Caporegime
- Joined
- 8 Jan 2004
- Posts
- 32,586
- Location
- Rutland
Windows seems pretty decent at this now. Whenever I'm gaming on my 7950X3D i can see it park an entire CCD without issue.
-
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
- Thread starter Grim5
- Start date
Associate
- Joined
- 11 Sep 2020
- Posts
- 871
- Location
- London
Good to know. Can you recall when it improved/was ‘fixed’? I remember the issues around launch but haven’t followed it since.
AMD Implements New CCD Connection in "Strix Halo" Ryzen AI Max Processors
Thanks to the informative breakdown by Chips and Cheese, we are learning that AMD's latest Ryzen AI processors for laptops, codenamed "Strix Halo," utilize a parallel "sea of wires" interconnect system between their chiplets, replacing the SERDES (serializer/deserializer) approach found in...
reads like quite a big change to Zen chiplet design
reduce inter-CCD latency and improve power consumption (in low power states presumably)
AMD Zen 6 rumours
Some early rumours coming in; this one not too major but it looks like OpenSil is nearing ready for deployment and Zen6 will be the Ryzen range to use OpenSil, before Agesa is completely dropped https://wccftech.com/amd-preps-replacement-agesa-opensil-next-gen-zen-6-ryzen-epyc/
forums.overclockers.co.uk
Thank you!AMD Zen 6 rumours
Some early rumours coming in; this one not too major but it looks like OpenSil is nearing ready for deployment and Zen6 will be the Ryzen range to use OpenSil, before Agesa is completely dropped https://wccftech.com/amd-preps-replacement-agesa-opensil-next-gen-zen-6-ryzen-epyc/
Would be great if true, might get a 24c chip variant if thats the case
Per core vs per core maximizing offsets PBO
if I push my better cores to lower PBO some of the worse cores need a reduced negative offset not sure why why
Say a worse core might need -4 when other cores at their best negative but when not so optimized for best cores he worse ones might need way -10
If this is the case does this mean that in a completely optimised and stablE PBO vs non optimised per core PBO
The completely optimised one as some cores have lower negative offset say -34 vs -26 - would be faster for gaming where a few cores are used
But in rendering/cinebench/Prime95 where all cores completely loaded the most optimized per core PBO will be slower as your worse cores have smaller offsets ?
I can't remember exacts but with my 7800x3d something like the following two are stable
-29, -30, -4, -19, -27, -30, -12, -34
Vs
-28, -28, -11, -12, -26, -26, -11, -28
I'm guessing scenario 2 faster for productive scenario 1 for gaming
When running prime95 for number 1 the clock speeds are lower than for number 2 as when all fully loaded clock is max the worse core
You'd think in case two say core 8 would run at -34 but it won't unless core 3 is above say -7 and then all cores stable when that core 3 is at -4
There must be some odd relative voltages or loads between cores
Note both above were Prime95 Small stable for over 5 hours and same again for core cycler
if I push my better cores to lower PBO some of the worse cores need a reduced negative offset not sure why why
Say a worse core might need -4 when other cores at their best negative but when not so optimized for best cores he worse ones might need way -10
If this is the case does this mean that in a completely optimised and stablE PBO vs non optimised per core PBO
The completely optimised one as some cores have lower negative offset say -34 vs -26 - would be faster for gaming where a few cores are used
But in rendering/cinebench/Prime95 where all cores completely loaded the most optimized per core PBO will be slower as your worse cores have smaller offsets ?
I can't remember exacts but with my 7800x3d something like the following two are stable
-29, -30, -4, -19, -27, -30, -12, -34
Vs
-28, -28, -11, -12, -26, -26, -11, -28
I'm guessing scenario 2 faster for productive scenario 1 for gaming
When running prime95 for number 1 the clock speeds are lower than for number 2 as when all fully loaded clock is max the worse core
You'd think in case two say core 8 would run at -34 but it won't unless core 3 is above say -7 and then all cores stable when that core 3 is at -4
There must be some odd relative voltages or loads between cores
Note both above were Prime95 Small stable for over 5 hours and same again for core cycler
these give a quick test of stability, but not 100% confirmation
then in two days you will get a random crash out of a multiplayer game and add +5 to all of above just in case
Sometime in March but no specific day yet.
Oh jeez still got a while to go then!
Associate
- Joined
- 9 Jan 2009
- Posts
- 48
- Location
- isle of wight
Some of the most recent articles say January, some day March. Waiting on the 9900x3d myself but all my Lian li stuff are still on pre order and looks like end of February for 5070ti so happy to wait. Be worth the upgrade from an i7 5960x and gtx 980 sli.any eta on 9950x3d? AMD are sure taking their time
AMD exec reveals Ryzen 9 9950X3D, Ryzen 9 9900X3D gaming performance — similar to Ryzen 7 9800X3D
Why would you spend extra for the same gaming performance?
Where did you read January? Their CES presser stated March 2025, and haven't read differently anywhere. Curious to see where you read that! TySome of the most recent articles say January, some day March. Waiting on the 9900x3d myself but all my Lian li stuff are still on pre order and looks like end of February for 5070ti so happy to wait. Be worth the upgrade from an i7 5960x and gtx 980 sli.
AMD exec reveals Ryzen 9 9950X3D, Ryzen 9 9900X3D gaming performance — similar to Ryzen 7 9800X3D
Why would you spend extra for the same gaming performance?www.tomshardware.com
Associate
- Joined
- 9 Jan 2009
- Posts
- 48
- Location
- isle of wight
Linked and quoted the post.
AH missed it - Thanks.
"Unfortunately, AMD hasn’t released pricing information for these chips yet, although we expect them to be released this January."
I am fairly certain that won't be the case! Wouldn't mind it though. Currently building a new rig and it's one of the last remaining pieces.
Associate
- Joined
- 9 Jan 2009
- Posts
- 48
- Location
- isle of wight
No worries, not many days of the month left to find out. I wonder if they mean release pricing information rather than the product.