Nvidia announces GeForce RTX 40 series GPUs based on Ada Lovelace architecture

Status
Not open for further replies.

Shawn Knight

Posts: 14,470   +171
Staff member
Highly anticipated: Nvidia has officially unveiled its next-gen GeForce RTX 40 series GPUs based on the new Ada Lovelace architecture and headlined by the flagship RTX 4090. Ada Lovelace is built on TSMC's 4N process and will debut in new RTX 4090 and RTX 4080 graphics cards. The former packs 76 billion transistors, 16,384 CUDA cores, and 24GB of GDDR6X memory from Micron.

According to Nvidia, an RTX 4090 with DLSS 3 is up to four times faster than an RTX 3090 Ti running DLSS 2, and efficiency gains mean it sticks with the same 450W power consumption as the RTX 3090 Ti.

DLSS 3 is Nvidia's latest Deep Learning Super Sampling neural-graphics technology. It is built on DLSS Super Resolution, and adds Optical Multi Frame Generation to create entirely new frames to significantly boost frame rates.

Nvidia has a deeper dive available on its website, but you should know that the tech is exclusive to RTX 40 series GPUs and is coming to some of the most popular game engines including Unreal Engine and Unity. More than 35 games and apps are already in the pipeline with the first coming in October.

Nvidia has announced two flavors of the GeForce RTX 4080 GPU. The RTX 4080 12GB features 7,680 CUDA cores alongside 12GB of Micron GDDR6X memory and with DLSS 3 enabled (vs DLSS 2), it's claimed to be faster than the RTX 3090 Ti -- it will set you back $899.

A higher-spec version with 9,728 CUDA cores and 16GB of GDDR6X memory will also be available, albeit for a bit more coin at a hefty $1,199.

Some feature and architectural highlights include:

  • Streaming multiprocessors with up to 83 teraflops of shader power — 2x over the previous generation.
  • Third-generation RT Cores with up to 191 effective ray-tracing teraflops — 2.8x over the previous generation.
  • Fourth-generation Tensor Cores with up to 1.32 Tensor petaflops — 5x over the previous generation using FP8 acceleration.
  • Shader Execution Reordering (SER) that improves execution efficiency by rescheduling shading workloads on the fly to better utilize the GPU resources. As significant an innovation as out-of-order execution was for CPUs, SER improves ray-tracing performance up to 3x and in-game frame rates by up to 25%.

  • Ada Optical Flow Accelerator with 2x faster performance allows DLSS 3 to predict movement in a scene, enabling the neural network to boost frame rates while maintaining image quality.
  • Architectural improvements tightly coupled with custom TSMC 4N process technology results in an up to 2x leap in power efficiency.
  • Dual Nvidia Encoders (NVENC) cut export times by up to half and feature AV1 support. The NVENC AV1 encode is being adopted by OBS, Blackmagic Design DaVinci Resolve, Discord and more.

The flagship RTX 4090 launches on October 12 starting at $1,599, with the RTX 4080 12GB and 16GB variants landing sometime in November from $899 and $1,199, respectively. All will be available from add-in partners including Asus, Gigabyte, MSI and Zotac, in both stock-clocked and factory overclocked variants.

Permalink to story.

 

godrilla

Posts: 583   +316
Nvidia's website shows power requirements for the 4090 being 850 watt psu minimum and can connect to the new PCIe5 power cable ( 450 or 600watt) or traditional 8 pin x3 or 4 power cables with adapter. FYI.

Only has support for DisplayPort 1.4a
 

Shawn Knight

Posts: 14,470   +171
Staff member
Nvidia's website shows power requirements for the 4090 being 850 watt psu minimum and can connect to the new PCIe5 power cable ( 450 or 600watt) or traditional 8 pin x3 or 4 power cables with adapter. FYI.

Only has support for DisplayPort 1.4a

Correct, 850W recommended for the total system and 450W for the GPU alone.
 

EdmondRC

Posts: 402   +580
These prices are in the stratosphere, it would be madness to buy any of these things day one. I predict Nvidia will be forced to reprice due to competitive pressures from AMD early next year. Maybe not, maybe AMD overprices their cards too, but with the MCM design, I would not be surprised if AMD severely undercuts Nvidia here, keeping their pricing structure close to the RDNA 2 structure. Even for the 12GB 3080, that's a price increase of 30% MSRP
 

DSirius

Posts: 368   +773
TechSpot Elite
I watched video "live" if we can call live a fake Jensen avatar presenting Ada Lovelace video card.
For me, it looked like Jensen tried too hard, building the hype for what may be under delivered at launch, especially the price and power consumption.
1599 $ is a paper launch price for 4090. Not buying these crap at all. Real price will be 2000$ at least for AIB customs models
Also, I found strange that Seasonic and EVGA (the best power supply manufacturers) are not among ATX3.0 power suppliers mentioned.
Especially EVGA :)
(ATX 3.0 power supplies will be available in October from ASUS, Cooler Master, FSP, Gigabyte, iBuyPower, MSI, and ThermalTake, with more models to come.)
 
Last edited:

EdmondRC

Posts: 402   +580
Actually it's not just ram the chip is superior ad103 vs ad104 on the 12 gig vram one. Some are saying the 12 gig 4080 should have been a 4070. But a desperate Nvidia will default to its shady practices like 3.5 gigs on the gtx 970.
Right, the 4080 12 GB is a ploy so that it doesn't look like they raised the price as much from 3080 to 4080. The real 4080 16GB is $1200, more than a 60% increase from the 3080 and the 12GB is still a 30% increase. I'm sure there will still be a 4070, but expect it to be at least $600 if not $700. This was even worse than I could have imagined, hard to get excited about new tech that you are being priced out of it. It's not that I can't afford these cards, but I have other priorities and I can't justify that kind of expense for a gaming hobby. I'm not sure where Nvidia is going to get their sales from considering mining is dead for now at least.
 
Last edited:

Fulljack

Posts: 103   +95
I really want to move to all-AMD build, but Nvidia doesn't make it easy with how well their graphics card are.

I know RT doesn't improves graphics that much when compared to well-built global illumination and cast shadow using modern rasterization technique, but it's still an improvement regardless.

Ampere are already faster 1.5 to 2x than RDNA2 in RT performance and now this? AMD needs to step up their game more, hopefully RDNA3 gives us a robust improvement in RT department.

that way, the consumers win with how fierce the competitions are.
 

waclark

Posts: 707   +451
That price jump for additional RAM is ridiculous.

My understanding is that there are more cores as well.

When getting a 40 series is nearly impossible, watch 30 series prices go right back up.
No kidding, It seems like it may already be happening. I'm looking at a 3080 and prices are holding pretty stable. If I can't get a 3080 12G or Ti for $600-650 USD I'll just get the 4080. For $100 over current prices you'll get 2x performance.
 

NeoMorpheus

Posts: 1,394   +2,962
Unlike many here, I will wait until after AMD releases their GPUs AND proper unbiased reviews are posted.

Meanwhile, lets enjoy the posts that even without proper reviews, the loyal ones are already posting, praising the gpu and already in line to purchase these gpus.
 
Last edited:

waclark

Posts: 707   +451
Right, the 4080 12 GB is a ploy so that it doesn't look like they raised the price as much from 3080 to 4080. The real 4080 16GB is $1200, more than a 60% increase from the 3080 and the 12GB is still a 30% increase. I'm sure there will still be a 4070, but expect it to be at least $600 if not $700. This was even worse than I could have imagined, hard to get excited about new tech that you are being priced out of it. It's not that I can't afford these cards, but I have other priorities and I can't justify that kind of expense for a gaming hobby. I'm not sure where Nvidia is going to get their sales from considering mining is dead for now at least.
Well, to be fair, the original 3080 only had 10G and was listed at $699. The 12G version had a much inflated MSRP, closer to $1,100-1,200. Current pricing of the 12G variant is sub-$799 right now. so for $100-200 more you're getting 2x performance. That's not horrible. Whether we should be paying $699 or $899 for a GPU is certainly a topic for discussion.

I guess we will wait for AMD to see where they land. Nividia could be having a 40 series sale even before they're available.
 

loki1944

Posts: 638   +474
"In fact, our new GeForce RTX 2080 and 2080 Ti graphics cards are so fast, you can reach 60 FPS at 4K, with max settings, in most of today’s games, giving you a super smooth, super detailed, super hi-res experience that can’t be found anywhere else."

I remember those lies
 

moon982

Posts: 106   +30
What is going on with the price increases.

Looking at wikipedia.

The 3080 is $699 and the 3080 12GB is $799 and 3080 Ti is $1199

When you look at the 4080 it is $899 for 12 GB and $1199 for 4080 16 GB.

When the 4070 come out next year it going to cost more than the 3080.

Where the 3070 is $499. When the 4070 come out next year it be lucky if it cost $700

All the price are really going up a lot this year.
 
Status
Not open for further replies.