Nvidia already makes a number of the finest graphics playing cards, however it’s additionally not resting on its laurels. Though the RTX 40-series, which has been bolstered by a refresh, continues to be very latest, Nvidia can also be engaged on its next-gen GPUs from the RTX 50-series.
The discharge date of RTX 50-series GPUs continues to be at the very least just a few months away, however varied rumors and leaks give us a greater concept of what to anticipate. Right here’s every part we learn about Nvidia’s upcoming technology of graphics playing cards.
RTX 50-series: pricing and launch date
We haven’t heard any specifics from Nvidia concerning the launch date simply but, however most estimates pin the launch of Blackwell across the finish of 2024 and the start of 2025. Some very tentative whispers even point out a doable RTX 50 refresh in 2026, however that’s approach too far into the long run to pay it a lot thoughts.
According to early rumors, Nvidia wasn’t supposed to be ready to launch the new graphics cards until 2025. This would give AMD a major edge, seeing as it’s rumored to launch RDNA 4 GPUs later this year. However, according to YouTuber and frequent leaker Moore’s Law Is Dead, Nvidia may not give AMD the breathing room it so badly needs.
Moore’s Law Is Dead said in a recent video that a source at Nvidia told him that “Blackwell is being prepared to be ready to launch in the fourth quarter of 2024,” but only if Nvidia wants it to. This depends on whether AMD’s RDNA 4 cards will be competitive enough to take away sales from Nvidia during the holiday season at the end of this year, as well as how Ada sales are going around that time.
No matter what, Nvidia is supposedly planning to “make a big deal about RTX 5000 efficiency at CES 2025.” This means that the GPUs are supposedly launching either at the end of 2024 or near the beginning of 2025, but a new report from UDN tells us that we might see a similar launch strategy for Blackwell as we did for Ada.
According to the report, Nvidia is getting ready to launch the RTX 5090 first in the final quarter of 2024, followed by the RTX 5080 a few weeks later. This mirrors the approach in the RTX 40-series, where the RTX 4090 hit the shelves first, followed by the RTX 4080 shortly after. The more budget-friendly graphics cards only came out early the next year, which is what might happen this time around, too. As of now, Nvidia is said to still be deciding the final specs of these two cards, meaning how cut-down the chips will be.
The pricing of these GPUs is pure speculation at this point. In this generation, Nvidia adopted a pricing strategy that can only be referred to as “expensive.” It might follow down that path and push the prices even higher, especially if the demand for AI GPUs remains as high as it is right now. After all, the current demand pushed the RTX 4090 way above $2,000, even though it launched at an already very high price point of $1,600. This certainly makes the RTX 5090 a worrying prospect., but Nvidia’s recent price cut in the RTX 40-series Super refresh gave many enthusiasts some hope.
Assuming the flagship 5090 will cost close to $2,000, the rest of the lineup is, unfortunately, likely to follow with price increases across the board. However, for Nvidia to remain the go-to against AMD, the prices can’t keep rising forever. There is some hope that Nvidia will realize this and keep its pricing more reasonable for the next generation, but it’s too early to tell.
RTX 50-series: specs
Nvidia RTX 50-series | |
Process node | TSMC 3nm or TSMC 5nm (N4P) |
Architecture | Blackwell |
Chip | GB202, GB203, GB205, GB206, GB207 |
Memory type | GDDR7 |
Maximum bus width | 384-bit/512-bit |
Display connectors | DisplayPort 2.1, HDMI 2.1 |
With the release of RTX 50-series GPUs still a while away, Nvidia hasn’t confirmed any specifications for any of the cards. In fact, we’re not even sure which models are coming. However, piecing together speculation from various hardware leakers gives us some idea of what we can expect. Remember to take the following with a healthy dose of skepticism until Nvidia itself spills the beans.
TSMC3
— kopite7kimi (@kopite7kimi) November 15, 2023
We all know for a undeniable fact that the follow-up to Ada Lovelace shall be referred to as Blackwell, honoring American mathematician David Blackwell. Rumor has it that will probably be manufactured by TSMC based mostly on a 3nm course of, however it’s unclear whether or not Nvidia shall be utilizing one in every of TSMC’s present 3nm nodes or a customized node.
The lineup is alleged to incorporate chips spanning from the high-end, RTX 4090-equivalent GB202 by means of the GB203, GB205, GB206, and entry-level GB207. This shall be an fascinating, maybe worrying, change if confirmed true. It might imply that the AD104 GPU powering the RTX 4070 would haven’t any successor within the subsequent technology. The RTX 5070 and RTX 5070 Ti may, due to this fact, make the most of the GB205 chip.
One of the talkative sources of knowledge on the RTX 50-series has been kopite7kimi on X (previously Twitter). The leaker revealed that we will anticipate the brand new GPUs to function help for DisplayPort 2.1, one thing that the Lovelace lineup doesn’t present, and likewise for HDMI 2.1.
I feel my persistence is right. So the distinction is that GB202 is 512-bit and AD102 is 384-bit.
— kopite7kimi (@kopite7kimi) March 11, 2024
Kopite’s newest replace talks concerning the reminiscence interface for Blackwell. The leaker now states that the flagship card will certainly have a 512-bit reminiscence bus, regardless of their earlier statements that it might keep on with 384-bit. The utmost bus width of Blackwell has been a really contentious matter amongst fashionable leakers, so it’s exhausting to know what’s true. Nevertheless, one factor that all of them agree on is that Nvidia will use the brand new GDDR7 reminiscence customary, which AMD is alleged to not be utilizing in its upcoming RDNA 4 lineup.
The leaker additionally up to date the expectations for the velocity of these GDDR7 reminiscence modules discovered within the RTX 50-series. Regardless of earlier rumors that we’d see 32Gb/s modules proper out of the gate, kopite7kimi says that Nvidia will use 28Gb/s for this technology. This nonetheless marks a stable improve over Ada, delivering as much as 1.8TB/s of reminiscence bandwidth on the rumored RTX 5090 — assuming the 512-bit reminiscence bus checks out.
A latest leak from YouTuber RedGamingTech gave us our first glimpse into what the specs might seem like throughout the completely different chips within the lineup. This doesn’t precisely translate to a one-to-one product comparability, although. In reality, the RTX 4090 didn’t even max out its AD102 chip, so it’s very doable that we’d see a repeat of this in Blackwell.
In accordance with the specs shared by RedGamingTech, the flagship chip might include 192 streaming multiprocessors (SMs), up from 144 within the AD102. The GB203 marks an enormous step down, with simply 108 SMs (up from 80 within the AD103). This means that, if the RTX 5090 will get most of these SMs and the RTX 5080 is given the GB203 chip, there’ll be an enormous efficiency hole between the 2 playing cards once more — similar to on this technology.
Extra hypothesis shared by kopite7kimi corroborates this. In accordance with the leaker, the GB203 chip shall be “half of GB202,” marking an analogous drop in efficiency as what we’ve seen within the RTX 4090 versus the RTX 4080. It’s price noting that RedGamingTech, not like kopite7kimi, believes that we’re getting a most bus width of 384 bits.
I feel GB203 is half of GB202, similar to GB102 and GB100. However I don't know if GB202 has a multi chip package deal.
— kopite7kimi (@kopite7kimi) March 11, 2024
The quantity of VRAM in GPUs has been a scorching matter as of late, and to that finish, RedGamingTech speculates that we’d see as much as 36GB of reminiscence within the RTX 5090. Nevertheless, these numbers aren’t finalized, so we’d find yourself with 24GB, like within the RTX 4090.
One final anecdote from the YouTuber states that the GB207, which means the least performant chip within the lineup, will doubtless solely seem in laptops at first. This tracks with what we’ve seen within the RTX 40-series, the place the RTX 4050 has solely made an look in laptops so far.
It’s too early to know the specifics of any particular person card at this level, and all of that is topic to vary. It’s doubtless that Nvidia will launch fashions starting from the RTX 5060 to the RTX 5090, with some Ti choices added into the combination. Let’s hope that it’s going to hold the specs balanced to supply a very good unfold of playing cards for lovers and entry-level customers alike.
RTX 50-series: structure
Nvidia is conserving the structure utilized in Blackwell chips hush-hush, however it gained’t keep that approach for much longer. With the GPUs just a few months away, we’ll be taught extra as the discharge date attracts nearer. In the interim, Nvidia talked concerning the architecture for its data center Blackwell GPUs, which may not be very indicative of what could happen in the consumer lineup — but there are still some interesting tidbits.
The first curious part is that the enterprise version of Blackwell is built on TSMC’s 4NP node, which is actually a 5nm process. Previous rumors indicated that the RTX 50-series might be built on a 3nm process, but that now seems quite unlikely, given the recent announcement. Moreover, the B200 GPU comes with a dedicated decompression engine. While there’s no telling if that will make it to the consumer GPUs, it could bring a major boost to the graphics cards.
Although Nvidia discussed the Blackwell architecture in relation to enterprises, it stayed silent on its consumer lineup. As a result, all we have is more speculation from various sources, but the information is often somewhat conflicting.
RedGamingTech talked about the Blackwell architecture at length in a recent video. The YouTuber referred to it as “one of the most influential graphics architectures,” predicting that the RTX 50-series will introduce significant improvements to things like path tracing and ray tracing, offering gains for both enthusiast-grade and midrange cards.
To that end, the YouTuber said we might see significant architectural changes, including a major redesign of Nvidia’s SMs. He also mentioned the addition of a denoising accelerator, either as a part of the chip or as a function of Nvidia’s Tensor cores. More importantly, RedGamingTech initially teased that Nvidia may use a multi-chip module (MCM) design. This means a design approach where multiple smaller chips are packaged together to form a single, larger, and more powerful processor. Switching to an MCM design over monolithic could give Nvidia a major edge, including scalability, higher yields, and more design flexibility.
Unfortunately, a recent update from the same YouTuber revealed that Nvidia won’t be using an MCM design in Blackwell. Reportedly, Nvidia initially planned to use dual GB202 dies glued together, possibly with some SMs cut, but ultimately decided against it. The YouTuber remarked that issues such as high prices, the latency between the two dies, and various difficulties in getting it to work made Nvidia stick to its previous architecture.
Take this with a healthy dose of skepticism. It’s possible that Nvidia may be planning to switch to MCM in the future, but such architectural changes are never made last minute, so that plan for Blackwell may have never existed. However, it’s also possible that Nvidia may introduce architectural changes instead of pushing for top performance to allow the new technology to mature before ramping up the performance in RTX 6000-series graphics cards a few years from now.
RTX 50-series: Performance
As the specifications of RTX 50-series graphics cards are still mostly a mystery, it’s hard to make accurate predictions about their performance. However, many have tried, which is why we have some juicy rumors to dig into while we wait for official benchmarks.
According to Moore’s Law Is Dead, the performance uplift between Ada and Blackwell may not be major. The YouTuber’s source mentioned that “Blackwell’s rasterization uplift over Ada will not be as impressive as [from] Ampere to Ada.” However, the source also said that Nvidia could make the RTX 5090 feel like a similar uplift “if it felt threatened.” That seems unlikely, seeing as AMD is reportedly stepping down from making high-end GPUs in the next generation, potentially leaving Nvidia as the only source of high-end graphics cards for the next couple of years.
Based on the above, we might be looking at performance gains along the lines of 30% to 50% for the flagship. Midrange and entry-level cards typically see a smaller boost in performance gen-on-gen, so those might be even less impressive.
However, on the other end of the spectrum is speculation from sources like RedGamingTech. The YouTuber claims in his video that we’re looking at an up to 2x increase in performance between Lovelace and Blackwell. He mentioned that the RTX 50-series should double the ray tracing performance compared to the RTX 40-series, as well as provide a performance boost of up to 2x. RedGamingTech is unsure if this means rasterization, though, so it’s hard to know the metric by which to measure these gains. He does, however, predict clock speeds reaching over 3GHz, which would be a sizable boost over Ada, but also says that this only applies to overclocked models.
In a later video, RedGamingTech added that we might see an up to 60% boost from one flagship to the next, meaning a 60% difference in performance between the RTX 4090 and the RTX 5090. The YouTuber also stressed that Nvidia’s focus was heavily on ray tracing and path tracing, with up to a 2.5x boost in those workloads. Again, approach all of this information with some skepticism.
The only real hint of performance figures we have right now comes from a slide made by Nvidia, but unfortunately, the slide talks about its next-gen high-performance computing (HPC) graphics card used in data centers. The graph, which measures GPU performance in GPT-3 175B inference, shows that the H200 GPU will be up to 18 times faster than the A100 — but that’s not Blackwell architecture yet. B100, the first Blackwell graphics card on the list, offers significantly higher performance, although Nvidia didn’t put a number on it. It looks to be about twice as fast as the H200.
While that’s exciting for those in need of an HPC GPU, gamers and other consumers will need to wait to find out the reality about the capabilities of RTX 50-series GPUs.
RTX 50-series: power draw
Prior to the release of the RTX 40-series, the flagship RTX 4090 was the subject of a lot of rumors, and its power draw was an especially hot topic. Some sources claimed that the GPU would have truly monstrous power consumption, even reaching up to 900 watts. We now know that those claims were false, as the RTX 4090 consumes 450 watts, and its connector supports up to 600W — while occasionally melting. It’s hard to imagine that Nvidia will push those numbers even higher in the next generation of GPUs.
The RTX 50-series has been unable to avoid some power-related controversies, though. Moore’s Law Is Dead recently revealed that Nvidia is planning to use a whole new connector, which would mark the fourth such change in a span of just three years. The YouTuber cites anonymous sources, claiming that Nvidia is switching to a 16-pin connector, all dedicated to 12V power delivery. However, many other sources are pointing out that this is unlikely.
Hardware Busters reached out to its own sources and confirmed that “no one is aware of a new connector.” Nvidia would have to be working with major PSU brands, especially after the issues with the 12VHPWR connector. If these brands don’t know anything about it, Nvidia might not be making these changes in this generation yet.
Assuming Nvidia sticks to the (somewhat controversial) 12VHPWR connector that it’s currently using, the maximum power consumption will remain at 600W. The flagship RTX 5090 might go on to see an increase in power draw if it offers significantly more performance, but it’ll still need to leave some room for potential overclocking, so a maximum of 500W seems reasonable.
For the rest of the lineup, it’s possible that Nvidia will try to keep things more conservative instead of pushing for higher power consumption. As pointed out by NotebookCheck, Nvidia’s current trend of increasing total board power (TBP) is still fairly new — especially on cards like the RTX 4080. Historically, xx80 cards stayed well under 300W, even dipping below 200W at times. In the last couple of generations, the RTX 3080 and the RTX 4080 both pushed the TBP to new heights, with each requiring up to 320W.
With power consumption as high as this, it doesn’t make a lot of sense for Nvidia to keep pushing for even higher wattages, especially seeing as AMD is likely to keep it more conservative in RDNA 4. If Nvidia dials it back a little, we might see the RTX 5080 with a TBP of around 250W to 280W. However, if Nvidia sticks to its current scheme, it might go in the other direction and hit as high as 350W.
Editors’ Recommendations