AMD Ryzen 7 9800X3D Overclocked to 5.46 GHz, Beating Ryzen 7 7800X3D by 27%

[AleksandarK]

We are days away from the official November 7 launch of AMD's Ryzen 7 9800X3D CPU with 3D V-Cache, and we are already seeing some estimates of the speedup compared to the last-generation Ryzen 7 7800X3D CPU. According to a Geekbench submission discovered by Everest (Olrak29_) on X, the upcoming AMD Ryzen 7 9800X3D has been spotted running at a clock speed of 5.46 GHz. This is a 260 MHz increase from the official boost frequency of 5.2 GHz, which indicates overclocking has been applied. If readers recall, the last generations of X3D processors had overclocking disabled, and this time, things are looking different thanks to the compute die being placed on top of SRAM. AMD attributes this to CCD being closer to the heat spreader instead of memory and allowing it to spread heat more effectively, ensuring a stable overclock.

Regarding performance, the Ryzen 7 9800X3D outperforms its predecessor, the Ryzen 7 7800X3D, by an impressive 27.4% in the single-core Geekbench v6 test and 26.8% in the multicore test. The last generation CPU scored 2,726 points in single-core and 15,157 points in multicore tests, while the new Zen 5 design has managed to produce 3,473 points in single-core and 19,216 in multicore tests. These results are approximately 27% improvement over the Zen 4, suggesting that the Zen 5 architecture benefits greatly from better SRAM bandwidth and capacity. While these results only come from synthetic benchmarks, they give us a picture of what to expect from this CPU. We have to wait for more real-world test cases to fully conclude the improvement factor.



View at TechPowerUp Main Site | Source

 

[JohH]

Neat. I don't care about overclocking but that's good for those who do.

 

[csendesmark]

It going to hit hard!

 

[Onasi]

I assume this is single core? In any case, not likely to actually be an improvement in real world compared to just regular boost with some PBO.

 

[nvidiaenjoyer]

how does geekbench respond to cache? does it see any gains from it at all or no?

if it does, then zen 5 is legit memory bandwidth starved, if adding more cache (and therefore reducing trips to ram) provides disproportionate gains in the bench scores compared to zen 4

if that's the case then im definitely looking forward to zen 6 with a better memory controller and, hopefully, improved infinity fabric - a 11950x3d with these improvements would be sweet

 

[Seeprime]

How much additional power does it take to get the higher results? I don't want to need my own power plant to get better performance.

 

[phints]

This is neat proof that it works, but the way these Ryzen CPUs are so efficiently designed the ideal setup is PBO + Curve Optimizer set in BIOS.

 

[DemonicRyzen666]

who wrote this?

SRAM doesn't increase bandwidth.

Larger L3 caches increased the hit rate with less cache misses.

 

[kondamin]

who wrote this?

SRAM doesn't increase bandwidth.

Larger L3 caches increased the hit rate with less cache misses.

SRAM bandwidth compared to zen 4, as in more via's between the ccd and the sram chip would allow for more bandwidth between both chips.

 

[DemonicRyzen666]

SRAM bandwidth compared to zen 4, as in more via's between the ccd and the sram chip would allow for more bandwidth between both chips.

That was a design change for Zen 5 architecture though.
All caches on Zen 5 have double the bandwidth of Zen 4.

 

[Code3]

I've always said that the 7800x3d was puny

 

[Vayra86]

I've always said that the 7800x3d was puny

Let's see what the power draw is at first Keeping in mind AMD already stretched the clocks by a good 200mhz.

I kinda like the X3D's efficiency. The hard single thread perf isn't always the limiting factor, 7800X3D already bumps into engine limits more often than not. It remains to be seen how much of this % translates into meaningful performance. But yeah, its good AMD has a product that separates itself from their previous one nonetheless.

 

[Frick]

I've posted this a lot lately, but

I'm really quite interested how this performs in sim games.

 

[AnotherReader]

That was a design change for Zen 5 architecture though.
All caches on Zen 5 have double the bandwidth of Zen 4.

L3 bandwidth hasn't quite doubled though it comes much closer to its theoretical peak now. It's pretty amazing considering that the per core interface to L3 is still 256 bits wide.

 

[ShiBDiB]

So an overclocked next generation cpu outperforms the last generation at stock clocks?

What in lazy reporting is this.

 

[phanbuey]

So an overclocked next generation cpu outperforms the last generation at stock clocks?

What in lazy reporting is this.

The last gen 'stock' part is a non-overclockable part- so 9800x3d can flat out just outperform the fastest gaming CPU on the market by 24% -- which is nice really.

 

[Frick]

So an overclocked next generation cpu outperforms the last generation at stock clocks?

What in lazy reporting is this.

We did get some actual CB numbers though. The bit about OC is inconsequential, as most overclocking is these days.

It's a non-overclockable part- it flat out just outperforms the fastest gaming CPU on the market by 24% -- which is nice really.

The 9800x3d will be overclockable though, I thought that was one of its "things".

 

[ShiBDiB]

It's a non-overclockable part- it flat out just outperforms the fastest gaming CPU on the market by 24% -- which is nice really.

That's... not true?

Yes it outperforms the last generation while stock, which I'd certainly hope (looking at you Intel). But it also is clearly overclocked.

 

[phanbuey]

I meant the 'last generation part at stock clocks' was the non-overclockable.

The 9800x3d does overclock -- I was responding to a post.

 

[Makaveli]

So I was right the 5.2 that we have seen was before PBO.

 

[ShiBDiB]

I meant the 'last generation part at stock clocks' was the non-overclockable.

The 9800x3d does overclock -- I was responding to a post.

Ya you responded to my reply and then completely changed yours to the opposite of what you said after I pointed out you being wrong lol.

 

[freeagent]

Good timing, I have a job interview tomorrow

 

[phanbuey]

Ya you responded to my reply and then completely changed yours to the opposite of what you said after I pointed out you being wrong lol.

I reworded it to be clearer. My point was your:
"So an overclocked next generation cpu outperforms the last generation at stock clocks?"
is really:
"So an overclocked next generation cpu outperforms the last generation at ANY / MAX clocks?"

-yes, and that's actually pretty exciting considering that last gen X3D is currently the fastest in gaming.

 

[oxrufiioxo]

Good timing, I have a job interview tomorrow

 

[RaceT3ch]

...Well my predictions were close...