Skip to main content
  1. Home
  2. Computing
  3. News

Nvidia’s RTX 4090 breaks all limits in a leaked benchmark

Add as a preferred source on Google

With the predicted launch window of the Nvidia GeForce RTX 40-Series inching closer, we’re starting to hear more about the reported performance of these next-gen graphics cards. Today, we got a real treat — rumored benchmarks of the flagship RTX 4090, spotted in the wild on Chiphell forums.

It’s all good news today. For one, the GPU managed to destroy its predecessors, and two, it hit a clock speed that’s still very rare to see — and that’s just in its engineering stage. Is Nvidia really going to offer as big a performance leap as it seems right now?

Nvidia RTX 4090 TimeSpy benchmark score.
Chiphell

It’s worth noting that the GPU is unnamed, but it definitely seems like the RTX 4090. The graphics card was tested in 3DMark Time Spy Extreme and it did very well, although it was paired with a mid-range processor (Intel Core i5-12400F) that didn’t help it achieve a higher score. Judging by the screenshot, the GPU scored 20,192 points, but the overall score is quite low, and this will likely be improved by pairing the GPU with a similarly high-end processor.

Recommended Videos

Going by the 20,192 score spells good news for the RTX 4090 — it beats its predecessor by a healthy amount. We’ve previously heard whispers of Nvidia’s next-gen GPUs being able to double the performance of the RTX 30-Series, and we’re not quite hitting those numbers, but it’s definitely close. VideoCardz notes that the GPU is 78% faster than the flagship RTX 3090 Ti and 90% faster than the RTX 3090.

Aside from the benchmark score, the leaker shared tantalizing details about the clocks and the temperature of the GPU. The graphics card managed to hit 3,015MHz in this test, meaning it surpassed the 3GHz barrier that was previously only speculated about. This is especially exciting when you consider that the leaker is presumably using an engineering sample, so the real performance of the card could be even better.

In terms of temperatures, the leaker revealed that the tests were conducted at a 30-degree Celsius ambient temperature (86 degrees F). The graphics card temps hovered around the 65°C mark, although at times, they did go higher. This GPU was air-cooled and the leaker notes that it’s very large, something that was already implied by recently revealed renders.

Render of an Nvidia GeForce RTX 4090 graphics card.
QbitLeaks

There’s always the possibility that this is an Nvidia GeForce RTX 4090 Ti rather than an RTX 4090, but it seems highly unlikely. The TDP is rated at 450 watts, implying an RTX 4090, and it’s likely too soon for samples of the RTX 4090 Ti to be out. In any case, the GPU did a fantastic job in this test, and since it falls in line with previous rumors and leaks, it might be true that Nvidia is readying to deliver a big performance increase with this generation.

We will need to wait for more benchmarks to appear to make a proper judgment, but in all likelihood, we won’t have to wait long. Nvidia is most likely going to announce its RTX 40-Series graphics cards at GTC on September 20, so after that, the flood gates are likely to open and we might learn more about the real performance of these GPUs.

Monica J. White
Monica is a computing writer at Digital Trends, focusing on PC hardware. Since joining the team in 2021, Monica has written…
AMD just made Ryzen laptop chips even more confusing, but here’s what’s actually new
The refreshed lineup brings more Zen 4 processors to mainstream and budget laptops.
AMD Ryzen 100 and 200 series

AMD has quietly expanded its mobile processor portfolio with 11 new Ryzen laptop processors, adding fresh models under both the Ryzen 200 and Ryzen 100 families. While that sounds straightforward enough, the bigger story isn't the chips themselves -- it's AMD's increasingly confusing naming strategy. The company has introduced seven new Ryzen 200 processors alongside four new Ryzen 100 models, but despite belonging to different series, many of them are actually built on the same Hawk Point silicon featuring Zen 4 CPU cores and RDNA 3 integrated graphics.

The Ryzen 200 series gets seven new CPUs

Read more
OpenAI is killing ChatGPT Atlas browser. I loved it, but it was an uphill race to the top
It was a trailblazer in a few ways, before it was copied down to its skeleton.
ChatGPT Atlas browser on a MacBook.

When OpenAI launched its own web browser, there was plenty of skepticism as to why a frontier AI lab is even bothering with making a browser in the first place. And yet, the company went ahead and launched ChatGPT Atlas with a heavy dosage of AI features built in. Well, the days of browser ambitions are over, and it will be put on cold ice in September this year.

OpenAI says it is sunsetting the short-lived browser in favor of pushing the new ChatGPT work desktop app, which already has a built-in browser as well as a cloud browser for AI agents. And now that ChatGPT is making its way to other browsers, such as Chrome, as an extension, there is little need for maintaining a dedicated browser project of its own.

Read more
Windows 11 Search is getting bigger, but only by 4 pixels
The change could be in preparation for the upcoming Ask Copilot feature
Windows 11 Laptop

If you have used Windows 11 Search after the June update, you may have noticed it feels a little less annoying. Microsoft recently made the Start menu and Search more responsive, and also fixed one of Search’s stranger limits by letting it find local files using just two characters.

Now, the company appears to be making a much smaller change. According to Windows Central, Microsoft accidentally revealed that the search box in the Taskbar and Start menu is getting 4 pixels taller. Four pixels sounds like the kind of change only a UI/UX designer could love, but screenshots from the Insider Preview build suggest it is visible once you know where to look.

Read more