• 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 5 rumours

KompuKare said:
www.hardwareluxx.de

Optimierte Sprungvorhersage: Zen 5 mit Windows-Update KB5041587 getestet - Hardwareluxx

Optimierte Sprungvorhersage: Zen 5 mit Windows 11 24H2 getestet.
www.hardwareluxx.de www.hardwareluxx.de
Click to expand...

Those numbers are interesting on the Intel side, so just thought I'd run the canned benchmark on 23H2 on the same settings as listed.

Do you have the test system breakdown, so I can have a look?

Average: 237.55 fps
Min: 190 fps

1mOKiRV.png
 
Last edited:
LtMatt said:
9950X 8000Mhz CL32, 1.65 VDD.
46U9eSm.png
Click to expand...
Hi LtMatt, want to try the new version of my ray tracer on your new chip?
easyupload.io

Easyupload.io - Upload Files and Share Them Easily

Uplaod files easily and share them with your friends and colleagues. File upload solution for everyone. Upload big files and get a link to share effortlesly.
easyupload.io easyupload.io
Think its memory limited on my PC, get ~220ms (250K triangles) but only getting ~60% CPU usage.
WASD-QE move camer, curser keys rotate camer. Un-check StopRender to start rendering.
 
Last edited:
Freddie1980 said:
I see lots of numbers there but which one shows the speed of the infinity fabric?

Is there any benefit of running ram faster then then IF?
Click to expand...
FCLK 2200 is the infinity fabric
generally it is best to run highest possible stable FCLK regardless of memory speed
 
Last edited:
alec said:
I suspect many of these workloads could be found in modern games. Games just started to utilise more than 8 cores. And there is a lot of inter-thread communication there

Normal multicore loads, your video encoding and rendering tasks, barely exchange data between threads. Start on a chunk of work, give it back when done. Won't even notice this latency increase

I'm glad that AMD thinks this is fixable. Secretly hope that SMT thread latency increase would be fixed too.
Click to expand...
I really hope you're right btw. The Zen 5 parts have incredible cache hit rates with a really stunning branch predictor, but we aren't seeing the sweet benefits in a lot of workloads. Maybe that's why, maybe not. When the latencies get fix we'll find out!

One thing though, utilising more than 8 cores is not the same as one core needing to access another core's memory.

The 2x CCD X3D parts relied on core parking to ensure that the CCD running the game was close to the 3D cache to reduce latency, but that's a different case.

The SMT thread latency is definitely something I hope is fixed, because that will be coming into play regularly.
 
Freddie1980 said:
I see lots of numbers there but which one shows the speed of the infinity fabric?

Is there any benefit of running ram faster then then IF?
Click to expand...
FCLK.
FredFlint said:
Hi LtMatt, want to try the new version of my ray tracer on your new chip?
easyupload.io

Easyupload.io - Upload Files and Share Them Easily

Uplaod files easily and share them with your friends and colleagues. File upload solution for everyone. Upload big files and get a link to share effortlesly.
easyupload.io easyupload.io
Think its memory limited on my PC, get ~220ms (250K triangles) but only getting ~60% CPU usage.
WASD-QE move camer, curser keys rotate camer. Un-check StopRender to start rendering.
Click to expand...
I’ll try it when I get home.
 
You must log in or register to reply here.
Back
Top Bottom