Discussion Speculation: Zen 4 (EPYC 4 "Genoa", Ryzen 7000, etc.)

Page 335 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Vattila

Senior member
Oct 22, 2004
805
1,394
136
Except for the details about the improvements in the microarchitecture, we now know pretty well what to expect with Zen 3.

The leaked presentation by AMD Senior Manager Martin Hilgeman shows that EPYC 3 "Milan" will, as promised and expected, reuse the current platform (SP3), and the system architecture and packaging looks to be the same, with the same 9-die chiplet design and the same maximum core and thread-count (no SMT-4, contrary to rumour). The biggest change revealed so far is the enlargement of the compute complex from 4 cores to 8 cores, all sharing a larger L3 cache ("32+ MB", likely to double to 64 MB, I think).

Hilgeman's slides did also show that EPYC 4 "Genoa" is in the definition phase (or was at the time of the presentation in September, at least), and will come with a new platform (SP5), with new memory support (likely DDR5).



What else do you think we will see with Zen 4? PCI-Express 5 support? Increased core-count? 4-way SMT? New packaging (interposer, 2.5D, 3D)? Integrated memory on package (HBM)?

Vote in the poll and share your thoughts!
 
Last edited:
Reactions: richardllewis_01

poke01

Golden Member
Mar 8, 2022
1,347
1,514
106
False. Amd use whatever benchmark that suits them the most. At the moment, there isn't a single benchmark that can capture the whole performance spectrum of zen. I think, to amd, bg is the most appropriate because of avx512.
geekbench is a good tool to use for ST. AMD and Intel use it for a reason. But yes it should not be the only tool to bench.
 
Reactions: Kaluan

AtenRa

Lifer
Feb 2, 2009
14,003
3,360
136
I think 12100/12400 users need to buy new mobo for the new i7s

nope


 

biostud

Lifer
Feb 27, 2003
18,387
4,939
136
1. No competition at $100 vs i3 12100
2. 12600K currently at $239 that gives higher ST and MT vs the 5600X at $200
3. At the $300 range i7 12700F is currently at $320 that decimates the $260 5700X in both gaming and MT workloads.

Plus socket 1700 platform has higher upgradability (Raptor Lake), AM4 is dead socket.

So in October the new Ryzen 7600X at $299 will have to compete against AL 6+4 at $200 (13400). 13400 system will be even cheaper because the platform also supports DDR4.

So it comes down to this, 7600X at $299 will have to compete against i7 13xxx and not i5s when adding the cost of the platform and DDR5 memory.
And again at the sub $200 range AMD will have no answer against Intel, unless they will cut prices and make 5700X compete at the $200-220 segment. Which again will have hard time compete in gaming against the $200-220 Intel CPUs.

And good luck finding a cheap B650 motherboard this year.

Isn't the low end market 12100/12400 primarily OEM builds?

I agree that Intel have a better product at this price range, but do the customers buying computers in this price range really care that much about which CPU is in their computer?
 

Vope45

Member
Oct 4, 2020
114
168
86
nope


You think a sub $100 mobo can run those i7s? I don't think so.
 

AtenRa

Lifer
Feb 2, 2009
14,003
3,360
136
Isn't the low end market 12100/12400 primarily OEM builds?

I agree that Intel have a better product at this price range, but do the customers buying computers in this price range really care that much about which CPU is in their computer?

Vast majority of sales is at $300-350 and bellow. i3/i5, ryzen 3/5
12400 and 5600x are the two most dominant CPUs in DIY




Also from US

 
Reactions: biostud

Gideon

Golden Member
Nov 27, 2007
1,705
3,913
136
I got a chuckle out of the worries about power consumption from two rather pro-Intel forum members here.

Ever since 10900K launched with it's 250W power draw, we have heard nothing but:

1. "Power consumption doesn't matter for usual desktop use".
2. "Nobody runs cinebench all day or taxes all the cores constantly"
3. "In gaming loads it will be nowhere near its PL2"

They have some merit and the last one is an actual fact.

Thw moving goalposts are just funny, as the moment AMD starts playing "the Intel game" (but still drawing less) with 230W PPT it's suddenly a serious issue?

None of these members have ever mentioned that Intel's powerdraw is "worrying" or a "bad sign".
 

DrMrLordX

Lifer
Apr 27, 2000
21,791
11,131
136
The lack of v-cache related info is sad. I would prefer to get that one, but i am not keen to wait another half year for it.

In the same boat. It's gonna feel bad man to wait ~6 more months, but it'll also feel bad when Raphael-X hits the scene and I'm running a 7950X. I won't be buying both.

So Nobody Will Acknowledge that I was Right About AVX512? So far Geekbench5 seems to like it because the Crypto performance is Off The Chart.

I will! For now! Wait until GB is patched to reduce the importance of crypto performance. GB6 incoming?

Quicksync and GPU encode is fine and good (enough) but nothing beats CPU.

Agreed! CPU is king in streaming for sure. It kind of makes me cringe watching streamers rave about how good the 12900k is at streaming in 720p (yay e-cores!). A 7950X will be happily running circles around it streaming in 1080p or one of the "hybrid" resolutions favored by Twitch.

During non-public Q+A with journalists, AMD said there will be no supply constraints on Zen 4.

Let's hope so. Though with all the customers abandoning wafers at TSMC, AMD may be flush with available silicon.

The prices show there is now very very little compettition.

If we want competition, Intel is going to have to wake up and do something. Raptor Lake may give budget buyers some interesting options, but at the mid-to-high end consumer desktop, it's AMD or bust.

Geekbench isn't that great as it doesn't stress the cpu in all aspects needed.

Depends on the CPU. If you're running a little 4c mobile CPU it can keep something like that pegged. A 12c or 16c desktop behemoth? Nah, not really.
 

Zucker2k

Golden Member
Feb 15, 2006
1,810
1,159
136
I got a chuckle out of the worries about power consumption from two rather pro-Intel forum members here.

Ever since 10900K launched with it's 250W power draw, we have heard nothing but:

1. "Power consumption doesn't matter for usual desktop use".
2. "Nobody runs cinebench all day or taxes all the cores constantly"
3. "In gaming loads it will be nowhere near its PL2"

They have some merit and the last one is an actual fact.

Thw moving goalposts are just funny, as the moment AMD starts playing "the Intel game" (but still drawing less) with 230W PPT it's suddenly a serious issue?

None of these members have ever mentioned that Intel's powerdraw is "worrying" or a "bad sign".
These are true, but until yesterday some of the loudest voices on here were in denial that the 7950x (and we now know the 7900x as well) are going to have a 230 watts PPT limit, the max socket power right from the gate. That accounted for a lot of the debate on here. These same voices, while denying this significant jump in power consumption on the part of Zen 4, pushed the unsubstantiated 350w power consumption on RPL.

I have to say that with AMD's 5nm full node advantage bringing up the bottom power ceiling this much, Intel may be left with no option but moving up. Let's just hope it doesn't get uglier than it already is.
 

Yosar

Member
Mar 28, 2019
28
136
76
2. 12600K currently at $239 that gives higher ST and MT vs the 5600X at $200

Maybe on some promotion on Newegg. In my country the difference is about $100. You can buy new mobo for this $100 and put 5600X in it.
In games (and this is processor for games market, not MT performance) it gives 12600K 3% advantage over 5600X.
People happily trade this 3% for mobo.

3. At the $300 range i7 12700F is currently at $320 that decimates the $260 5700X in both gaming and MT workloads.

Simply no. Both are other categories. And in gaming (1920x1080) 12700F is 7% faster if you game on 6950XT. I would not call this decimate.
Oh, and 5700X is about $80 cheaper in my country than 12700F. Price of 12700F is actually much much closer to R9 5900X than R7 5700X. And this comparison doesn't look good for 12700F at all in ST and MT.
And this is before Zen 3 price cuts which will be inevitable due to Zen 4 launch.

Plus socket 1700 platform has higher upgradability (Raptor Lake), AM4 is dead socket.

There is no upgradability for socket 1700. Unless you upgrade every generation which people mostly don't. You can say about upgradability if at least 3-4 processor generations are on the same socket. 1700 is as dead as AM4 basically at this moment.
Actually no. You can have a really great upgade in future in 5800X3D if you are gamer.
And for bonus you have no problems with compatibility of games with intel hybrid architecture (some games still don't launch even).
 

coercitiv

Diamond Member
Jan 24, 2014
6,376
12,754
136
Probably a bit late to do so, and also difficult in a not-breaking-platform-compatibility-generation.
Rest assured that Intel's socket has lots of breathing room, especially considering they let mobo makers choose default specs on a model by model basis. I'd have to check the stock config on my MSI board on it's latest UEFI revision, but the initial firmware had no power limit in place for the 12700K. It simply ran as high as thermals allowed.

The same applies for AM5 too, although probably it will require pro-active user decision like enbling PBO (or equivalent). IIRC the x670 board from MSI that is equivalent with my Z690 MSI board is rocking 14 x 80A stages, an upgrade from 14 x 55A on my board. The newer Intel boards will definitely follow this trend. It's a VRM "arms race".

Oh, almost forgot, RIP PSUs!
 

coercitiv

Diamond Member
Jan 24, 2014
6,376
12,754
136
In case anyone missed this:

Dr. Lisa Su: 'With our launch of Zen 4 we don't expect any supply constraints.'
From an AMD standpoint, we have dramatically increased our overall capacity, in terms of wafers, as well as substrates and on the back end. So with our launch of Zen 4 we don't expect any supply constraints.

Logistically it takes a little bit longer for things to get into region. So we're having this event at the end of August and we're on sale on September 27. One of the reasons for that time, frankly, is to make sure that we do have product in region so that people can really look at buying across the board.
 

moinmoin

Diamond Member
Jun 1, 2017
4,993
7,763
136
A base level price for a cpu at 300usd is just plain high. Full stop. And it can be seen at the valuation of amd. This industry is haunted by the tendency towards natural monopoly and the resulting high prices.
It's not the base level price, it's just the high entry level price to an overall costly new platform still to be launched with costly new memory and (if consequently taking PCIe 5 NVMe) costly new storage. Over time there will be cheaper products.

But as Zen 3 already has shown if faced with the question of supplying low margin mass market or high margin specialist markets AMD will pick the latter, happily leaving the former to Intel which is still more competent at creating true mass market products in high quantity. Will be interesting to see how that plays out this round with AMD already publicly stating that supply won't be an issue this time.

(I don't see how any of that relates to monopolies though. It was and is a duopoly, and a rather healthy one at that.)
 
Reactions: inf64

Abwx

Lifer
Apr 2, 2011
11,161
3,858
136
For efficiency, they compared 7950x vs 5950x in CB R23 nT.

Footnotes for both slides:
View attachment 66795
This slide is straightforward. I am happy with much higher performance at low TDP, that means mobile will see significant improvement in performance.


This slide doesn't tell us much info to compare.
In the case of Up to ~49% higher performance(same power), we don't know what TDP they are actually talking about. It must be somewhere between 65-105W TDP based on the increase in performance and looking at the first slide.
In the case of Up to ~62% lower power(same performance), we don't know what is the R23 score they are talking about.
I personally think they are talking about R23 score of 5950x at 170W TDP(230W PPT).
This would also mean 7950x needs only 65W TDP(88W PPT) for the same performance. Why do I think so? Because Zen4 is most efficient at 65W TDP(88W PPT) based on the first slide.

BTW I checked 5950x review(Link) at ComputerBase. They didn't test CB R23, but at least they tested CB R20, there shouldn't be much of a difference.
If I apply the first slide to CB R20 scores, then It looks like this for 7950x.
CB R20TDP: 65W
(PPT: 88W)
TDP: 105W
(PPT: 142W)
Performance gain from
higher TDP
5950x79821043531%
7950x13889 (+74%)14296 (+37%)3%
This doesn't look correct, only 3% difference in performance with 62% higher TDP(65->105W)? Nonsense.


That s accurate, one slide is a 5950X/7950X comparison while the other, with 62% lower power or 59% higher perf, is the average for the whole stack from 5600X to 5950X in respect of 7600X to 7950X.
 
Reactions: lightmanek

Joe NYC

Platinum Member
Jun 26, 2021
2,323
2,929
106
Sorry to disappoint you but raptor lake does not matter at all. For alder lake users, raptor is a very very small side grade. For all other users, amd offer a much much better core and platform. Hope this helps.

I agree.

RPL is not enough of an upgrade for ADL users.

RPL platform is inferior and dead end.

RPL CPU that will need extreme OC, extreme memory, extreme TDP just to barely keep up.
 

biostud

Lifer
Feb 27, 2003
18,387
4,939
136

IGP has 2 CUs. Should still be plenty for 2D but you won't be gaming on it, even old games. I wonder if it has the encoders.

Also the IO die confirmed to be 122 mm2. AMD's definitely taking a hit on margins pricing it this low.
[/QUOTE]

AMD has said that this is nice to have to be able to turn your computer on and use it, even if you don't have a GPU or if you're running some troubleshooting with your GPU. Also I suspect it will be a power saving feature when they start using desktop zen4 CPUs in laptops for high performance laptops with a dedicated GPU.
 
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/    |