Monday, January 20th 2020

Rumor: NVIDIA's Next Generation GeForce RTX 3080 and RTX 3070 "Ampere" Graphics Cards Detailed

NVIDIA's next-generation of graphics cards codenamed Ampere is set to arrive sometime this year, presumably around GTC 2020 which takes place on March 22nd. Before the CEO of NVIDIA, Jensen Huang officially reveals the specifications of these new GPUs, we have the latest round of rumors coming our way. According to VideoCardz, which cites multiple sources, the die configurations of the upcoming GeForce RTX 3070 and RTX 3080 have been detailed. Using the latest 7 nm manufacturing process from Samsung, this generation of NVIDIA GPU offers a big improvement from the previous generation.

For starters the two dies which have appeared have codenames like GA103 and GA104, standing for RTX 3080 and RTX 3070 respectively. Perhaps the biggest surprise is the Streaming Multiprocessor (SM) count. The smaller GA104 die has as much as 48 SMs, resulting in 3072 CUDA cores, while the bigger, oddly named, GA103 die has as much as 60 SMs that result in 3840 CUDA cores in total. These improvements in SM count should result in a notable performance increase across the board. Alongside the increase in SM count, there is also a new memory bus width. The smaller GA104 die that should end up in RTX 3070 uses a 256-bit memory bus allowing for 8/16 GB of GDDR6 memory, while its bigger brother, the GA103, has a 320-bit wide bus that allows the card to be configured with either 10 or 20 GB of GDDR6 memory. In the images below you can check out the alleged diagrams for yourself and see if this looks fake or not, however, it is recommended to take this rumor with a grain of salt.
Source: VideoCardz
Add your own comment

173 Comments on Rumor: NVIDIA's Next Generation GeForce RTX 3080 and RTX 3070 "Ampere" Graphics Cards Detailed

#26
EarthDog
Otonel88
I think what ever the potential of the Ampere architecture, the performance will be around 10% - 15% improved over current generation.
I highly doubt it. This past generation we saw the least amount of performance increase from flagship to flagship from NV (and that was 25-30%).
Posted on Reply
#27
Xaled
Evet a %100 improvement means nothing, if the price would increase %95 and the price/performance ratio would stay the same
Posted on Reply
#28
Dave65
I'll wait for FAT Navi and then compare with Nvidias offering.
Posted on Reply
#29
Otonel88
EarthDog
I highly doubt it. This past generation we saw the least amount of performance increase from flagship to flagship from NV (and that was 25-30%).
What I mean, is that if the Ampere architecture potential is 50%, the first ampere GPUs will release with an improvement of 10%-15%. Leaving the rest of potential for future cards. I think Nvidia will not give all the potential of Ampere with the next 30xx cards, just so they have something left in the tank for other future cards. Plus they will do architecture improvements in the future potentially getting even more from Ampere.

Xaled
Evet a %100 improvement means nothing, if the price would increase %95 and the price/performance ratio would stay the same
Honestlly I think the price will stay either the same or increase. Don't think it will decrease. We saw nothing but price increases from Nvidia for the past years.
Posted on Reply
#30
EarthDog
Otonel88
What I mean, is that if the Ampere architecture potential is 50%, the first ampere GPUs will release with an improvement of 10%-15%. Leaving the rest of potential for future cards. I think Nvidia will not give all the potential of Ampere with the next 30xx cards, just so they have something left in the tank for other future cards. Plus they will do architecture improvements in the future potentially getting even more from Ampere.
In other words, you are calling these their mid-range cards??? Ik still confused... 10-15% over what?

Also, it has been generations since we have seen architectural changes in mid life cycle for gpu... IPC is what it is.
Posted on Reply
#31
Otonel88
EarthDog
Big Navi is due out at the same time...

... wondering if it will still be 'big' compared to these (or w.e NV's high-end will be) or what...
It will be 'big navi' compared to current generations. Don't think AMD can take Nvidia in the top tier cards just yet. Historically they where always a bit behind for top tier cards.
Posted on Reply
#32
Xaled
EarthDog
I highly doubt it. This past generation we saw the least amount of performance increase from flagship to flagship from NV (and that was 25-30%).

And the RTX 2xxx series came 3 years after. But still Nvidia could've made more improvement in the RTX series but there were no need to and there won't be need in this series as well
Posted on Reply
#33
repman244
Xaled

But still Nvidia could've made more improvement but there were no need to and there won't be need in this series as well
5970 is a dual GPU card, the only valid comparison is the 5870.
Posted on Reply
#34
Otonel88
EarthDog
In other words, you are calling these their mid-range cards??? Ik still confused... 10-15% over what?

Also, it has been generations since we have seen architectural changes in mid life cycle for gpu... IPC is what it is.
10% - 15% over top of the range Nvidia cards like 2080 super or 2080 TI.
If they are running tests and at the moment and the new Ampere generation has an improvement of 50% over the current high end cards then the GPU to come in the next years could have improvements over the current generations as it follows:
3080 (+15% improvement over 2080)
4080 (+30% improvement over 2080) .. and so on.

They will release the performance in batches over the next years cards.

(I am just speculating on percentages, but I guess you got my point)
Posted on Reply
#35
Xaled
repman244
5970 is a dual GPU card, the only valid comparison is the 5870.
Who even mentioned the 5xxx series. I am comparing Nvidia to Nvidia
Posted on Reply
#36
EarthDog
Otonel88
10% - 15% over top of the range Nvidia cards like 2080 super or 2080 TI.
If they are running tests and at the moment and the new Ampere generation has an improvement of 50% over the current high end cards then the GPU to come in the next years could have improvements over the current generations as it follows:
3080 (+15% improvement over 2080)
4080 (+30% improvement over 2080) .. and so on.

They will release the performance in batches over the next years cards.

(I am just speculating on percentages, but I guess you got my point)
Then that isnt midrange. You are confusing. Lol

With Turing, they released high end first and midrange after.

Pre
Posted on Reply
#37
Otonel88
EarthDog
Then that isnt midrange. You are confusing. Lol
I am talking about what Nvidia could do with Amepere over the current top cards from the current architecture
Posted on Reply
#38
EarthDog
Otonel88
I am talking about what Nvidia could do with Amepere over the current top cards from the current architecture
Yep. It's going to do a lot more than that, flagship to flagehip.

Xaled
And the RTX 2xxx series came 3 years after. But still Nvidia could've made more improvement in the RTX series but there were no need to and there won't be need in this series as well
Isnt that the same base arch though but die shrieked? I dont recall. Regardless, least amount of improvement in a loooong while.
Posted on Reply
#39
FordGT90Concept
"I go fast!1!11!1!"
londiste
nm has not been favourable to higher clocks so far but has been considerably better in power consumption and efficiency.
That's at TSMC. This is Samsung.

Judging by the core count, I expect Ampere to be quite different from Turing. Like a Kepler to Maxwell jump in architecture over Pascal to Turing.
Posted on Reply
#40
Xaled
EarthDog
Yep. It's going to do a lot more than that, flagship to flagehip.
I agree it can be up to 50٪ better, maybe even more, but i really doubt that Nvidia would play all of its cards, as there is no need to and the prices would increase in the same amount
Posted on Reply
#41
EarthDog
Xaled
I agree it can be up to 50٪ better, maybe even more, but i really doubt that Nvidia would play all of its cards, as there is no need to and the prices would increase in the same amount
To be clear, I dont think it will be 50% either, but it will surely be more than 10-15%.
Posted on Reply
#42
64K
Otonel88
10% - 15% over top of the range Nvidia cards like 2080 super or 2080 TI.
If they are running tests and at the moment and the new Ampere generation has an improvement of 50% over the current high end cards then the GPU to come in the next years could have improvements over the current generations as it follows:
3080 (+15% improvement over 2080)
4080 (+30% improvement over 2080) .. and so on.

They will release the performance in batches over the next years cards.

(I am just speculating on percentages, but I guess you got my point)
What are you basing this on?

Let's look at past performance increases to get some clarity about what we can probably expect from Ampere with the lower process node and new architecture.

RTX 2080 (non Super) showed an increase of 37% in average performance over the GTX 1080

GTX 1080 showed an increase of 67% in average performance over the GTX 980



Why in the world are you expecting a lousy 15% increase in performance from Ampere over Turing?
Posted on Reply
#43
Xaled
Comparing names is just wrong, you should compare the best GPU to the best one of the previous generation
Posted on Reply
#44
Razbojnik
lynx29
going to be a great year for high end gaming.

I'm leaning towards big navi and ryzen 4700x, but if intel/nvidia high end ends up being 30+% fast for a 200-300 bucks more... than yeah I prob will go green.
Yeah...pity there are no games worthy of such powers, I mean when these cards come an average 250$ card will be able to rock 1440p and ray tracing on ultra in all games...plus games are being delayed like hell lately. we're getting there where there will be no point in buying expensive gpu's for gaming, just like there's no point in buying expensive cpu's for gaming...a 250$ cpu maxes all. I'm not complaining...I guess? Right? Hmm, there is a gap here. Gpu's are getting insanely powerful...while we are getting games of lesser quality. Come to think of it, there's a lot of things to complain about in here.
Posted on Reply
#45
64K
Xaled
Comparing names is just wrong, you should compare the best GPU to the best one of the previous generation
Ok let's look at high end GPUs

The RTX 2080 Ti showed an average increase in performance over the 1080 Ti of 33%

The 1080 Ti showed an average increase in performance over the 980 Ti of 75%

Posted on Reply
#46
theoneandonlymrk
EarthDog
Big Navi is due out at the same time...

... wondering if it will still be 'big' compared to these (or w.e NV's high-end will be) or what...
I think it bigger but not big personally, Navi 21lite is a console gpu , Navi 21 is due soon n thought big, but Navi 23 is alleged the one.

I hope these Nvidia 3070/3080 specs are wrong personally, not that there is much wrong with them per se just i am ever hopeful for some performance doubling to kick in soon so as to make 8k useable and this sounds like a mere step up in performance.

no i am not going to use 8k but expect something that can do 8k to do 4k very well and yes AMD are not doing what i want either before i get trolled.
Posted on Reply
#47
Razbojnik
kapone32
Multi GPU is for anyone that wants it. I have not replaced my Vega 64s with 5700Xts for exactly that reason. If Multi GPU was really dead MBs would not be touting it and giving you 2, 3 and 4 SLI bridges. As much as people complain about it being troublesome it is not as bad as people make it out to be. There are plenty of Games that support Multi GPU anyway. As an example I get 107 FPS average at 4K playing Jedi Fallen Order @ Ultra settings.
For the money you spent...I can't say I'm impressed with those frames...and I can only imagine how awful the stuttering in 4k is lol
Posted on Reply
#48
Otonel88
EarthDog
To be clear, I dont think it will be 50% either, but it will surely be more than 10-15%.
Whatever that percentage will be, we will not get all of that at once for the next high end cards.
Posted on Reply
#49
medi01
64K
RTX 2080 (non Super) showed an increase of 37% in average performance over the GTX 1080
It's sounds like it's more than what it actually is.

Posted on Reply
#50
EarthDog
Otonel88
Whatever that percentage will be, we will not get all of that at once for the next high end cards.
All I am saying is that we will get A LOT more than 10-15% gains out of the box from the new generation when comparing apples to apples (tier to tier). The only differences we've seen the past couple of generations are clock swaps and in the case of the super cards, core/spec differences, in order to extract more performance out of these cards. But 10-15% is putting a 2080 Ti on water and overclocking it... next gen will show double that, at minimum.
Posted on Reply
Add your own comment