Skip to main content

Nvidia’s next-generation GPUs could destroy Xbox Series X if leaks are true

Nvidia’s Turing-based RTX GPUs are some of the most powerful you can get your hands on today, but the next-generation is on its way. It appears that the company has begun testing its next-generation Ampere architecture, and a new series of leaked benchmarks are showing some promising results.

Recommended Videos

Several Geekbench 5 benchmarks were spotted by a Twitter user and then picked up by PCGamesN that detail some significant performance results and specifications for Nvidia’s upcoming GPU.

Stock photo of Nvidia T4 GPU
NVIDIA Newsroom/NVIDIA Corporation

Nvidia appears to have tested two Ampere GPUs with the codename of Fat Man and Little Boy, and what’s curious about these leaked early benchmarks is that the cards were tested on a non-high end system running an Intel Z370 Coffee Lake motherboard, according to PCGames N.

Get your weekly teardown of the tech behind PC gaming
Check your inbox!

The more powerful card, Fat Man, has a whopping 60% more streaming multiprocessors compared to the current GeForce RTX cards with a total of 118. If Nvidia maintains its current core-to-SM ratio from today, that will translate to 7,552 CUDA cores for Fat Man.

The card also comes 24GB of video RAM, matching what’s available on the premium Titan RTX card from today. Fat Man was tested with an unusually low boost clock speed of 1.11GHz.

PCGames N estimated that Fat Man to be capable of more than 33.5 TFLOPs at that clock speed. If Nvidia increases the speed on the card at launch, Fat Man’s computational capabilities could go even higher, making it a card that would be suitable for high end data analysis, deep learning applications, and artificial intelligence.

It’s unclear if Nvidia will be using Ampere for a rumored GeForce RTX 3080 launch, but it was previously speculated that the architecture may not be targeted at gamers, at least not initially given the massive amounts of video memory.

The less powerful Little Boy features 108 SMs but comes with double the memory of Fat Man. It’s clocked at 1.01 GHz in the leaked Geekbench test result. This should translate to roughly 6,912 CUDA cores and nearly 28 TFLOPs of compute power. Though not a direct apples-to-apples comparison of TFLOP performance, Microsoft recently revealed its new Xbox Series X comes tops out with 12 TFLOPs of compute power.

Microsoft / Microsoft

The TFLOP estimates assume that Nvidia will double its floating point units on Ampere, which would double the TFLOP performance. If Ampere doesn’t come with double the FP units, then the cards may max out with 17 TFLOPs and 14 TFLOPs, according to TechRadar.

Given that we still don’t know much about Ampere’s overall architecture, the CUDA core calculations as well as TFLOP approximations can drastically change when the cards are unveiled. And according to previous rumors, the Ampere graphics cards could potentially match AMD’s 7nm manufacturing process, giving it even higher efficiency than Turing.

With Nvidia’s annual GPU Technology Conference (GTC) slated to happen next month, where the company is expected to unveil Ampere.

Chuong Nguyen
Silicon Valley-based technology reporter and Giants baseball fan who splits his time between Northern California and Southern…
Wait, what? Nvidia’s RTX 50-series might be ready to go this month
RTX 4090.

Update: According to MEGAsizeGPU on X (formerly Twitter), the following might be a mistranslation. The source reportedly talks about Nvidia finalizing the design for Blackwell in September instead of launching the card. No details about the release date have been confirmed. The original article follows below.

Nvidia's RTX 50-series graphics cards are coming -- we know that for a fact. But when? The release dates of these graphics cards have been the topic of much speculation. Early leakers predicted that they'd launch in late 2024, but the general consensus slowly shifted toward an early 2025 release instead.

Read more
Nvidia’s next-gen GPUs may be delayed due to ‘design flaws’
Nvidia introducing its Blackwell GPU architecture at GTC 2024.

Dark clouds are looming over the future of Nvidia's best graphics cards. According to a new report, Nvidia told some of its partners that it will be delaying its upcoming Blackwell GPUs, and is now aiming for an early 2025 release instead. Delays are one thing, but the cause is perhaps the most worrying part of it all -- design flaws. What does this mean for Nvidia's RTX 50-series?

This worrying report originates from The Information, which cites two sources who helped produce the Blackwell chip, as well as its server hardware. Bloomberg recounts that the chips may be delayed by over three months at this point. Nvidia is preparing B100 and B200 chips for some of the world's most prominent tech companies, including Meta, Google, and Microsoft, so these delays could hit pretty hard.

Read more
Nvidia may be working on a surprising new budget GPU
Nvidia GeForce RTX 3050 graphics card.

This is certainly unexpected. Hints of an upcoming Nvidia graphics card showed up in the PCI-ID database, and it's definitely not what you think. Instead of working on the RTX 50-series or refreshing a 40-series GPU, Nvidia is bringing back the slowest RTX 30-series card -- the RTX 3050 -- but with an Ada Lovelace chip. In theory, that could make this the first RTX 30-series GPU to have access to Deep Learning Super Sampling 3 (DLSS 3).

The GPU in question is a laptop version of the RTX 3050. It's hard to say why Nvidia would choose to release new versions of that particular card in 2024 (or beyond), but the RTX 3050 A exists -- VideoCardz found traces of it in the latest Nvidia drivers. Just the fact that there's an RTX 3050 in the works is surprising, but that it's based on the AD106 chip is what baffles me.

Read more