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

Page 102 - 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
809
1,412
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).

Untitled2.png


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:
  • Like
Reactions: richardllewis_01

andermans

Member
Sep 11, 2020
151
153
76
We have something that is enough for estimations.
Thoses curves are the processes typical perf/power curves, no matter what is the uarch.

27.png



So if we use the 11800H from that chart, and go for 16 Willow Cove cores -> 24 Willow Cove cores at 165W TDP (looks like the top TDP spec for Alder lake?), that would mean going from 2 11800Hs at 82.5W to 3 11800Hs at 55W. Eyeing the graph that would be 12500 *2 = 25000 -> 10000 * 3 = 30000, or a 20% increase.

Of course illustration only, wrong process, wrong cores, ignoring uncore scaling etc.
 
  • Like
Reactions: Tlh97 and lobz

Abwx

Lifer
Apr 2, 2011
11,557
4,349
136
@Abwx

10ESF isn't represented on that chart.

If 10ESF has 15% better perf/watt then all you have to do is a translation of the 11800H curve by 7%, that is 7% better perf at iso frequency, likewise it will consume 13% less at same throughput (isoperf).

Also if IPC is x % better then power will increase by x % at isofrequency.

Edit : looking at the 11800H curve it seems that the improvement Intel are talking about is not within this displayed range but for frequencies that are well above the one that can be estimated here.

From 47.5W to 95W power increase by a factor two while perf increase by a 1.4 factor, that s quite remarkable and say that the process cant be improved the slightest within this frequency range, as it s a about a perfect square law wich is the best mosfets transistors can do by their very principle.
 
Last edited:

DrMrLordX

Lifer
Apr 27, 2000
22,065
11,695
136
I remember the time when people complained whenever some talked about AMD in threads related to Intel. Now I just had to look twice to be sure I'm in the Zen 4 thread. (Not intended to be a complaint, just an observation, carry on.)

At least that chart had Zen3 on it!

If 10ESF has 15% better perf/watt then all you have to do is a translation of the 11800H curve by 7%, that is 7% better perf at iso frequency, likewise it will consume 13% less at same throughput (isoperf).

You don't know that that calculation applies across the entire v/f curve.
 

Abwx

Lifer
Apr 2, 2011
11,557
4,349
136
You don't know that that calculation applies across the entire v/f curve.

Here the curve goes up to 4GHz, and as said mosfets display this square law only in the first part of a v/f curve (conduction, and hence frequency, increase as a square of voltage).

You can see that AMD Zen 3 curve is about the same up to a point where there s an inflexion, at wich point the power increase as a cube of frequency, this behaviour is due to the parasistic resistance of the mosfets that become large in respect its intrinsic conduction.

Methink that Intel s difficulty with 10nm wasnt that their process was that bad but that they absolutely wanted something that would give them a 10% frequency advantage reliably.
 

eek2121

Diamond Member
Aug 2, 2005
3,100
4,398
136
I remember the time when people complained whenever some talked about AMD in threads related to Intel. Now I just had to look twice to be sure I'm in the Zen 4 thread. (Not intended to be a complaint, just an observation, carry on.)
Happens in the Intel thread all the time. :D

At least that chart had Zen3 on it!



You don't know that that calculation applies across the entire v/f curve.

Well that and the goal posts have been moved thanks to a new architecture. You can't simply compare a different chip from Skylake to a different chip from tiger lake. That is just as invalid as a node change (oh and, perf/watt is NOT the only thing that improved with 10ESF). Zen 4 will have very different power/thermal parameters from Zen 3. Zen 3 also has very different thermal parameters from Zen 2, despite being on the same process. Zen 3 runs quite a bit hotter, significantly faster, and while I haven't checked the exact TDP calculations, it would not surprise me if they have changed. Despite that, TDP values and power consumption both remained the same.

None of us have any idea WHAT to expect from either Zen 4 or Raptor Lake. Many have low expectations from Intel, because they've been under delivering as of late. Many put their head in the clouds because AMD has been OVER delivering as of late. I suspect the reality will be much more neutral, or definitely different than most people think, since thoughts/feelings tend to skew such conversations for many.

I am going to make a bold guess as to where both Intel and AMD chips will land upon Zen 4/RPL-S chips: within 10% of each other for at least 85% of the desktop SKUs. Bookmark this comment. You heard it here first folks! 🤣
 

eek2121

Diamond Member
Aug 2, 2005
3,100
4,398
136
Igor’s Lab has an article on thermal and power management of Raphael. 2 things of interest that I did not see before:

  1. Internal documents call the 16-core part an APU.
  2. Internal documents show 120W and 170W TDP SKUs.
 

Ajay

Lifer
Jan 8, 2001
16,094
8,109
136
Igor’s Lab has an article on thermal and power management of Raphael. 2 things of interest that I did not see before:

  1. Internal documents call the 16-core part an APU.
  2. Internal documents show 120W and 170W TDP SKUs.
Igors Lab said:
There are no more secrets here and the changes compared to Zen 3 are more of a minor evolution, but the actual basic principle remains the same. So you can also infer Zen 3 from a lot of things. By the way, it’s interesting that AMD now also refers to the 16-core internally as an APU, which hides a graphics unit in the I/O die. But now we come to the power management on the next page, because the waste heat and the temperatures as a consequence have to come from somewhere.

So, yes, we've already seen these reported. Apparently, enthusiast grade CPUs are headed for high-end Consumer desktops; seems a bit odd, but okay.
 

A///

Diamond Member
Feb 24, 2017
4,351
3,158
136
Wasn't it last week we were told the so-called 170-175 watt TDP part was a fabrication? Interesting to see it back on the menu but I'm also curious what's causing that bump in power. Is it the iGPU? The bump to 120 watts from a baseline of 105 watts for the 5800X and 5900X and likely also the 5600X would point towards the entire range getting iGPUs and people who tossed the idea away even as recently should take note of this revelation. Though the 5950X and 5900X share the same TDP, so I'm a little confused about the jump for the 16c part and not the 12c, unless both are getting the 170 watt treatment?
 

Hougy

Member
Jan 13, 2021
80
62
61
Gracemont performs better than Zen 2 and worse than Zen 3. Throw 16 of them in a chip along with the Raptor Cove, and assuming Intel isn't power limited, the chip will be 40-50% faster than a 5950X. That means a 6950X will need to be 40%-50% faster to tie with the Intel chip. That is why AMD needs a 24 core SKU. Note that Raptor Lake is currently rumored to launch BEFORE Zen 4. If Raptor Cove is power limited, then AMD doesn't need to make up such a wide gap.



125W PL1, 241W PL2. That is from the Gigabyte leak. Raptor lake apparently has a lower PL2 (I would have to dig up the Igor's lab article).
The flagship Raptor lake could perform better than the flagship Zen 4 chip in multithreaded benchmarks if it has more cores but these benchmarks are useless to most consumers. What matters to games and the apps most people use is single threaded performance.

You are way overestimating the performance of Gracemont because you're only considering IPC and not the fact that it clocks lower and doesn't have SMT.

Zen 4 could have better single threaded performance, much lower power consumption and be cheaper, making it a huge win.
 

eek2121

Diamond Member
Aug 2, 2005
3,100
4,398
136
So, yes, we've already seen these reported. Apparently, enthusiast grade CPUs are headed for high-end Consumer desktops; seems a bit odd, but okay.
Yes, many of us have known it for months, but some were in denial.
Wasn't it last week we were told the so-called 170-175 watt TDP part was a fabrication? Interesting to see it back on the menu but I'm also curious what's causing that bump in power. Is it the iGPU? The bump to 120 watts from a baseline of 105 watts for the 5800X and 5900X and likely also the 5600X would point towards the entire range getting iGPUs and people who tossed the idea away even as recently should take note of this revelation. Though the 5950X and 5900X share the same TDP, so I'm a little confused about the jump for the 16c part and not the 12c, unless both are getting the 170 watt treatment?
There appear to still be 105W parts. My initial thought was the 170W part is a 24c part and the 120w part is the 16c part, with the rest being 105W parts, but the APU could also be the differentiator.
 
  • Like
Reactions: BorisTheBlade82

Mopetar

Diamond Member
Jan 31, 2011
8,114
6,770
136
So, yes, we've already seen these reported. Apparently, enthusiast grade CPUs are headed for high-end Consumer desktops; seems a bit odd, but okay.

I don't think it's odd at all. First not everyone who can benefit from a 16 core CPU needs a HEDT platform. Second, not everyone needs a particularly powerful GPU either and with the market the way it is right now being able to get an APU with 16 cores could save a lot of money.
 
  • Like
Reactions: Tlh97

jpiniero

Lifer
Oct 1, 2010
15,223
5,768
136
Wasn't it last week we were told the so-called 170-175 watt TDP part was a fabrication? Interesting to see it back on the menu but I'm also curious what's causing that bump in power. Is it the iGPU? The bump to 120 watts from a baseline of 105 watts for the 5800X and 5900X and likely also the 5600X would point towards the entire range getting iGPUs and people who tossed the idea away even as recently should take note of this revelation. Though the 5950X and 5900X share the same TDP, so I'm a little confused about the jump for the 16c part and not the 12c, unless both are getting the 170 watt treatment?

Most likely a post-Raphael product that has more than 16 cores.
 

Ajay

Lifer
Jan 8, 2001
16,094
8,109
136
I don't think it's odd at all. First not everyone who can benefit from a 16 core CPU needs a HEDT platform. Second, not everyone needs a particularly powerful GPU either and with the market the way it is right now being able to get an APU with 16 cores could save a lot of money.
Good point. Like scientific computing but with fairly simple output (graphs, less complex 3D models, etc.). Actually, video editing, if the APU has some acceleration for popular codecs.
 

eek2121

Diamond Member
Aug 2, 2005
3,100
4,398
136
Good point. Like scientific computing but with fairly simple output (graphs, less complex 3D models, etc.). Actually, video editing, if the APU has some acceleration for popular codecs.

There is also the fact that the higher core count parts are typically binned better.

I can’t wait for Zen 4 TBH so I can spend more money. AMD is robbing me of every penny I have at this point. I went from a 1950X to a 3900X to a 5950X. That doesn’t even count the other machines in my household. Hopefully Zen3D isn’t an outrageously good product as well…
 
  • Like
Reactions: Tlh97 and Joe NYC

Mopetar

Diamond Member
Jan 31, 2011
8,114
6,770
136
Well, one of the two chiplets (at least with Zen2).

Both chiplets hit the minimum bin requirements, but they aren't paired in any way beyond that.

Depending on the actual distribution in terms of silicon quality and where they set the bins it's pretty unlikely to get two evenly matched chiplets.

The vast majority of CPUs sold will have a disparity between those chiplets of at least a certain threshold.
 

Abwx

Lifer
Apr 2, 2011
11,557
4,349
136
142W isn't actually the max sustainable power for AM4. It can handle over 300W. Nothing AM4 pulls that kind of power by default, but it's possible.

OEMs can even do a 500W sustainable, the 142W sustained capability is the requirement from AMD for AM4 validation of the boards, they can increase this number for next boards the same way they did increase the power requirements from AM3 to AM3+.
 

eek2121

Diamond Member
Aug 2, 2005
3,100
4,398
136
142W isn't actually the max sustainable power for AM4. It can handle over 300W. Nothing AM4 pulls that kind of power by default, but it's possible.

Yeah I thought the quote was kind of odd. On my 3900X I had PBO turned on and set to 175W. I never had heat issues (except my office when CPU and GPU was active lol)