Saturday, November 8th 2025
AMD Ryzen 7 9700X3D Appears in Leaked PassMark Benchmark
A new leaked PassMark benchmark has revealed what appears to be AMD's next 3D V-Cache processor, the Ryzen 7 9700X3D, a new CPU to refresh the company's mid-range gaming lineup. The listing shows it is an eight-core, 16-thread Zen 5 CPU with 96 MB of L3 cache and a boost clock of up to 5.8 GHz, however, that figure is likely from early testing or overclocked conditions. The 9700X3D scored 4,687 points in single-threaded and 40,438 points in multithreaded benchmarks. That's about 1-5% ahead of the existing Ryzen 7 9800X3D. With the Ryzen 7 9700X3D, AMD might follow the same path as the 5700X3D, offering slightly lower clocks and price while maintaining strong gaming performance.
It's also rumored, as we reported earlier, that AMD is developing a higher-end Ryzen 7 9850X3D (120 W, 8 cores, 16 threads, 96 MB L3 cache, 5.6 GHz/4.7 GHz), a Ryzen 9 9950X3D2 (200 W, 16 cores, 32 threads, 192 MB L3 cache, 5.6 GHz/4.3 GHz), and an entry-level Ryzen 5 7500X3D (65 W, 6 cores, 12 threads, 96 MB L3 cache). With CES 2026 just two months away, we can expect AMD to release some of these now-rumored CPUs there, if not all four of them.
Update Nov 8th: This PassMark benchmark has been identified as a faked result. Redditors used Linux commands to submit spoofed identification data, using a standard 9700X, apparently to see if it was possible. The higher clocks than the 9800X3D would be unusual for a lower binned X3D part. Many news sites reported on the story, and the shenanigans serve as a warning regarding trusting the veracity of unverified benchmarks.
Source:
Videocardz
It's also rumored, as we reported earlier, that AMD is developing a higher-end Ryzen 7 9850X3D (120 W, 8 cores, 16 threads, 96 MB L3 cache, 5.6 GHz/4.7 GHz), a Ryzen 9 9950X3D2 (200 W, 16 cores, 32 threads, 192 MB L3 cache, 5.6 GHz/4.3 GHz), and an entry-level Ryzen 5 7500X3D (65 W, 6 cores, 12 threads, 96 MB L3 cache). With CES 2026 just two months away, we can expect AMD to release some of these now-rumored CPUs there, if not all four of them.
Update Nov 8th: This PassMark benchmark has been identified as a faked result. Redditors used Linux commands to submit spoofed identification data, using a standard 9700X, apparently to see if it was possible. The higher clocks than the 9800X3D would be unusual for a lower binned X3D part. Many news sites reported on the story, and the shenanigans serve as a warning regarding trusting the veracity of unverified benchmarks.
35 Comments on AMD Ryzen 7 9700X3D Appears in Leaked PassMark Benchmark
Unless it has higher boost clocks which the 9850X3D should have.
I'm really curious with this substantial uptick in clock speeds how it's going to play out with 12 core CCD's and X3D, I hope it boosts past 6ghz with a commensurate increase in cache (144mb for 12 cores.)
I'm guessing that's what you'd pull the trigger on as well.
I don't know why, but I have a feeling that this one is going to be locked so it doesn't shove the 9800X3D and have both for sale.
Alternatively, they might retire it in favour of 9700/9850 being the 8 core skews (or just litter the range with as many as possible to keep us all guessing)
But I have that itch boys. That terrible itch. And I cannot afford to scratch it.
intels big cache cpus long time away with no prove of performance, bartlett sneak preview can stop this rampage
Reddit post shows how they did it
I do hope that when they do push that bigger CCD, they also do it with conservative power management like the 7800X3D has it. I don't need an inefficient CPU for the last 200 mhz.
I believe that temperature does have a direct relation with every processor's stability. As a specific example, the 7800X3D's SoC and CCD voltages were locked to protect the delicate 3D V-Cache.
In the 9000X3D series chips, the 3D V-Cache has been repositioned below the CCD. This change allows the primary heat source, the processor cores, to interface directly with the cooling solution. According to AMD, the second generation 3D V-Cache design is now (up to 46%) more thermally-resistant whichs enables higher sustained clock frequencies in both single and multi-threaded workloads.
The 1.3 V limit was to stop degradation of the cache, which isn't really the major heat producing element, but had all of the thermal energy produced from the rest of the chip pass through it.
This could also be alleviated by backside power delivery and gate all around transistors, with less leakage.
That said, my media PC has a 4060 and a 12400f and....it plays most of my games just fine. Sins of a solar empire struggles, but thats it. Factorio hypothetically favors the x3d but since 2.0 came out and broke some mods I havent seen a map that doesnt work right on the i5.
The time to scratch that itch is when we get 12 core CCD x3d chips.
Maybe apply that logic to leaks and rumours for the front page too.
Yeah it’s juicy, but if it’s wrong you damage the sites credibility.
Or give it a clearly visible tag that it’s just a rumour.
These things can be programmed to fail.
wccftech.com/amd-ryzen-cpu-with-dual-x3d-chiplets-is-reportedly-fake-and-doesnt-exist/
The provided screenshot is missing the "Turbo speed" and "Typical TDP" fields, plus CPU cache specs, and reports the turbo speed as the base clockspeed.
But also I don't know if Passmark blanks things out for CPUs it's not supposed to know about. Even then, you can usually grab those specs even if you don't know the CPU model.