Nvidia has not released any information on its upcoming RTX 40 series graphics cards, but one thing is certain: Nvidia will release next-gen GPUs at some point. To bring you up to speed, we’ve rounded up all the rumblings around the virtual water cooler on RTX 40 series graphics cards, when they are coming and how they will perform.
Before we dive in, it’s important to note that Nvidia hasn’t announced anything regarding the RTX 40 series cards, also known as Lovelace. The company has not even confirmed their existence. All of the information below is based on rumors, leaks and speculation. We also have previous launches to compare, so we’ve only included rumors that seem likely based on previous launches.
Pricing and availability
Leak Greymon55 says the next-gen GPUs will arrive in October 2022. That includes the RTX 40 series and AMD’s upcoming RDNA 3 graphics cards. Although Nvidia hasn’t confirmed anything, September or October 2022 makes sense for the upcoming lineup. Series 30 RTX cards launched in September 2020 and RTX series 20 cards launched in September 2018.
The next-gen GPU will be in October of next year, in about a year.
– Greymon55 (@ greymon55) September 7, 2021
Even though October is not the right month, all signs point to a launch in the second half of 2022. Nvidia has maintained a two-year release cadence for the past few generations. It’s safe to assume we’ll see the cards next year, likely with an announcement in September and a launch in October.
Additionally, Nvidia is also expected to release an intergenerational update in early 2022. The 30-series RTX Super graphics card is rumored to launch in the early months of the year, with a possible announcement at CES 2022. . It’s always like that. than a rumor, of course.
As for the price of the RTX 40 series, we don’t know at this time. Nvidia has maintained its pricing by moving from the RTX 20 series to the RTX 30 series. Here’s where the current lineup is at right now:
- RTX 3090: $ 1,499
- RTX 3080 Ti: $ 1,199
- RTX 3080: $ 699
- RTX 3070 Ti: $ 599
- RTX 3070: $ 499
- RTX 3060 Ti: $ 399
- RTX 3060: $ 329
We could see the same prices this time around, although we could see a slight increase. By upgrading from the GTX 1080 to the RTX 2080, Nvidia increased the price by $ 100. Considering the current GPU pricing situation, the suggested price doesn’t mean much. Hopefully this is not the case when the RTX 40 series cards launch.
The same leaker who pushed back the release date also said that the RTX 40 series graphics cards will be built on chipmaker TSMC’s N5 node. This builds on earlier rumors that Nvidia uses a 5nm node, which makes sense. Currently, RTX 30 series graphics cards use Samsung’s 8nm process and AMD’s RX 6000 cards use TSMC 7nm process.
A smaller manufacturing process means Nvidia is able to squeeze more cores onto the same GPU chip. The rumor mill suggests that the new die will be able to hold 18,432 CUDA cores, nearly 8,000 more than the RTX 3090. This is a huge leap, suggesting that the RTX 40 series cards could see the same improvement. performance than the RTX 30 series.
There are a few caveats to this, however, which we’ll explore in more detail in the sections to come. Based on rumors, TSMC’s N5 node seems likely, but it is possible that the cards could use the N5P node as well. This is an improved version of the 5nm process. Instead of straight down jumps, TSMC uses improved processes between each step, so 7nm goes straight to 5nm with improved process in between, and so on.
Nvidia would use a monolithic design for RTX 40 series cards, while future cards from AMD would use a multi-chip module (MCM). Nvidia would switch to an MCM design with its Hopper graphics cards, which are rumored to launch after Lovelace.
We haven’t heard anything about ray tracing or Tensor Hearts, but it’s safe to assume they’ll appear on Lovelace. RT cores handle ray tracing in real time, while Tensor cores are responsible for the AI computations required for Nvidia’s Deep Learning Super Sampling (DLSS). These are tentpole features of the Nvidia graphics cards, so we would expect them to appear on the RTX 40 series cards.
The RTX 40 series cards are at least a year away, so we don’t know much about performance yet. Considering the number of cores assumed, a card with the full die could offer up to a 76% increase over the RTX 3090. These are only theoretical performance, however. Many other factors apply, so don’t expect a 76% increase in the real world.
One thing is for sure, we will get a huge improvement in performance including power consumption. Maybe we can get 100tflops fp32 in one card.
– kopite7kimi (@ kopite7kimi) May 22, 2021
Leaker Kopite7kimi, who has become a trusted source for GPU leaks, speculated that the cards might be capable of up to 100 teraflops of FP32 computing power. It is important to reiterate that this figure is purely speculative. The leak manager was not making the request based on the information he had.
Overall, rumors suggest a 2.2x increase over the current GPU core. Yet, it is too early to tell. Even though the design is finalized right now, Nvidia still needs to tweak the drivers and optimize the cards to run the latest games. We don’t know where the performance will be when the cards launch, but the rumor circuit suggests it could be a big improvement.
A significant increase in performance is often accompanied by an increase in power demands. And rumors suggest that the RTX 40 series cards could be more power hungry than any previous version from Nvidia. Various leakers have confirmed that the cards will require a lot of energy, although the exact number is still pending.
400W appears to be the minimum for the most powerful card, although some leaks suggest it could require up to 500W. For context, the RTX 3090 requires 350W of power, and Nvidia recommends at least a block of power. 750 W power supply. This is the minimum recommendation, so you should have 800 W or more power supply for the RTX 3090.
If the top card pushes towards 500W, you will probably need a power supply of 1000W or more. Nvidia broke the 250W cap with its 30-series RTX graphics cards. Previously, even the most powerful consumer cards were below 250W. The RTX 3080, RTX 3080 Ti, and RTX 3090 all require over 300W, which suggests that Nvidia isn’t too concerned about the power limits of its high-end models.
Nvidia hasn’t confirmed anything, and it’s possible that power consumption will decrease when the design is finalized. Still, it looks like manufacturers may need to invest in a new power supply for a high-end card for the next generation.
- Other Rumors Suggest Nvidia Lovelace AKA RTX 4000 Series Will Use TSMC N5
- New leak reveals that Nvidia is working on GeForce RTX 3090 Super
- Upgraded Nvidia GeForce RTX 3060 could arrive in early 2022
- Rumors are circulating about a monstrous GeForce RTX 3090 Super on the way
- Nvidia could tackle GPU shortages by turning to RTX 2060 again – but with an upgrade
- Here’s where to buy Nvidia GeForce RTX 3070
- Intel’s first Alchemist graphics card could fight Nvidia GeForce RTX 3070
- Next-gen GPUs could launch in October 2022 – both by AMD and Nvidia
- Best Buy will have Nvidia RTX 30 Series cards in stock in select stores tomorrow