The anticipation for the Nvidia GeForce RTX 5090 is high, as consumers expect another record-breaking GPU. Here’s everything we know about the upcoming card, including its speculated price, specifications, and rumored performance.
The Nvidia GeForce RTX 4090 is the fastest consumer graphics card on the market, and its successor has high expectations to live up to. The RTX 5090 is the highly anticipated follow-up to the RTX 4090, and consumers are expecting big things from this upcoming card.
Nvidia has yet to release any concrete details about this GPU, and most of the information we have about it has come from industry rumors and leaks. As always, this means that until Nvidia releases any official statements, this information should be taken with a grain of salt.
Speculations on the release window
According to Taiwanese newspaper United Daily News, the RTX 5090 is slated for release in the fourth quarter of 2024. This date would mark two years since the RTX 40 series was revealed. According to the leak, the move is aimed at “helping graphics card manufacturers ship their products this year and increasing the average unit price of graphics cards.”
Regular leaker Moore’s Law is Dead claims that the RTX 5090 is expected to launch first, and that it will be the only RTX 50 series card to launch in 2024. His colleague Kopite7Kimi, however, disputed this claim. Kopite expressed his belief that the RTX 5080 will be released first, with the RTX 5090 to follow in late 2024.
Specs leaked
The RTX 4090 is expected to use the GB202 chip. This chip is the flagship of the new Blackwell architecture lineup, which was officially announced by Nvidia at GTC 2024.
Analysts predict that the GB202 gaming chip will see significant performance gains over the current Ada Lovelace AD102 chips. The L1 cache in particular is expected to see significant performance gains.
Kopite7Kimi, a reliable tipster, has indicated that the GB202 will use the same process node as the AI-focused GB101 chip. Additionally, Kopite7Kimi suggested that the chip will have 12 GPCs, with 8 TPCs each, for a total of 96 TPCs.
Panzerlied, another reliable hardware leaker, has shared some possible specs for the RTX 5090, which if accurate, could show some very impressive improvements.
As for the memory bus, there is some debate. Early reports indicated that Nvidia plans to reuse the existing 384-bit memory bus from Ada chips. However, more recent reports suggest that Nvidia will use a 512-bit memory bus, but only on flagship chips such as the GB202 that is expected to be used in the RTX 5090.
A rumor from May 2024 also from Panzerlied suggests that the RTX 5090 will feature a very “dense” memory configuration. The leaker specifies that the GPU will use four memory modules on top, with five more on each side and two below.
Panzerlied provided additional information in July 2024 that the RTX 5090’s base clock speed could reach 2.9 GHz. This is getting close to the 3 GHz barrier and represents a stunning 30% increase in base clock speed over the RTX 4090.
The Boost clock speed hasn’t been specified yet, but with such a high base clock speed, the RTX 5090 should be able to hit speeds well over 3GHz with the Boost clock engaged.
For comparison, here are the currently rumored specs of the RTX 5090 versus the RTX 4090:
A Twitter/X user called Everest We found that Seasonic included the data for the RTX 50 series of graphics cards in the PSU wattage calculator. While this information is not completely accurate, it is in line with the expected numbers.
Assuming the data is accurate, customers may want to upgrade their PSU before purchasing an RTX 5090. Seasonic’s data puts the RTX 5090’s TDP at 500W.
Benchlife contributed to the RTX 5090 TDP rumors from early September 2024Benchlife has close ties to Nvidia’s board partners and is well placed to have access to hard numbers. It says the RTX 5090 is expected to be a 550W card.
Kopite7Kimi The RTX 5090 was rumored to draw 600W of power. However, other data from Benchlife suggests that this is for a high-power cooling module for the card, and the 600W figure refers to the heat dissipation capacity, not the TDP.
Speculations on the design
If the memory configuration suggested above is accurate, it will require an entirely new PCB design to accommodate this unusual layout. This also reinforces rumors that the RTX 5090 will use a 512-bit memory bus, as a large bit width would be required for such a dense memory configuration.
Related to this, there is a leaked end of May 2024which once again comes from Kopite7Kimi. This leak suggests that the RTX 5090 Founders Edition will use an entirely new PCB layout. Specifically, the card will be split into three sections: the main board, a rigid I/O board, and a separate PCIe slot section.
According to the leak, the motherboard and I/O board are needed to house the increasing number of components required by the Blackwell architecture. The rigid I/O board could be used to house the display controllers, power supply, or other components that don’t require active cooling. The PCIe slot, on the other hand, will act as a riser, containing no circuitry of its own.
Based on these leaked design specifications and looking at Nvidia’s previous multi-PCB designs, it’s possible that this layout could potentially make the RTX 5090 thinner than its predecessors.
Once again, Kopite7Kimi confirms this theory. A user on Twitter/X There was some speculation as to whether the RTX 5090 would use the bulky 4-slot cooler originally designed for the unreleased RTX 4090 Ti. Kopite responded that the RTX 5090 Founders Edition would instead use a dual-slot cooler with a dual-fan layout.
Kopite also said that unlike its rivals AMD, Nvidia is not using a chiplet build. The GB202 chip is likely to power the RTX 5090, and Kopite’s leak suggests that this chip will continue to use Monolithic GPU Dies.
Speculations on performance
Jugo Mobile
According to Panzerlied’s leak, the RTX 5090 will end up being around 70% faster than the RTX 4090. However, it’s unclear how or where this figure could be replicated. The leaker is likely talking about pure, rasterized performance, with no DLSS frills to boost performance.
That ends up being around 1.7x faster than the current fastest gaming GPU, meaning the Blackwell generation could end up being one of the biggest generational leaps in GPUs we’ve seen in quite some time.
Speculations on features
Current data suggests that the RTX 5090 will support Display Port 2.1 as well as HDMI 2.1, and will be an update to the Ada Lovelace series, which did not offer DisplayPort 2.1 support.
Jugo Mobile
The power connector has proven to be a sore point. There have been numerous reliability issues reported with the 12VHPWR connector used in the RTX 4090 and 40 Super cards, with some consumers reporting melted connectors.
Another reliable and persistent leaker, Moore’s Law is Dead, has reported that Nvidia is considering switching to a new connector type for its next-gen cards. The leak goes on to say that Team Green is working on an entirely new type of 16-pin connector. So far, however, no other leaks have confirmed this rumor.
Price speculation
While very little information is currently available on what to expect regarding the pricing structure of the RTX 50 series, some analysts are predicting that the RTX 5090 will retail with an MSRP of $1,699 at launch.
The RTX 4090 has been unchallenged in its category. With the RTX 5090 set to have similar dominance in the extremely high-end segment of the market, this could give Nvidia carte blanche to charge whatever it wants.
Should we wait for the RTX 5090?
Customers looking to buy a high-end GPU may be asked to wait until late 2024. If the RTX 5090 launches during the expected timeframe, the wealthy will have a shiny new toy to play with. Those looking for a bargain could also take advantage by picking up discounted RTX 4090 or RTX 4080 Super cards when retailers look to clear out their old stock.
