• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

*** AMD "Zen 4" thread (inc AM5/APU discussion) ***

The OC Air score was at 226 watts, the 6.5Ghz one was at 271 watts.

I'm assuming the 39,000 stock score was at 170 watts.
If it is at 170w then color me impressed. I find it unlikely, but id be happy to be wrong. Thats 40% faster than a tuned 12900k at 170w. Good stuff if true
 
Last edited:
Just gonna say it, still don't think multithreaded performance is that important. 8 cores is enough.

Show me a game that really benefits from more than 8 cores, and maybe I will change my mind.

Maybe >8 cores might be useful for some work related tasks, like rendering or compiling code?
 
Last edited:
But unless you tune both you arent putting them in similar conditions. The numbers of zen4 are from xoc with exotic cooling.
the stock doesnt mention type of cooling used ? the numbers look impressive for multi for the top end guess people who are buying them want them for multi core workloads

will be interesting to see single core , but we dont still know what this translates into gaming :)
 
Last edited:
Just gonna say it, still don't think multithreaded performance is that important. 8 cores is enough.

Show me a game that really benefits from more than 8 cores, and maybe I will change my mind.

Maybe >8 cores might be useful for some work related takes, like rendering or compiling code?
Spiderman does for sure. Ecores give me around 20+% better mins compared to just 8p cores.
 
The 5800X3D is getting 90 FPS minimums with 'only' 8 cores, ahead of the 12900K on a DDR4 setup (RT enabled).

I bet 120+ FPS will be possible on a '7800X3D'.
 
Last edited:
Its very responsive to memory bandwidth. 12900K i'm assuming DDR4: 96 FPS
12900K DDR5 6400 131 (+36%)

Zen 4 will get the same boost just from DDR5.
It scales with memory and cores cause it basically unpacks assets on the fly. The bigger zen 3s are memory bottlenecked thats why they are pretty close to the 5600x. Also ht is broken in this game, disabling them gives you a LOT of performance.

My 12900k with HT off was hitting 85-90% usage, so do the math for the 7600x. It will probably be restricted by the multicore performance.
 
There doesn't seem to be much relationship between the framerate and the number of cores.

Other factors like the L3 cache amount, IPC, clock rate and frequency of the RAM seem to count for more.

If those E-Cores on the 12900K were disabled, I doubt it would make a big difference.
 
So? Nobody said it will be unplayable. Im saying it scales beyond 6 cores, so why would you buy 6 or 8 when you can get 14 for similar or even less money?

The difference here between the i3 12100 (4 cores) and the 12900K (8 cores) is 23% and most of that will be down to clock speed.

The only standout difference is DDR4 vs DDR5.

6 cores isnt doing badly here and this is at 1080p
iVY0oAJ.jpg

1080p without RT 8 core vs 16core AMD hardly anything in it
mxXJnbW.jpg
 
Last edited:
It scales with memory and cores cause it basically unpacks assets on the fly. The bigger zen 3s are memory bottlenecked thats why they are pretty close to the 5600x. Also ht is broken in this game, disabling them gives you a LOT of performance.

My 12900k with HT off was hitting 85-90% usage, so do the math for the 7600x. It will probably be restricted by the multicore performance.

you say memory bottlenecked but look at the 5800X3D , also zen4 shouldnt be memory bottlenecked right ? soon after add in X3D on the Zen4 chips anyway not long to wait now see what the 7600xt does in this game
 
Last edited:
There doesn't seem to be much relationship between the framerate and the number of cores.

Other factors like the L3 cache amount, IPC, clock rate and frequency of the RAM seem to count for more.

If those E-Cores on the 12900K were disabled, I doubt it would make a big difference.
I tried it, it makes a huge difference. 20+% better minimums and averages with the ecores on
 
Back
Top Bottom