AMD Ryzen 9000 Zen 5 Single Thread Performance at 5.80 GHz Found 19% Over Zen 4

[btarunr]

An AMD Ryzen 9000 "Granite Ridge" desktop processor engineering sample with a maximum boost frequency of 5.80 GHz was found to offer an astonishing 19% higher single-threaded performance increase over an AMD Ryzen 9 7950X. "Granite Ridge" is codename for the Socket AM5 desktop processor family that implements the new "Zen 5" CPU microarchitecture. The unnamed "Granite Ridge" processor comes with an OPN code of 100-0000001290. Its CPU core count is irrelevant, as the single-threaded performance is in question here. The processor boosts up to 5.80 GHz, which means the core handling the single-threaded benchmark workload is achieving this speed. This speed is 100 MHz higher than the 5.70 GHz that the Ryzen 9 7950X processor based on the "Zen 4" architecture, boosts up to.

The single-threaded benchmark in question is the CPU-Z Bench. The mostly blurred out CPU-Z screenshot that reveals the OPN also mentions a processor TDP of 170 W, which means this engineering sample chip is either 12-core or 16-core. The chip posts a CPU-Z Bench single-thread score of 910 points, which matches that of the Intel Core i9-14900K with its 908 points. You've to understand that the 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 straight off the bat, we see that "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.



Many Thanks to TumbleGeorge for the tip.

View at TechPowerUp Main Site | Source

 

[thesmokingman]

Now that's an enticing bump if true. Time for the leaks...

 

[R0H1T]

Should wait for better benches, any way IMO the biggest show stopper would still be Strix point/Halo with up to 256bit wide memory support!

 

[a111087]

another site that was reporting on this said that 14900k gets 942 points... they even complained that amd is still behind 14900k
which one is it?

 

[thesmokingman]

another site that was reporting on this said that 14900k gets 942 points... they even complained that amd is still behind 14900k
which one is it?
View attachment 349096

It's hard to compare from different sources and obviously bring salt. Anyways onto more leaks...

 

[Tomorrow]

Intel has had historically higher scores than AMD in CPU-Z. Wait for more tests (not Geekbench etc) to get a clearer picture.
Also this was 14900K boosting to 6Ghz where as 9950X reportedly boosts to 5.8Ghz (likely at lower power than 14900K , too).

 

[Onasi]

It’s a single synthetic benchmark, so we’d have to take it for what it is. But seeing as how Zen 5 is a major architecture upgrade a 15-20% clock for clock ST uplift is more than in line with what’s expected. This isn’t exactly that since a 100 MHz difference is there, but overall… it tracks.

 

[john_]

At 5.8GHz doesn't just equals 14900K. It equals an overclocked and unstable 14900K.
Also CPU-z benchmark is for years considered one of the Intel friendly ones.

While I doubt, I hope AMD to be considering bringing the X3D chips the same day with the regular ones. They can put a ridiculous high price if they want on them, but it will be stupid if they don't announce them together with the regular ones. They have to finally start understanding the power of marketing. Zen 5 will have a totally different, much higher level of acceptance, if an 8 core 9800X3D annihilates everything in gaming benchmarks with differences of 20-50%. If they fear internal competition, they can start that chip at $550. Zen 4 and AM5 would had much higher success if the X3D chips where introduced together with the new platform.

 

[wolf]

Interesting if true, should mean the non X3D chips batt closer to a 14900K in games, without the power draw and instability. Then X3D chips will be something else again. Also looking forward to the bigger GPU / 256 bit memory chips based on this.

 

[hsew]

I just hope power and thermals are good… The never-ending pursuit of diminishing returns is getting old. Intel has somehow managed to drag AMD into a new, pointless clock-speed race starting with Zen 3. I hope they’ve managed to keep things under control for Zen 5.

 

[R0H1T]

Depends on how much Intel OC's Arrow(Panther?) lake

 

[john_]

Intel has somehow managed to drag AMD into a new, pointless clock-speed race starting with Zen 3.

AMD tried to promote it's chips as super efficient. They did that keeping 12 and 16 core chips at 8 core chips power consumption levels. Then users online where praising Intel's chips for being 1% faster in single threaded benchmarks and games while using twice the power. What was expected from AMD to do, other than offer users what they wanted? That +1% performance for a +50% power increase.
Intel didn't drag AMD to anything. Users and tech press did. They are so desperate to keep offering wins to Intel, that they made efficiency look like a secondary, unimportant feature.

another site

Tom's is pro-Intel at least 20+ years now. The title that, that article uses, is what Intel trolls post left and right from yesterday.

 

[Bwaze]

CPU-Z Bench isn't really a very good representative of games, and as others have noticed, that score is more in line with a 13900k...

 

[Crackong]

another site that was reporting on this said that 14900k gets 942 points... they even complained that amd is still behind 14900k
which one is it?

I suppose that 14900k isn't running at Intel baseline ?

 

[Makaveli]

At 5.8GHz doesn't just equals 14900K. It equals an overclocked and unstable 14900K.
Also CPU-z benchmark is for years considered one of the Intel friendly ones.

While I doubt, I hope AMD to be considering bringing the X3D chips the same day with the regular ones. They can put a ridiculous high price if they want on them, but it will be stupid if they don't announce them together with the regular ones. They have to finally start understanding the power of marketing. Zen 5 will have a totally different, much higher level of acceptance, if an 8 core 9800X3D annihilates everything in gaming benchmarks with differences of 20-50%. If they fear internal competition, they can start that chip at $550. Zen 4 and AM5 would had much higher success if the X3D chips where introduced together with the new platform.

You will not be getting X3D chips the same time as regular just based on production. I don't see this changing!

 

[Cifu]

For the record: the 2CCD solutions historically have had the upper hand in boost clock:

7950X: 5750Mhz
7700X: 5550Mhz
7700: 5350Mhz
7800X3D: 5050Mhz

Not to mention the AMD firstly let the Zen4 to boost even more with the base 100Mhz clock tuning. This door was closed after the AGESA 1.0.0.3, after that only the eCLK tuning remained.

 

[A Computer Guy]

another site that was reporting on this said that 14900k gets 942 points... they even complained that amd is still behind 14900k
which one is it?
View attachment 349096

Is that in consideration of Intel's new baseline profile? (I'm guessing probably not)

 

[R0H1T]

I suppose that 14900k isn't running at Intel baseline ?

Spoiler

 

[stimpy88]

When AMD first released the Zen2 architecture, CPU-Z's author (or Intel) decided that he didn't like the Zen2 out-performing the Intel chip at the time, so a new benchmark version was released, reducing the AMD scores (Intel scores stayed the same) by some 15%. I have never taken the CPU-Z benchmark seriously after that, as it's apparently just an Intel sponsored benchmark.

 

[ARF]

When AMD first released the Zen2 architecture, CPU-Z's author (or Intel) decided that he didn't like the Zen2 out-performing the Intel chip at the time, so a new benchmark version was released, reducing the AMD scores (Intel scores stayed the same) by some 15%. I have never taken the CPU-Z benchmark seriously after that, as it's apparently just an Intel sponsored benchmark.

Absolutely. After version 1.78.3, CPU-Z became extremely uninteresting, hence I accept 1.78.3 as the "true" CPU-Z benchmark.

AMD Ryzen 7 1800X Review (Page 9)

We review AMD Ryzen 7 1800X Processor. The 8-core processor will be tested on an X370 motherboards AMDs most high-end processor SKU really is a top dog being amd enthusiast class eight-core processor!...

www.guru3d.com

Nevertheless, I think AMD is extremely passive and doesn't do anything in its fundamental interests.
Why not release an AMD sponsored benchmark called x86-64-Z, or something like this?

 

[Vayra86]

Absolutely. After version 1.78.3, CPU-Z became extremely uninteresting, hence I accept 1.78.3 as the "true" CPU-Z benchmark.

AMD Ryzen 7 1800X Review (Page 9)

We review AMD Ryzen 7 1800X Processor. The 8-core processor will be tested on an X370 motherboards AMDs most high-end processor SKU really is a top dog being amd enthusiast class eight-core processor!...

www.guru3d.com

Nevertheless, I think AMD is extremely passive and doesn't do anything in its fundamental interests.
Why not release an AMD sponsored benchmark called x86-64-Z, or something like this?

How is that good advertising? 'Look we have an optimized for AMD benchmark here, let's see if we win!'

It is much more powerful advertising to beat Intel on their own turf. A truly good processor should be able to, right? Now that doesn't mean I'm saying use CPU Z. But you wouldn't want 'AMD-Z' for the very same reason.

 

[ARF]

How is that good advertising?

Yes.

'Look we have an optimized for AMD benchmark here, let's see if we win!'

It is much more powerful advertising to beat Intel on their own turf. A truly good processor should be able to, right? Now that doesn't mean I'm saying use CPU Z. But you wouldn't want 'AMD-Z' for the very same reason.

CPU-Z can always release a new version in which the intel processors are up to 100% faster. Don't underestimate them.
What I want is a truely neutral benchmark, which can happen only if it is AMD sponsored.

 

[londiste]

AMD tried to promote it's chips as super efficient. They did that keeping 12 and 16 core chips at 8 core chips power consumption levels.

Which 8-core chips runs at 230W power limit? That is what 7900X and 7950X have.

 

[Bwaze]

It's strange that there are so different results out there for stock 14900k in reviews - is this again motherboard schenanigans overcklocking at default settings? The most widely published number is also a pre-release "leak" from September 2023, with the score of 978, clearly an overclock.

 

[napata]

Is that in consideration of Intel's new baseline profile? (I'm guessing probably not)

It's not but it's a single thread workload so it's not as if you go above 253W in those on a 14900K. CPU-Z single thread is irrelevant anyway.