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

#151
Nkd
AusWolf
Awesome numbers, awesome prices... what's even more awesome is that the number of CUDA cores suggest a Ti version coming out later for each segment.
ti is gone. It’s going to be super refresh from now on. That’s one of the reasons they changedto 3090 instead of 3080 ti at launch.
Posted on Reply
#152
Makaveli
Imouto
RIP AMD.

This time around they won't be able to compete even in price. I see the higher ups at AMD hovering their mouses over emails asking their providers to stop the RDNA2 production.
You have no clue what you are talking about thanks for posting to show everyone else.
Posted on Reply
#153
Zubasa
Berfs1
Actually, 3090 has higher performance/$ than 2080 ti.
And that is somehow something unique for a new generation of GPUs?
We got that in all the generations of GPUs before Turing.
Posted on Reply
#154
Chrispy_
PowerPC
I actually think it's admirable that they aren't ripping people off with too much VRAM. Well, they are doing it with 24GB on the 3090, but you have to be gullible to think you need that much VRAM. That extra amount of VRAM is usually just to inflate the price for no reason.
see my post here
I've already run out of VRAM this year, far bigger games are coming.
Posted on Reply
#156
Raendor
Maybe game devs should learn again how to compress the textures without losing much quality. When games these days weight 100 gigs of which the most part is textures - it’s just stupid.
Posted on Reply
#157
ZoneDymo
Chrispy_
see my post here
I've already run out of VRAM this year, far bigger games are coming.
^ I mean dont get me wrong, a lot is good here, but look at the most recent TPU game review, Horizon Zero Dawn, that game uses over 8gb of Vram on 4k.

And that is today..... I mean here you buy that 500 dollar RTX3070 to get RTX2080Ti performance but it has only 8gb of ram...that is going to be a problem soon I think.
Posted on Reply
#158
Berfs1
Zubasa
And that is somehow something unique for a new generation of GPUs?
We got that in all the generations of GPUs before Turing.
Actually, from my numbers, 600 to 700 series was a (very slight) decrease in average performance/$. 10 to 20 series was a more noticeable average performance/$.
Posted on Reply
#159
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 ...
Posted on Reply
#160
Nkd
lynx29
WTF JUST HAPPENED!!!!!! A 2080 TI FOR $500 OMFG LMAO THIS IS THE GREATEST DAY EVER... HOLY CRAP....


this is unbelievable... Nvidia just came out swinging. I don't see Big Navi matching this I really don't...
hold your horses and wait for reviews. Browsing NVIDIA’s website it seems those numbers are based off rtx and dlss on. Not just pure rasterizarion. NVIDIA’s is just combining all three now I believe.
Posted on Reply
#161
Zubasa
Berfs1
Actually, from my numbers, 600 to 700 series was a (very slight) decrease in average performance/$. 10 to 20 series was a more noticeable average performance/$.
Both of them are Kepler, that generation is pretty much a rebrand for the most part. The GTX 780 / Ti was the only new die of note in the 700-series.
Basically the whole RX400 vs RX500 rubbish.
Posted on Reply
#162
Berfs1
ZoneDymo
^ I mean dont get me wrong, a lot is good here, but look at the most recent TPU game review, Horizon Zero Dawn, that game uses over 8gb of Vram on 4k.

And that is today..... I mean here you buy that 500 dollar RTX3070 to get RTX2080Ti performance but it has only 8gb of ram...that is going to be a problem soon I think.
Okay but you can afford 50$ of ram right? Cus if it goes past the VRAM limit, you can go to the RAM. We had this instance earlier where 8GB and 1080 Ti was actually not a bad idea (at the time), because the 1080 Ti had more VRAM, but if you paired with a 1060 3GB, you definitely needed 16 GB. NOW, RAM is cheaper, only 100$ for 32GB. Just get that, and you will be fine.
Posted on Reply
#163
Vayra86
So much for all the BS of the last few weeks then, huh.

I suppose Nvidia is ready to make their RT push and take a much smaller margin on it. That 3070 is pretty neat at 499, but first... raw performance plx. RT performance is an irrelevant metric.
Posted on Reply
#164
Berfs1
Nkd
hold your horses and wait for reviews. Browsing NVIDIA’s website it seems those numbers are based off rtx and dlss on. Not just pure rasterizarion. NVIDIA’s is just combining all three now I believe.
Not true, the core counts are literally doubled. MORE than doubled.
Posted on Reply
#165
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.
Posted on Reply
#166
Berfs1
Zubasa
Both of them are Kepler, that generation is pretty much a rebrand for the most part.
Basically the whole RX400 vs RX500 rubbish.
Ok congrats... 30 series on average has much lower price/performance. Yay, its better.
Posted on Reply
#167
Vayra86
ZoneDymo
^ I mean dont get me wrong, a lot is good here, but look at the most recent TPU game review, Horizon Zero Dawn, that game uses over 8gb of Vram on 4k.

And that is today..... I mean here you buy that 500 dollar RTX3070 to get RTX2080Ti performance but it has only 8gb of ram...that is going to be a problem soon I think.
I think its clear they try to sell you a 3090 if you REALLY want 4K.

As they should. Its still early adopter territory and you shouldve known better if you thought that 1080ti would carry you in the long run. Not you - royal you.
Posted on Reply
#168
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
Posted on Reply
#169
Chrispy_
Nkd
hold your horses and wait for reviews. Browsing NVIDIA’s website it seems those numbers are based off rtx and dlss on. Not just pure rasterizarion. NVIDIA’s is just combining all three now I believe.
This is what I'd heard in a couple of streams discussing the new RT performance too. Nvidia are going to be pushing DLSS hard this generation.
Posted on Reply
#170
Nkd
chstamos
Welp! I don't have the live stream. Is RTX 3070 supposed to be twice as fast as 2080 Ti without raytracing, too, or only with RT on?

If it's twice the rasterization performance this is actually an amazing new lineup! I was set for disappointment, thought we were going to have another turing debut on our hands, which I had found very underwhelming. This is actually great!
well looking at the website and under performance 3080 is twicea as fast as 2080 super with rtx and dlss on. They show numbers from 3 games.They didn’t really show any game numbers specifically. So looks like their overall performance number is based on rtx and dlss on.
Posted on Reply
#171
PowerPC
Chrispy_
see my post here
I've already run out of VRAM this year, far bigger games are coming.
VRAM is one of the most expensive parts of the GPU right now and so far, the least important when it comes to performance. You can increase the VRAM, but then you'll have to increase the price considerably, just look at 3090... I mean, these cards are for today. I doubt anybody can do anything about VRAM prices. Game devs really need (and probably will) adapt to this reality, unless these prices change. I think you can find flaws with anything, if you want.
Posted on Reply
#172
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.
Posted on Reply
#173
Berfs1
Makaveli
You have no clue what you are talking about thanks for posting to show everyone else.
How sure are you that YOU know what you are talking about? AMD has not been able to compete with the 2080 Ti, except with the *rumored* (and never released) RX 5950 XT. At least Nvidia made something with MORE THAN 2x THEIR CURRENT FLAGSHIP CARD!!
Nkd
well looking at the website and under performance 3080 is twicea as fast as 2080 super with rtx and dlss on. They show numbers from 3 games.They didn’t really show any game numbers specifically. So looks like their overall performance number is based on rtx and dlss on.
They didn't show exact numbers because they can vary. Because not everyone has the exact same CPU in the exact same configuration.
Posted on Reply
#174
RedelZaVedno
Digital Foundry benchmarks 3080 = 65-90% faster than 2080... AWESOME!!!
Posted on Reply
#175
medi01
Mm, did number of transistors per CU go down?
RedelZaVedno
Digital Foundry benchmarks 3080 = 65-90% faster than 2080... AWESOME!!!
So far so good for the "two times faster" promise, lol.
Posted on Reply
Add your own comment