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

AMD Threadripper 3990X is breaking CPU world records just days after release

Add as a preferred source on Google

Threadripper 3990X
AMD’s 3990X Threadripper CPU should dominate in the HEDT space. Image used with permission by copyright holder

AMD’s Threadripper 3990X isn’t just a super-powerful CPU on paper. Just days after its general release, it’s smashed a number of world records, breaking those set by its 32-core sibling, the 3970X, as well as Intel’s best Xeons, and even some dual-socket machines. AMD’s 3990X is making just about every other processor on the planet look pedestrian in comparison.

Recommended Videos

AMD’s Ryzen processors have always offered impressive multithreaded performance, from the first generation to the third. But with the instructions per clock improvements, the Zen 2 architecture, the chiplet design, and the 7nm process node shrink of the third-generation, it’s broken new ground. Alongside up to 16-cores in a mainstream desktop CPU with the 3950X, AMD’s Threadripper 3990X has ushered in a whole new era of high-end desktop performance. Its 64 cores, 128 threads, 288MB of combined cache, and 4GHz+ boost frequency, make it the most powerful CPU ever outside of a server room.

And even there it’s competitive.

The first record to fall following the 3990X’s debut on February 7 was GeekBench3 — Multi Core. Supercooled by liquid nitrogen and running on an all-core boost frequency of 5,375MHz, the 3990X scored a massive 293,771 points. That’s 10,000 points more than another 3990X benchmarked that same day, and more than 100,000 points more than the 3970X, which originally took the record in December last year.

That was followed by a GPUPi for CPU record, where the same 3990X running at 5,475MHz across all cores stole the top spot, again beating out the 3970X. The same again with Cinebench R15, and wPrime 1024m a couple of days later.

What’s so impressive about the Threadripper 3990X in each of these cases isn’t just the speed with which these records were taken, but the dramatic fashion. Between 2013 and June 2019, just before the release of third-generation Ryzen CPUs, the GeekBench3 multi-core score had increased from 38,000 to 135,000. Just six months later, the 3990X has more than doubled that score.

In Cinebench R15, the score increased from 1,000 in 2013, to 10,000 (with four Xeon CPUs) before the release of third-generation Ryzen CPUs. The 3990X increased that to 18,000 overnight.

These benchmarks are designed with multi-core CPUs and multiple CPUs in mind, so they scale far better than your average game or piece of desktop software. But the performance leap made with the 3990X is nothing short of staggering. Especially when you consider the state of the market. Intel’s 10th-generation HEDT CPUs, like the 18-core 10980XE, launched in the last few months of 2019 with a near-universal 50% price cut over their predecessors. And they were still dead on arrival, with no competition to offer AMD’s Threadrippper 3000 CPUs, even falling behind the mainstream Ryzen 3000 CPUs in some cases.

AMD didn’t need to release a 64-core Threadripper CPU. Its 32-core 3970X was more than enough to take the performance crown from Intel in all the benchmarks that matter for these kinds of chips. It may even eat into AMD’s own server business. AMD Epyc CPUs sport up to 64 cores but offer lower boost clocks than their Threadripper counterparts. They enjoy eight memory channels and more PCIExpress lanes, but a 64 core Threadripper CPU is, in some cases, faster than the 64 core Epyc alternative.

We’re already seeing some of this cannibalization of AMD’s own server market with the likes of Blur Studios using Threadripper 3000 CPUs for movie rendering. That’s a job typically reserved for server chips.

Terminator: Dark Fate meets 3rd Gen AMD Ryzen™ Threadripper™

In releasing the 3990X, AMD has leapfrogged everyone, including itself, and ended up so far ahead Intel may well take years to recover, with no sign of its upcoming CPU generations offering anything like the power of AMD’s frontrunner processors. Especially at the top end.

Jon Martindale
Jon Martindale covers how to guides, best-of lists, and explainers to help everyone understand the hottest new hardware and…
Google’s new Magic Pointer Play Store listing reveals a Gemini shortcut built for Googlebooks
The unannounced app turns the cursor into a contextual AI tool for search, image creation, and shopping
Plant, Text, Business Card

Google has quietly published a new Play Store listing for Magic Pointer, an unannounced app built for Googlebooks. Updated on July 10, the app turns the cursor into a Gemini shortcut that can act on whatever a user selects on screen.

Magic Pointer can send an image to Lens, generate a related image, or surface a shopping action without forcing users to open a separate chatbot. Regular Android devices currently show as incompatible, so the listing offers an early preview rather than a broad release.

Read more
You can stop using AI, but this new report says you probably can’t escape it
A UK survey found that most people feel AI exposure is unavoidable, raising harder questions about consent, privacy, and whether opting out is still realistic
AI Chatbots

More people are trying to use less AI, but avoiding it altogether may already be impossible.

A survey of 2,055 UK adults found that 42% deliberately limit how much AI they use. Another 70% said avoiding AI exposure would be difficult or impossible, even when they actively wanted less of it.

Read more
The face on an AI interviewer may matter as much as the decision it makes
Researchers found that race and gender matching changed how fairly rejected applicants viewed an automated interview, even though everyone received the same outcome
File, Computer Hardware, Electronics

An AI hiring system can treat every applicant the same and still leave some people feeling targeted. Researchers found that rejected candidates judged an automated interview differently depending on the race and gender of the avatar delivering the result.

Around 220 participants completed a simulated interview for a fictional customer support role with one of four photorealistic AI avatars. Everyone was rejected, yet perceptions of fairness shifted with the interviewer’s appearance. An algorithm audit could miss that reaction because candidates don’t experience the system as raw code. They experience a face asking questions and judging their answers.

Read more