems ready to roll out since the DXR portion of the architecture is already in PS5 and Xbox. I think AMD is finally beginning to execute a little more aggressively which should hopefully bring some much needed competition at the high end. I won't go near a $1000 gpu again, Titan XP was my last one.

Normally I'd agree with that sentiment, except I've experience every AMD GPU release, and well, having little Navi be less efficient without RT really doesn't bode well for Big Navi.
 
Sorry but I think you are still way off. If you look at the actual real world boost clock not the advertised boost clock there is only just a few percentage difference. The 2080 super in that review has an average boost clock of 1919 compared to the 2080 TI at 1824 if you look at its review. That is only a 5% difference real world boost difference tested by the same site. Even if I give you perfect scaling on that, that would still only make the 2080 TI 23% faster than the 2080 super even at 4K. That is pathetic scaling at 4k for a card with over 40% more cores, nearly 40% more ROPs and higher bandwidth.

Given those clocks, the 2080ti should be 35% faster making the 18% core scaling look bad. However, I suspect that the 2080ti is slightly bandwidth starved.
You keep throwing the 40% number around, but you need to factor in cores/bandwidth as well as frequency.
Using clocks in your link, the 2080ti still only has 25% more bandwidth than the 2080s. Of this is the bottleneck, the scaling doesn't look so bad
 
Normally I'd agree with that sentiment, except I've experience every AMD GPU release, and well, having little Navi be less efficient without RT really doesn't bode well for Big Navi.

You probably aren't wrong, but when we're talking about enthusiast class products I'm not really sure things like "efficiency" or "power requirements" really matter all that much. At that level the only thing that really matters is raw performance (and maybe price). If Big Navi if performance competitive with enthusiast class cards then it being less efficient isn't going to matter.
 
makes no sense. unless they released it with HDMI 2.1. then i could see people jumping on it for the nvidia certified 4k TV's to do 120hz. i see all the press releases with LG lately, but then you're like well...unless you are downscaling this thing to a lower res, you can't even take advantage of it. And if Ampere is delayed, maybe a re-release with HDMI 2.1 makes sense? Idk.
 
  • Like
Reactions: mikeo
like this
Back
Top