Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)

Page 208 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

adroc_thurston

Diamond Member
Jul 2, 2023
3,322
4,790
96

Markfw

Moderator Emeritus, Elite Member
May 16, 2002
25,742
14,772
136
No the recent AMD-blocks-DLSS FUD.
AMD utterly failed to control the narrative and he's like the head honcho whose job is to do just that.

Yeah most people who want 24C+ probably have money for a TR.
But I can get Genoa cheaper than threadripper on ebay.... Building my 5th Genoa box right now.
 

Timmah!

Golden Member
Jul 24, 2010
1,463
729
136
I hate to agree since I am always wanting more cores. But there is threadripper for mainstream "moar cores" and then server .
I just happen to think, that after 3 generations topping at 16 cores and soon another one to be added, it would be nice if from Zen6 on, the desktop ceiling got increased to 24C and TR then started at 32. Especially, since the server core count gets increased by each passing gen, from 64 to 96 and probably 128 or whatever number in the future.
 

H433x0n

Golden Member
Mar 15, 2023
1,068
1,272
96
I just happen to think, that after 3 generations topping at 16 cores and soon another one to be added, it would be nice if from Zen6 on, the desktop ceiling got increased to 24C and TR then started at 32. Especially, since the server core count gets increased by each passing gen, from 64 to 96 and probably 128 or whatever number in the future.
There’s nothing to gain going past 16C without an expansion of memory channels and PCIe lanes. The consumer doesn’t even buy many 16C processors, the best selling SKUs are still 8C.

I could see an argument for it if the highest selling SKU was the 7950X but I don’t even know if it’s in the top 3 of Zen 4. I know that B650 outsells X670 by like 3 to 1, so that doesn’t really show a ton of demand for these higher end platforms.
 
Reactions: Tlh97 and moinmoin

Markfw

Moderator Emeritus, Elite Member
May 16, 2002
25,742
14,772
136

Glo.

Diamond Member
Apr 25, 2015
5,761
4,666
136
I'd rather have large big.LITTLE combo, in terms of core counts, even something as stupid as 8+32, rather than just plain 16-24 performance cores.
 
Reactions: Fjodor2001

TESKATLIPOKA

Platinum Member
May 1, 2020
2,428
2,914
136
I'd rather have large big.LITTLE combo, in terms of core counts, even something as stupid as 8+32, rather than just plain 16-24 performance cores.
You mean 8 Zen5 + 32 Zen5C or the Intel version?
If It's the first, I can understand that, but If It's the latter, then doesn't make sense. It would perform like 24 performance cores and at some threads count you will even lose performance.
 

Joe NYC

Platinum Member
Jun 26, 2021
2,331
2,942
106
I just happen to think, that after 3 generations topping at 16 cores and soon another one to be added, it would be nice if from Zen6 on, the desktop ceiling got increased to 24C and TR then started at 32. Especially, since the server core count gets increased by each passing gen, from 64 to 96 and probably 128 or whatever number in the future.

If you look at DIY sales chart, and this is typically people who buy higher performing PCs / CPUs, the volume of sales falls dramatically past 8 cores. 12 - 16 cores are low volume.

There would be another drop like that, from 16 cores to 24 cores, leaving it with very low volume.

That potential volume of sales is what determines if AMD introduces 24 core CPUs. As long as it is miniscule, and AMD would lose money on it, it is not going to happen. If there is a sudden resurgence of 16 cores, there maybe hope for 24 cores.

One place where it could come from is if AMD decides to go after low end servers. For now, AMD (Forrest Norrod) calls that segment "low calorie". We will see if it changes...
 

Joe NYC

Platinum Member
Jun 26, 2021
2,331
2,942
106
I just happen to think, that after 3 generations topping at 16 cores and soon another one to be added, it would be nice if from Zen6 on, the desktop ceiling got increased to 24C and TR then started at 32. Especially, since the server core count gets increased by each passing gen, from 64 to 96 and probably 128 or whatever number in the future.

Forgot to add those sales charts from DIY market in Germany:

PS: I made a prediction that 7800x3d will cause the socket AM5 to exceed AM6 by end of June. Well, I was 1/4 right. It is the 7800x3d that caused the AM5 to finally outsell AM4, but it happened 5 months later, at the end of November.

 
Reactions: lightmanek

Abwx

Lifer
Apr 2, 2011
11,167
3,862
136
Forgot to add those sales charts from DIY market in Germany:

PS: I made a prediction that 7800x3d will cause the socket AM5 to exceed AM6 by end of June. Well, I was 1/4 right. It is the 7800x3d that caused the AM5 to finally outsell AM4, but it happened 5 months later, at the end of November.


There was a surge due to black friday but AM4/AM5 are still quite close, here the graph he posted :

 
Last edited:

Saylick

Diamond Member
Sep 10, 2012
3,385
7,151
136
I'd rather have large big.LITTLE combo, in terms of core counts, even something as stupid as 8+32, rather than just plain 16-24 performance cores.
Funny, I think I would prefer something like the opposite of 8+32, or said in another way, I think I would prefer something closer to what Apple does for their Mac SOCs: a handful of efficiency cores with the bulk of the CPU as big cores, like a 12+4. If I truly had an embarrassingly parallel workload, I think I'd want either a GPU to do the work or at least equal performance on a per-core basis.
 

Glo.

Diamond Member
Apr 25, 2015
5,761
4,666
136
Funny, I think I would prefer something like the opposite of 8+32, or said in another way, I think I would prefer something closer to what Apple does for their Mac SOCs: a handful of efficiency cores with the bulk of the CPU as big cores, like a 12+4. If I truly had an embarrassingly parallel workload, I think I'd want either a GPU to do the work or at least equal performance on a per-core basis.
This could also fit the bill.
 

Glo.

Diamond Member
Apr 25, 2015
5,761
4,666
136
You mean 8 Zen5 + 32 Zen5C or the Intel version?
If It's the first, I can understand that, but If It's the latter, then doesn't make sense. It would perform like 24 performance cores and at some threads count you will even lose performance.
For me? Even 4Z5+8Z5C wouls be enough, if it would provide power efficiency.
 

moinmoin

Diamond Member
Jun 1, 2017
4,994
7,765
136

moinmoin

Diamond Member
Jun 1, 2017
4,994
7,765
136
Whats odd, it says I bought it, so I know its the right post, but mine is on a supermicro motherboard !
That's great for you. Maybe not all of the 4 chips offered were locked to Dell and you were lucky to get one that isn't. Or the seller included this line just to be on the safe side just in case but didn't have the capacity to verify it themselves.
 

DrMrLordX

Lifer
Apr 27, 2000
21,802
11,157
136
I hate to agree since I am always wanting more cores. But there is threadripper for mainstream "moar cores" and then server .
Mainstream is going to struggle with memory bandwidth as core counts go higher, which is probably one of the reasons why we'll be stuck at 16C on AMD platforms for awhile. The costs associated with building motherboards for quad channel (or more!) would irritate OEMs. Never mind that mainstream boards can easily hit $1k now . . .
 

Mopetar

Diamond Member
Jan 31, 2011
8,005
6,449
136
Adding a lot of v-cache to higher core-count CPUs would do a lot to alleviate bandwidth issues from a lack of additional memory channels. The other possibility is to go with LPDDR memory that can operate at higher frequencies and provide more bandwidth on the same bus, but I don't know if the market that wants more cores is down for soldered memory.
 
Reactions: igor_kavinski
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |