News Ryzen 9000 Zen 5 CPU trails Core i9-14900K in leaked benchmark — Granite Ridge 5.8 GHz CPU shows Core i9-13900K-like single-threaded performance in...

Article:: ""The leaker claims that the Zen 5 processor had a boost clock speed of 5.8 GHz, the highest we've seen from a mainstream Ryzen chip. For comparison, the Ryzen 9 7950X, the current flagship, has the top boost clock speed of 5.7 GHz, so we're looking at a 100 MHz improvement on the unidentified Zen 5""

That's because for Zen 5, AMD has changed the Infinity Fabric interconnect for the CCD-CCD and IOD communication, and the IMC (Memory Controller) has been given an upgrade as well.

Hence the CPU runs at a frequency 100 MHz faster than the Ryzen 9 7950X.

Also, since this is a 170W TDP engineering sample, this is 100% either a 16-core or a 12-core SKU. 8-core chip seems out of equation.

That's because ZEN 5 CPU lineup is supposed to feature the same core count as the current Ryzen 7000 series, with 16, 12, 8 and 6 cores. There won't be any smaller or dense ZEN 5c cores at least initially.

Only Zen 5 cores.

AMD Ryzen Granite Ridge CPUs should be available in four configurations include the top 16 core part, & followed by 12, 8, and 6 core variants.

https://www.chiphell.com/forum.php?mod=viewthread&tid=2608542&extra=page=1&mobile=no

fo2mKEf.png
 
That's because for Zen 5, AMD has changed the Infinity Fabric interconnect for the CCD-CCD and IOD communication, and the IMC (Memory Controller) has been given an upgrade as well.

Hence the CPU runs at a frequency 100 MHz faster than the Ryzen 9 7950X.

Also, since this is a 170W TDP engineering sample, this is 100% either a 16-core or a 12-core SKU. 8-core chip seems out of equation.

That's because ZEN 5 CPU lineup is supposed to feature the same core count as the current Ryzen 7000 series, with 16, 12, 8 and 6 cores. There won't be any smaller or dense ZEN 5c cores at least initially.

Only Zen 5 cores.

AMD Ryzen Granite Ridge CPUs should be available in four configurations include the top 16 core part, & followed by 12, 8, and 6 core variants.

https://www.chiphell.com/forum.php?mod=viewthread&tid=2608542&extra=page=1&mobile=no

fo2mKEf.png
How can we say this was not the 9700X 8 core part? They only tested single-threaded performance. Was one of the earlier tests associated with the part number the 12-core or 16 core part? The 170w TDP does not mean it was impossible for this test to have been on an 8c part, no? Either way I am happy with a 18.5% increase in ST.
 
Last edited:
The leaker also stated that the "non-X3D chip easily takes out Zen 4's X3D" but didn't provide cache-sensitive or gaming benchmarks to corroborate those claims.

That's because the standard Zen 5 CPUs are going to outperform Zen 4 3D V-Cache chips in gaming, due to improvements to Zen 5 architecture, and other upgrades like Infinity Fabric interconnect and IMC .

Ryzen 9000 Zen 5 CPU trails Core i9-14900K in leaked benchmark


The Zen 5 engineering sample was right up there with the previous Core i9-13900K, but it lags behind the newer Core i9-14900K, which produced a 3.5% higher single-threaded score.

First of all you need to realize this is an early sample, and it is likely that with the final silicon, the chip's performance should be much better with firmware support. And I also expect the IPC to end up close to 15%.

i9-14900K boosts one of its P-cores to 6.00 GHz, to yield the 908 points that's part CPU-Z's reference scores.

So "Zen 5" has a higher IPC than the "Raptor Cove" P-core powering the i9-14900K. Its gaming performance might end up higher than the Ryzen 7000 X3D family.

Also, we should never trust CPU-Z scores, at least for AMD Ryzen CPU benchmarks, Zen architecture.

CPU-Z app never took advantage of Zen4's improvements to the arch like, micro-op cache, branch prediction, L2 cache capacity etc, but other apps did. I expect the same with ZEN 5.

With AMD Excavator lineup to Zen 1: There was roughly 52% IPC Increase
  • Zen 1 -Zen+: 3% IPC Increase
  • Zen --Zen 2: 15% IPC Increase
  • Zen --Zen 3: 19% IPC Increase
  • Zen 3 -- Zen 4: 13% IPC Increase
  • Zen 4 -- Zen 5: ~10-15% IPC increment ? (speculation)
 
Last edited by a moderator:
Also, never trust CPU-Z benchmark, at least for AMD Ryzen CPU benchamrks.
  • AMD Excavator to Zen 1: 52% IPC Increase
  • Zen 1 To Zen+: 3% IPC Increase
  • Zen + To Zen 2: 15% IPC Increase
  • Zen 2 To Zen 3: 19% IPC Increase
  • Zen 3 To Zen 4: 13% IPC Increase
  • Zen 4 To Zen 5: ~10-15% IPC Increase
If AMD Excavator was a baseline of 100 that means:

Zen1 is 152
Zen+ is 156.56
Zen2 is 180.04
Zen3 is 214.25
Zen4 is 242.11
Zen5 is purported to be (10%-15%) 266.32-278.42
 
Even if someone doesn't want to believe the CPU-Z benchmark, when you compare AMD to AMD instead of Intel to AMD, the jump looks really good to me. Extrapolate this to the MT score and it seems like they'll increase overall metrics by a substantial margin (or I'd hope they do).

Grain of salt and all, but to me this is not a negative in the bigger picture.

Regards.
 
First of all you need to realize this is an early sample, and it is likely that with the final silicon, the chip's performance should be much better with firmware support. And I also expect the IPC to end up close to 15%.
This line is parroted every time someone doesn't like the leaked benchmark result of a prerelease CPU of their favorite team, and it almost always ends up that the benchmark is pretty close to accurate. AMD is scheduled to announce Zen5 next week at Computex. It's not getting released then, but we sure aren't looking at "early engineering" samples at this point in the development time line.
 
  • Like
Reactions: artk2219
This line is parroted every time someone doesn't like the leaked benchmark result of a prerelease CPU of their favorite team, and it almost always ends up that the benchmark is pretty close to accurate. AMD is scheduled to announce Zen5 next week at Computex. It's not getting released then, but we sure aren't looking at "early engineering" samples at this point in the development time line.

Yes, I know that might sound like a broken record, and I also agree with your point, but the CPU-Z screenshot itself says that this CPU is an engineering sample.

AMD Eng Sample: "100-0000001290"

AMD-Ryzen-9000-Granite-Ridge-Zen-5-Desktop-CPU.jpg
 
  • Like
Reactions: artk2219
Well as I sit here with my cursed 105w 5950X (that won't boost worth crap, 4.49ghz all core and 4.68ghz single core in CPU-Z test even with removed PBO limits) scoring 632/12593 ST/MT in the bench vs the "reference 5950X" of 648/11906, and the 647 ST score mentioned in the article, I wouldn't say it's exactly consistent, and it also doesn't -mean- anything.

I'm also struggling to come up with a sales pitch. Sure it's faster than Zen 3 by a couple dozen percent, but is it really worth having to also buy a new motherboard and RAM as well? Same with Zen 4, it may be 15% faster but unless you're rocking a 4080 or 4090 are you really going to notice any difference in games, if then? Similarly, if someone is rocking a pre-Pandemic system what's the argument to sway them to Socket AM5 vs AM4 or Intel LGA whatever the last generation or so was?
 
Yes, I know that might sound like a broken record, and I also agree with your point, but the CPU-Z screenshot itself says that this CPU is an engineering sample.

AMD Eng Sample: "100-0000001290"

AMD-Ryzen-9000-Granite-Ridge-Zen-5-Desktop-CPU.jpg
We're certainly not looking at a retail sample here, but it also doesn't say anything about early engineering sample either. We're close enough to release at this point that engineering samples that are floating around should be pretty representative of final performance. This is still a Zen architecture CPU, we're not looking at a completely new architecture that will require significant performance optimizations.
 
  • Like
Reactions: artk2219
I'm also struggling to come up with a sales pitch. Sure it's faster than Zen 3 by a couple dozen percent, but is it really worth having to also buy a new motherboard and RAM as well? Same with Zen 4, it may be 15% faster but unless you're rocking a 4080 or 4090 are you really going to notice any difference in games, if then? Similarly, if someone is rocking a pre-Pandemic system what's the argument to sway them to Socket AM5 vs AM4 or Intel LGA whatever the last generation or so was?
Why release annual car models that are usually slight tweaks of what was sold the year before? Why would anyone upgrade from last year's model? They wouldn't. The average age of car in the US is 12.6 years old. That's why there is a new model released every year, so the people that are upgrading feel like they are getting something brand new. The average age of a PC is probably 5 years or more old. For someone upgrading something that old and expecting the new system to go another 5 years or more, that's who Zen 5 will be marketed to.
 
Why release annual car models that are usually slight tweaks of what was sold the year before? Why would anyone upgrade from last year's model? They wouldn't. The average age of car in the US is 12.6 years old. That's why there is a new model released every year, so the people that are upgrading feel like they are getting something brand new. The average age of a PC is probably 5 years or more old. For someone upgrading something that old and expecting the new system to go another 5 years or more, that's who Zen 5 will be marketed to.
Someone like me. I'm stuck with an ancient Elitebook with a i7 8665u. Quad-core and slow AF! I'm excited for this release to start building my new AM5 based PC... Hahaha this laptop is ancient! Lol
 
  • Like
Reactions: artk2219
Well as I sit here with my cursed 105w 5950X (that won't boost worth crap, 4.49ghz all core and 4.68ghz single core in CPU-Z test even with removed PBO limits) scoring 632/12593 ST/MT in the bench vs the "reference 5950X" of 648/11906, and the 647 ST score mentioned in the article, I wouldn't say it's exactly consistent, and it also doesn't -mean- anything.

I'm also struggling to come up with a sales pitch. Sure it's faster than Zen 3 by a couple dozen percent, but is it really worth having to also buy a new motherboard and RAM as well? Same with Zen 4, it may be 15% faster but unless you're rocking a 4080 or 4090 are you really going to notice any difference in games, if then? Similarly, if someone is rocking a pre-Pandemic system what's the argument to sway them to Socket AM5 vs AM4 or Intel LGA whatever the last generation or so was?
This release is for people like me. I'm stuck on an ancient Elitebook with a i7 8665u quad core. It's slow AF and it's time for something modern. So I am excited and ready for Ryzen 9000 to build an AM5 based PC. And not everyone is using CPUs for gaming. So that point is moot as well.
 
  • Like
Reactions: artk2219
So, we are now preemptively trashing another AMD product even before launching?

Ok, fine, i’ll bite, how much faster is your beloved intel?

3.5%…..aaannnd thats it?

This is sad.
 
  • Like
Reactions: artk2219
Well as I sit here with my cursed 105w 5950X (that won't boost worth crap, 4.49ghz all core and 4.68ghz single core in CPU-Z test even with removed PBO limits) scoring 632/12593 ST/MT in the bench vs the "reference 5950X" of 648/11906, and the 647 ST score mentioned in the article, I wouldn't say it's exactly consistent, and it also doesn't -mean- anything.

I'm also struggling to come up with a sales pitch. Sure it's faster than Zen 3 by a couple dozen percent, but is it really worth having to also buy a new motherboard and RAM as well? Same with Zen 4, it may be 15% faster but unless you're rocking a 4080 or 4090 are you really going to notice any difference in games, if then? Similarly, if someone is rocking a pre-Pandemic system what's the argument to sway them to Socket AM5 vs AM4 or Intel LGA whatever the last generation or so was?
I will be going Zen 5 from Zen 3 I've been on my build for 5 years now. I want to flip before the resale value is totally down the drain.

And looks like I will be getting a 30% ipc increase in the jump so well worth it. I'm already good on gpu for now so this upgrade is only CPU / Ram / Motherboard rest of my build stays the same.
 
How can we say this was not the 9700X 8 core part? They only tested single-threaded performance. Was one of the earlier tests associated with the part number the 12-core or 16 core part? The 170w TDP does not mean it was impossible for this test to have been on an 8c part, no? Either way I am happy with a 18.5% increase in ST.
While technically possible, 170W 8-core would be a huge departure from previous generations.
8-core 8700G or 7700 are rated at 65W.
The 7800X3D is the only one at 120W. The 7800X at 105.
170W is a ridiculous amount for 8-core and only Intel has pushed the consumption so high for low-core count CPUs, and precisely to achieve this 3.5% advantage in single core.

What's missing in this article is the power consumption indeed when running single-core bench. Intel so far has a slight advantage but burning >2X the power.