Wednesday, September 2nd 2020

NVIDIA Announces GeForce Ampere RTX 3000 Series Graphics Cards: Over 10000 CUDA Cores

NVIDIA just announced its new generation GeForce "Ampere" graphics card series. The company is taking a top-to-down approach with this generation, much like "Turing," by launching its two top-end products, the GeForce RTX 3090 24 GB, and the GeForce RTX 3080 10 GB graphics cards. Both cards are based on the 8 nm "GA102" silicon. Join us as we live blog the pre-recorded stream by NVIDIA, hosted by CEO Jen-Hsun Huang.

Update 16:04 UTC: Fortnite gets RTX support. NVIDIA demoed an upcoming update to Fortnite that adds DLSS 2.0, ambient occlusion, and ray-traced shadows and reflections. Coming soon.
Update 16:06 UTC: NVIDIA Reflex technology works to reduce e-sports game latency. Without elaborating, NVIDIA spoke of a feature that works to reduce input and display latencies "by up to 50%". The first supported games will be Valorant, Apex Legends, Call of Duty Warzone, Destiny 2 and Fortnite—in September.
Update 16:07 UTC: Announcing NVIDIA G-SYNC eSports Displays—a 360 Hz IPS dual-driver panel that launches through various monitor partners in this fall. The display has a built-in NVIDIA Reflex precision latency analyzer.
Update 16:07 UTC: NVIDIA Broadcast is a brand new app available in September that is a turnkey solution to enhance video and audio streaming taking advantage of the AI capabilities of GeForce RTX. It makes it easy to filter and improve your video, add AI-based backgrounds (static or animated), and builds on RTX Voice to filter out background noise from audio.
Update 16:10 UTC: Ansel evolves into Omniverse Machinima, an asset exchange that helps independent content creators to use game assets to create movies. Think fan-fiction Star Trek episodes using Star Trek Online assets. Beta in October.
Update 16:15 UTC: Updates to the AI tensor cores and RT cores. In addition to more numbers of RT- and tensor cores, the 2nd generation RT cores and 3rd generation tensor cores offer higher IPC. Making ray-tracing have as little performance impact as possible appears to be an engineering goal with Ampere.
Update 16:18 UTC: Ampere 2nd Gen RTX technology. Traditional shaders are up by 270%, raytracing units are 1.7x faster and the tensor cores bring a 2.7x speedup.
Update 16:19 UTC: Here it is! Samsung 8 nm and Micron GDDR6X memory. The announcement of Samsung and 8 nm came out of nowhere, as we were widely expecting TSMC 7 nm. Apparently NVIDIA will use Samsung for its Ampere client-graphics silicon, and TSMC for lower volume A100 professional-level scalar processors.
Update 16:20 UTC: Ampere has almost twice the performance per Watt compared to Turing!
Update 16:21 UTC: Marbles 2nd Gen demo is jaw-dropping! NVIDIA demonstrated it at 1440p 30 Hz, or 4x the workload of first-gen Marbles (720p 30 Hz).
Update 16:23 UTC: Cyberpunk 2077 is playing big on the next generation. NVIDIA is banking extensively on the game to highlight the advantages of Ampere. The 200 GB game could absorb gamers for weeks or months on end.
Update 16:24 UTC: New RTX IO technology accelerates the storage sub-system for gaming. This works in tandem with the new Microsoft DirectStorage technology, which is the Windows API version of the Xbox Velocity Architecture, that's able to directly pull resources from disk into the GPU. It requires for game engines to support the technology. The tech promises a 100x throughput increase, and significant reductions in CPU utilization. It's timely as PCIe gen 4 SSDs are on the anvil.

Update 16:26 UTC: Here it is, the GeForce RTX 3080, 10 GB GDDR6X, running at 19 Gbps, 238 tensor TFLOPs, 58 RT TFLOPs, 18 power phases.
Update 16:29 UTC: Airflow design. 90 W more cooling performance than Turing FE cooler.
Update 16:30 UTC: Performance leap, $700. 2x as fast as RTX 2080, available September 17. Up to 2x faster than the original RTX 2070.
Update 17:05 UTC: GDDR6X was purpose-developed by NVIDIA and Micron Technology, which could be an exclusive vendor of these chips to NVIDIA. These chips use the new PAM4 encoding scheme to significantly increase data-rates over GDDR6. On the RTX 3090, the chips tick at 19.5 Gbps (data rates), with memory bandwidths approaching 940 GB/s.
Update 16:31 UTC: RTX 3070, $500, faster than RTX 2080 Ti, 60% faster than RTX 2070, available in October. 20 shader TFLOPs, 40 RT TFLOPs, 163 tensor cores, 8 GB GDDR6
Update 16:33 UTC: Call of Duty: Black Ops Cold War is RTX-on.

Update 16:35 UTC: RTX 3090 is the new TITAN. Twice as fast as RTX 2080 Ti, 24 GB GDDR6X. The Giant Ampere. A BFGPU, $1500 available from September 24. It is designed to power 60 fps at 8K resolution, up to 50% faster than Titan RTX.

Update 16:43 UTC: Wow, I want one. On paper, the RTX 3090 is the kind of card I want to upgrade my monitor for. Not sure if a GPU ever had that impact.
Update 16:59 UTC: Insane CUDA core counts, 2-3x increase generation-over-generation. You won't believe these.
Update 17:01 UTC: GeForce RTX 3090 in the details. Over Ten Thousand CUDA cores!
Update 17:02 UTC: GeForce RTX 3080 details. More insane specs.

Update 17:03 UTC: The GeForce RTX 3070 has more CUDA cores than a TITAN RTX. And it's $500. Really wish these cards came out in March. 2020 would've been a lot better.
Here's a list of the top 10 Ampere features.

Update 19:22 UTC: For a limited time, gamers who purchase a new GeForce RTX 30 Series GPU or system will receive a PC digital download of Watch Dogs: Legion and a one-year subscription to the NVIDIA GeForce NOW cloud gaming service.

Update 19:47 UTC: All Turing cards support HDMI 2.1. The increased bandwidth provided by HDMI 2.1 allows, for the first time, a single cable connection to 8K HDR TVs for ultra-high-resolution gaming. Also supported is AV1 video decode.

Update 20:06 UTC: Added the complete NVIDIA presentation slide deck at the end of this post.

Update Sep 2nd: We received following info from NVIDIA regarding international pricing:
  • UK: RTX 3070: GBP 469, RTX 3080: GBP 649, RTX 3090: GBP 1399
  • Europe: RTX 3070: EUR 499, RTX 3080: EUR 699, RTX 3090: EUR 1499 (this might vary a bit depending on local VAT)
  • Australia: RTX 3070: AUD 809, RTX 3080: AUD 1139, RTX 3090: AUD 2429
Add your own comment

502 Comments on NVIDIA Announces GeForce Ampere RTX 3000 Series Graphics Cards: Over 10000 CUDA Cores

#176
Fluffmeister
nguyen
Jup, RTX 3080 will straight up murdering 2080 Ti, 40-50% faster than 2080 Ti and 80% faster than RTX 2080 just in rasterization

Edited: @Fluffmeister just posted the DigitalFoundry link like 5s faster than I did :D
Yeah just beat ya too it! Sorry fella! But yeah, the 3080 smacks the 2080 pretty hard.
Posted on Reply
#177
Nkd
nguyen
Jup, RTX 3080 will straight up murdering 2080 Ti, 40-50% faster than 2080 Ti and 80% faster than RTX 2080 just in rasterization

Edited: @Fluffmeister just posted the DigitalFoundry link like 5s faster than I did :D
Well looks like Navi is going to be just competing with 3080 from rasterization standpoint looking at specs. I don’t see them tryin got compete with 3090 because math doesn’t add up unless Ofcourse they have more cores all of sudden then rumored. Plus not sure if amd really wants to compete at 1000+. Just not something they shoot for.
Posted on Reply
#178
Vayra86
Fluffmeister

Gets juicy from about 6 minutes in, compares a cranked 2080 to a 3080.

Go to 6.16 to be precise and well.... gee.
I think its always best to take a worst case, much salt scenario on these comparisons because you're not usually looking at the toughest scenes games will throw at you.

Realistically I think what 3080 will give you, is +75% over a 1080ti / 2070S. Because let's face it, 2080 wasn't much of its own tier.

Which is STILL pretty damn decent at 700. I have to say. Its also good to see that the x80 pulls up the absolute performance already. That's a lot better than Turing's 2080ti doing it.
Posted on Reply
#179
PowerPC
Fluffmeister

Gets juicy from about 6 minutes in, compares a cranked 2080 to a 3080.

Go to 6.16 to be precise and well.... gee.
So Doom Eternal performance of 3080 vs 2080 is increased by a whooping 81% without RTX because that game doesn't even support RTX afaik.

And my prediction a couple of weeks ago was around 60%, which was actually made fun of on this forum hahaha. Even 60% would still have been a lot actually and much more than at least two of the last increases of generations.
Posted on Reply
#180
randompeep
Out there wishing people would calm their asses down on VRAM battle. Like the heck you do with dozens and dozens?! Isn't this the hardware you'd be supposed to game the next 3-5 years like you do with a next-gen console, huh ? How many of y'all are gonna do it in 4k 120 Hz or 8k (just to call it out) ?
And for the VRAM 'horny' peeps, go hunt the used market! Big deals are already on globally
Posted on Reply
#181
Zubasa
medi01
So far so good for the "two times faster" promise, lol.
Ampere got more than double the CUDA cores, but the rest of the card (ROPs / Bandwidth etc) didn't get this kind of increase.
So the performance checks out in the end.
Vayra86
I think its always best to take a worst case, much salt scenario on these comparisons because you're not usually looking at the toughest scenes games will throw at you.

Realistically I think what 3080 will give you, is +75% over a 1080ti / 2070S. Because let's face it, 2080 wasn't much of its own tier.

Which is STILL pretty damn decent at 700. I have to say. Its also good to see that the x80 pulls up the absolute performance already. That's a lot better than Turing's 2080ti doing it.
Yeah 70-ish% increase is like Maxwell vs Pascal which is very nice, finally we get a "real" generational increase.
Posted on Reply
#182
Chrispy_
medi01
So
kopite7kimi
was right about pricing.

According to him/her Big Navi soundly beats 3070 (rendering 3070Ti useless).
This means we have:

3070 << Big Navi < 3080

It also explains the unexpectedly conservative pricing.
big navi is still a ways off though, right?
Posted on Reply
#183
mrthanhnguyen
Dont forget the real benchmark is not from Nvidia slide.
Posted on Reply
#184
chodaboy19
Wow the performance increase is truly brutal across the board!

After some more driver and game optimization, the performance will just keep going up!
Posted on Reply
#185
Chomiq
Imouto
That's one of the points, though. Consoles are obsolete even before launch.
100 mil + of PS4 sold says otherwise.
sutyi
No Ti branding this time around supposedly. I think with the super models they'll double the VRAM on higher SKUs.


3080S 20GB / 3070S 16GB model what you should keep an eye out if they are forced by AMD.
They have a big price gap between 3080 and 3090. They can hold out until flagship RDNA2 and then drop 3080 Ti to fit this scenario.
Posted on Reply
#186
Vayra86
Nkd
Well looks like Navi is going to be just competing with 3080 from rasterization standpoint looking at specs. I don’t see them tryin got compete with 3090 because math doesn’t add up unless Ofcourse they have more cores all of sudden then rumored. Plus not sure if amd really wants to compete at 1000+. Just not something they shoot for.
I think Navi is going to struggle catching this 3080 to be honest. AMD has yet to surpass 2070S performance convincingly, and now they're making an 80% jump ahead? Not likely, unless they make something absolutely gargantuan. But let's not dive into the next pond of speculation... my heart... :p

By the by, do we have TDPs for these Ampere releases already? The real numbers?
Posted on Reply
#187
HD64G
raptori
The " twice as fast" or 3070 is faster than 2080Ti , is it the general performance ? or RT performance only ? or we don't know yet ...
That increase in performance is clearly the FPS with RTX ON.
Posted on Reply
#188
RedelZaVedno
randompeep
Out there wishing people would calm their asses down on VRAM battle. Like the heck you do with dozens and dozens?! Isn't this the hardware you'd be supposed to game the next 3-5 years like you do with a next-gen console, huh ? How many of y'all are gonna do it in 4k 120 Hz or 8k (just to call it out) ?
And for the VRAM 'horny' peeps, go hunt the used market! Big deals are already on globally
Just look at Microsoft FS 2020. 2080TI manages only 31FPS (with deeps to 21) over NY City at 4K/ULTRA and wants to use 12.7GB of VRAM. 3080 will hopefully get us to 55 fps (with deeps to 40ies). Having 16GB would probably smoothen these fps deeps further. Needless to say 3080 is still a godsent for flight simmers :)
Posted on Reply
#189
sutyi
Chomiq
100 mil + of PS4 sold says otherwise.

They have a big price gap between 3080 and 3090. They can hold out until flagship RDNA2 and then drop 3080 Ti to fit this scenario.
Most leaky info points to AIB partners pushing towards a more transparent naming scheme for the product stack. Super had a better a ring to it.

I'm not doubting they'll release a something between the 3080 and the 3090 down the line, but 80% it will not be called a 3080Ti.
It will probably be called a 3080 Super and it will have 20GB memory and a 350W TGP. They'll release it around 899-999 when they have enough cut down cores.
Posted on Reply
#190
efikkan
medi01
According to him/her Big Navi soundly beats 3070 (rendering 3070Ti useless).
This means we have:
3070 << Big Navi < 3080
It also explains the unexpectedly conservative pricing.
That's a pretty bold statement when nobody have compared them in real workloads yet.

Don't get me wrong, I hope Big Navi™ will push the prices a bit down further.
Posted on Reply
#191
Icon Charlie
HD64G
That increase in performance is clearly the FPS with RTX ON.
I agree.

I want to see standard testing from several different sites on a large amount of games and programs is when judgement should be placed.

Starting at 1080p. and go up from there. I want to see real performance gains without RTX on.
Posted on Reply
#192
PowerPC
With the VRAM discussion, you also have to mention that the 30x0 cards all have 60% faster VRAM than the previous generation (it makes the process of freeing up and writing to Memory much faster) and therefore will probably lead to a much smaller VRAM bottleneck than Turing had if you even encounter this as a problem, which most games so far don't even on 4K. 8-10 GB is still pretty good. But they'll probably release Ti variants of all these cards with increased VRAM, but also a much higher prices.
Posted on Reply
#193
Chomiq
sutyi
Most leaky info points to AIB partners pushing towards a more transparent naming scheme for the product stack. Super had a better a ring to it.

I'm not doubting they'll release a something between the 3080 and the 3090 down the line, but 80% it will not be called a 3080Ti.
It will probably be called a 3080 Super and it will have 20GB memory and a 350W TGP. They'll release it around 899-999 when they have enough cut down cores.
Thing is, Super is perfectly fine if you're doing a refresh. Ti is helpful if you need something quick to one up the competition. They can do Super refresh in 12 months, but Ti would be a direct response if RDNA2 actually managed to match 3080 in non-RT performance.
Posted on Reply
#194
Zubasa
chodaboy19
Wow the performance increase is truly brutal across the board!

After some more driver and game optimization, the performance will just keep going up!
You might want to tone down the excitement a bit, generally nVidia GPUs don't tend to gain a lot of performance unless there is some glaring bug in the driver.
nVidia is generally pretty good at getting most of the performance out of their cards from the get go.
For example, the relative performance between Pascal and Turing more or less stayed the same in most cases.
Posted on Reply
#195
medi01
efikkan
That's a pretty bold statement when nobody have compared them in real workloads yet.
How NV/AMD "know" what is being cooked in the other kitchen always was a mystery to me (and I wondered how much of it could be blocked on driver level).

Nevertheless, 3080 and below pricing is... not usual, is it?
Posted on Reply
#196
Metroid
3080 because is the best cost benefit price/performance gpu, I want more gddr memory, 10gb is not enough, I like Nvidia cooling design however I will have to wait for reviews and then decide which 3080 to buy, 3080 nvidia reference cards will only come with 10gb, so need to wait and see which company will pack more gddr and have the best cooling solution.

3070 = $499
3080 = $699
3090 = $1499

3090 is too expensive for the performance it will give, I mean is better to buy 2 x 3080 and put on sli, I dont think the 3090 will be even 50% faster than the 3080.

Also nvidia made wonders, I mean 2080 is 12nm and 3080 is 8nm, in that sense we should expect only 50% faster and yet is a lot more than that. If we take 1080 which is a 14nm and compare x 3080 8nm which is not even a 7nm, the performance will be much more than 100% and for that I'm happy.
Posted on Reply
#197
Zubasa
medi01
How NV/AMD "know" what is being cooked in the other kitchen always was a mystery to me (and I wondered how much of it could be blocked on driver level).

Nevertheless, 3080 and below pricing is... not usual, is it?
Depends on how you think of it, Turing's pricing was "unusual" in the bad way. :laugh:
TBH, I was kind of expecting nVidia to raise the price on those as well.
Posted on Reply
#198
medi01
HD64G
That increase in performance is clearly the FPS with RTX ON.
On one hand there is that 2+ times CU cores, on other, lower clocks and lower number of transistors per CU (???).

If increasing number of CUs 2.5 times only brings 1.5 times better perf, I'd call it disappointing.
Zubasa
Depends on how you think of it, Turing's pricing was "unusual" in the bad way. :laugh:
ASP price bump started when AMD Raja times, which is Pascal.
Posted on Reply
#199
SIGSEGV
I knew that 3090 would be more costly. Well, it is very obvious that the main target of this card is for intermediate or even beginner researchers and professionals, especially for whom they use CUDA libraries for having fun with deep learning. Well, with 3090 hefty prices, I prefer to rent google cloud AI to serve my passion.
I really hope AMD has a secret recipe to counter these ampere GPU lineups.
Posted on Reply
#200
medi01
Metroid
the best cost benefit price/performance gpu
It actually is 3070 (which one could safely say without even watching anything) but still, wench for baitmarks, it shouldn't take long for actual tests from places not known to be NV's bentover stops to roll them out.
Posted on Reply
Add your own comment