All rumors so far

AMD and Nvidia have been going head-to-head for years as the only consumer desktop graphics card manufacturers. Although Intel's Arc series brought Team Blue into the game, the Nvidia GeForce RTX 4000 series dominates our list of the best GPUs – and for good reason. Not only do the Ada Lovelace GPUs offer amazing ray tracing capabilities, but you can also enjoy smoother frame rates thanks to a combination of DLSS 3.5 and Nvidia's proprietary frame generation technology.

Related

Which GPU should you buy in each $100 price range?

There is something for everyone here, whether you are a budget user or a gaming enthusiast

In addition, Team Green's upcoming series is expected to be even better than the current generation. Nvidia's new graphics cards, named after mathematician David Blackwell “Blackwell”, could bring high-bandwidth GDDR7 memory, support for PCIe 5.0, and many more features. In this article, we summarize all the rumors and leaks about the GeForce RTX 5000 family.

Nvidia GeForce RTX 5000 series: when we want to see them and how much we want to pay

Nvidia typically follows a two-year release cycle for consumer graphics cards, with current-generation hardware refreshed roughly every year. With Team Green unveiling the Super variants of its current RTX 4000 GPUs in January 2024, it was assumed that the Blackwell series would be unveiled in late 2024, although all signs now point to an announcement at CES 2025.

Previously, HardwareLUXX, a German tech website, reported that we won’t see the “Ada Lovelace-Next” graphics cards before 2025. On the other hand, Youtuber Moore's Law is deadclaimed that the newer cards could appear as early as the fourth quarter of 2024. In an older X (formerly called Twitter) threadLeaker kopite7kimi speculated that the new GPUs could launch “at the end of the year,” before noting that the release of the new GPUs could be pushed back to 2025 due to sanctions. Separately, the leaker also stated that the RTX 5080 (codenamed GB203) will launch before the top-of-the-line RTX 5090.

The prices of the new graphics cards are currently anyone's guess. The RTX series is known for its overpriced nature, and the exorbitant pricing trend is likely to continue with the Blackwell family. Assuming we get an RTX 5090 instead of a 5090 Ti as the flagship GPU, it should have the same (or slightly higher) launch price of $1,599 as the current generation GeForce RTX 4090.

Nvidia GeForce RTX 5000 series: What we saw

Higher bandwidth GDDR7 memory on most GPUs

Many GPU enthusiasts were disappointed with the Ada Lovelace family because Nvidia decided to adopt the GDDR6X and GDDR6 memory from older GPUs. Fortunately, the rumor mill claims that the flagship GeForce RTX 5000 series graphics card(s) will utilize the latest GDDR7 memory. For comparison, the GeForce RTX 4090 had a memory bandwidth of 1,008 GB/s, which is the same as the last generation GeForce RTX 3090 Ti. In contrast, the RTX 5090's memory bandwidth is rumored to be 1,536 GB/s, which is an astonishing 52% increase!

There were also numerous rumors about the memory bus of the GeForce 5090. After many contradictory statements about the RTX 5090, kopite7kimi revealed that the crème-de-la-crème Blackwell family GPU would feature a 512-bit memory bus. In June, they published a table detailing the memory bus on each GPU in the upcoming family. Interestingly, the RTX 5050 (GB207) is the only graphics card in the series to feature the outdated GDDR6 video memory.

Speaking of which, aside from a few rumors that the RTX 5090 will have 36GB of GDDR7 memory, there haven't been too many updates on the GPU's VRAM size. Still, like the last-gen RTX 4090, the GPU will have at least 24GB of VRAM. As for the other high-end model, kopite7kimi earlier quoted that the GB203 card may have half the memory of the GB202 variant. Assuming it's not a “multi-chip package,” this means a theoretical RTX 5080 will have between 12 and 18 GB of memory.

Higher TDP next to 16-pin connectors

In early July, power supply manufacturer Seasonic accidentally added the TDP requirements of Nvidia's next-gen GPUs to its online wattage calculator. The leaked specs were first spotted by Twitter user Everest (@Olrak29_) and removed by Seasonic shortly after, but not before news outlets like VideoCardZ tracked the power consumption of all next-gen GPUs.

Obviously, there are some power-hungry graphics cards in the Blackwell family, with the RTX 5090 requiring 50W more than its last-gen counterpart. While the RTX 5080 and 5070 beat the power consumption of their predecessors by 30W and 20W respectively, the RTX 5060 is quite a monster. With a massive TDP of 170W, the RTX 5060 requires 55W more than the RTX 4060 and 5W more than the 16GB variant of the RTX 4060 Ti. Additionally, all GPUs will reportedly feature 16-pin connectors, which is quite surprising considering there's even an RTX 5050 in the mix. Regardless, you may need to upgrade your PSU if you plan on equipping your PC with the upcoming Team Green GPUs.

Possible MCM design

Unlike AMD, Nvidia relies on monolithic chips for its graphics cards. With a monolithic chip, all the circuitry is on the same chip, making it difficult to shrink the transistors without increasing the overall size of the chip. In contrast, a multi-chip module (MCM) approach adds separate chips (called chiplets) connected via interposers on a single substrate. This increases the scalability of MCM-based chips and allows manufacturers to build in better specifications.

According to leaker kopite7kimiNvidia's data center chips GB100 and GB102 are Special feature an MCM design instead of the traditional monolithic architecture. Likewise, the high-end GB202 chip is expected to bring the MCM architecture to consumer GPUs. However, this new design will likely not be included in the GB203, GB205, GB206, and GB207 chips, as Nvidia will use the same old monolithic architecture for its budget and mid-range GPUs.

New TSMC 4N manufacturing process (with higher core count)

Nvidia has been using newer manufacturing processes in the last few GPU generations, and according to rumors, the Blackwell series will be no exception. Team Green's upcoming GPU family will be based on the TSMC 4N process node, which is based on the TSMC 5 manufacturing process. Recently kopite7kimi Also confirmed that the flagship GB202 chip will have the same manufacturing process as GB100, the chip for data center GPUs. However, the leaker noted that the chip density of the GB202 SKU would increase by 30%.

In a (n older tweet, kopite7kimi revealed that there will be no significant increase in the number of Graphics Processing Clusters (GPCs) and Texture Processor Clusters (TPCs) in the new GPUs. In addition, Panzerlied claimed that in addition to a 15% higher frequency and 78% more cache, the GeForce RTX 5090 could also feature a 50% scaling improvement, which could translate into a higher number of 24,576 CUDA cores.

Support for PCIe 5.0 and DisplayPort 2.1

Although the Ada Lovelace GPUs work with PCIe 5.0 motherboards, they cannot fully utilize the high-speed interface standard. However, the Blackwell GPUs are expected to be able to take advantage of the lightning-fast data transfer speeds of PCI Express Gen 5.0.

It is also rumored that the GeForce RTX 5000 graphics cards will be a step ahead of their predecessors in terms of display capabilities. According to kopite7kimiThe new GPUs will be compatible with the DisplayPort 2.1 standard, which supports much higher bandwidth, refresh rate and resolution than the last generation DisplayPort 1.4.

Nvidia GeForce RTX 5000 Series Rumors: To Believe or Not to Believe

While many of these leaks have been published by trusted sources, you should take them with a grain of salt. Not only have the rumors surrounding these GPUs changed time and time again, but many of the leaks surrounding older Nvidia graphics cards have been debunked shortly after the products were launched. For example, there was quite an uproar in the community when the flagship RTX 4090 from the Ada Lovelace family was believed to have a maximum TDP of 800W. Thankfully, the card debuted with a TDP of 450W instead.

While there are many reasons why Nvidia will launch the Blackwell family in 2025, the company hasn't said anything for now. Either way, we'll know more about the upcoming GPU family soon. But if you can't wait for the Blackwell series, the RTX 40 series GPUs are still worth a look. The GeForce RTX 4070 in particular remains our favorite GPU, offering solid performance without costing a fortune.

Related

MSI GeForce RTX 4070 Ventus 3X in review: 1440p gaming at a bargain price

Affordable RTX 4070 performance with impressive 1440p results