Skip to main content

Nvidia RTX 40-series could have incredible TFLOP performance

Nvidia’s upcoming GeForce RTX 40 series of GPUs could possibly more than double the teraflops (TFLOPS) count of the 3090 Ti, and may even exceed 100 TFLOPS.

Reliable tipster Greymon55, who has provided several rumors pertaining to Nvidia’s next-gen graphics cards, has reaffirmed his stance regarding the graphical compute performance (FP32) of Lovelace video cards.

Recommended Videos

https://twitter.com/greymon55/status/1485833625178284038

He tweeted that “if Lovelace gets up to 2.5 GHz, you’ll get about 90t of FP32.” Previously, he referred to the possibility of the GeForce RTX 40 series offering between 85 to 92 TFLOPS in FP32 in November 2021. For reference, Lovelace is believed to be Nvidia’s codename for its next-generation GPUs.

As pointed out by NotebookCheck, TFLOPS (tera (trillion) floating operations per second) has increasingly become a commonly used metric to determine how powerful a graphics card is in general. As opposed to gigahertz (GHz), which determines a processor’s clock speed, TFLOP measures a computer’s performance.

Nvidia’s Ampere architecture allowed the RTX 3090 to be powered by 35.58 TFLOPS in FP32 (single-precision floating-point format), while the highly anticipated GeForce RTX 3090 Ti GPU is said to deliver 40 TFLOPS.

An RTX 4090 video card (based on the AD102 GPU) offering 90 TFLOPS would therefore be 2.5 times more than the RTX 3090’s teraflops count. NotebookCheck highlights how an upgraded RTX 4090 Ti based on the next-gen Lovelace microarchitecture could theoretically exceed 100 TFLOPS. Comparatively, the PlayStation 5 and Xbox Series X sport 10.28 TFLOPS and 12.15 TFLOPS, respectively.

With such a high TFLOPS count, it reinforces a previous rumor that the RTX 4090 or RTX 4090 Ti could require a huge 1,200-watt power supply unit (PSU).

Greymon55 also pointed out that Lovelace video cards would offer “double performance” of the GeForce RTX 3090. When considering the rumored technical specifications, that statement could very well become a reality — in addition to 18,432 CUDA cores, the RTX 4090 is said to sport 24GB of GDDR6X memory, joined by a clock speed between 2.3GHz to 2.5GHz. Greymon55’s sources indicate that next-gen Nvidia GPUs will arrive in September 2022.

https://twitter.com/greymon55/status/1483642725891117057

As for the GeForce RTX 3090 Ti, the graphics card was reportedly scheduled for a release on January 27, but Nvidia has apparently stopped the production of the card for now. In any case, when it does eventually launch, it’s set to become the most expensive GPU ever ($4,500). That price record will most likely be broken when its successor, the RTX 4090 Ti, hits the market.

Nvidia is reportedly spending several billion dollars to use TSMC’s 5nm process node required for GeForce RTX 40 GPUs.

Zak Islam
Former Digital Trends Contributor
Zak Islam was a freelance writer at Digital Trends covering the latest news in the technology world, particularly the…
Nvidia RTX Pro 6000 Blackwell sets a new standard for workstation GPUs
Nvidia RTX Pro 6000 Blackwell GPU

Nvidia announced a host of new GPUs at its GTC 2025 conference on Tuesday, including the RTX Pro Blackwell series. The highlight component of the series is the RTX Pro 6000 Blackwell GPU, developed for workstations. The RTX Pro 6000 Blackwell GPU is also available in variants to support desktops, laptops, and data centers. The component is intended to aid professional designers, developers, data scientists, and creatives, the Verge noted. 

The RTX Pro naming sequence allows the Blackwell-based GPU to stand out against previous architectures, including Ada Lovelace, Ampere, and Turing, as well as the standard consumer product GPUs, Tech Radar Pro noted. 

Read more
Nvidia RTX 50 series owners can unlock free GDDR7 memory speed boost
Screenshots of MSI Afterburner over a colorful background.

Popular graphics card monitoring and overclocking tool, MSI Afterburner, has received a beta update enabling owners of Nvidia’s RTX 50-series GPUs to boost their GDDR7 memory speeds by up to 10%. This enhancement allows data transfer rates to reach up to 36 gigatransfers per second (GT/s), surpassing Nvidia's standard specifications.

The RTX 50-series graphics cards are equipped with GDDR7 memory modules rated between 28GT/s and 32GT/s. However, Nvidia often sets default data transfer rates slightly lower, at 28GT/s for most models and 30GT/s for the RTX 5080, to ensure stability and longevity. The new update to MSI Afterburner unlocks the potential to exceed these factory settings, offering enthusiasts the opportunity to maximize their hardware's performance.

Read more
Nvidia claims RTX 5000 shipped better than 4000 but gamers are still waiting
The RTX 5090 sitting on a pink background.

Nvidia is trying to make its GeForce RTX 5000 series seem more impressive to the media by suggesting that the latest GPUs are selling better than the previous generation. However, many pundits aren’t buying the claim.

PC Mag pondered whether Nvidia has orchestrated a “paper launch” of the RTX 5000 series, suggesting that there might not be much of a product available for consumers. The majority of the people with their hands on the GPUs, especially the high-end models such as the 5090 and 5080 appear to be reviewers, influencers, and other determined enthusiasts as opposed to everyday gamers, who are still using prior generation GPUs at higher rates.

Read more