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

Page 7 - 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
807
1,411
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

TESKATLIPOKA

Platinum Member
May 1, 2020
2,508
3,009
136
That s not at 45W, the number is for the first run, in the power charts we can see that the laptop consume 105-110W in Cinebench with an external monitor...
You are right, the numbers for ADL are much higher than for Zen 3.
R9 PRO 6950HS uses only Ø64.7 (49.2-68) in CB R15 with external monitor.
i9-12900HK uses Ø106.1 (102-113) in CB R15 with external monitor.
i9-12900HX uses Ø177.1 (166.7-232) in CB R15 with external monitor.

edit: I checked some reviews at ultrabookreview.com and 6800U vs 6900HX vs 12900H looks like this in CB R15 at 15th run.
R9-6900HX (35W)i9-12900H (35W)R7-6800U (25W)
CB R15192516611598
 
Last edited:

Asterox

Golden Member
May 15, 2012
1,037
1,821
136
Genoa Wrecking all kinds of havoc at Geekbench5 Top Multi Score

Top Multi-Core Geekbench 5 CPU Results


Over 96,000 points in Geekbench is just crazy even knowing that Geekbench does not scale linearly.

Spectacle, but in real world what can you do with Geekbench. :grinning:

A beauty for the eyes, and the software used.

- Cinema4d/Octane
- After Effects
- Premiere Pro

 

Kaluan

Senior member
Jan 4, 2022
504
1,074
106
That is because it uses actual workloads. In my experience, it has been pretty accurate at measuring IPC and/or performance and IIRC Dr. Ian Cutress himself stated it was comparable to SPEC.

Real world workloads don’t scale perfectly or linearly.

Geekbench is a solid benchmark that receives way too much hate. Most of that hate comes from the wide variability of results that are published from a wide range of hardware/software configurations ranging from complete garbage to sunshine perfect.

A random result found on their website can be taken with a grain of salt. A result in a carefully controlled environment on the other hand is invaluable for measuring performance.
Yeah I don't know about that, whenever I see a benchmark have results up for something like a 5950X varying from 9K to 17,5K points I can't think of anything but how garbage it is.

Particularly when almost all other benchmarks out there give much more consistent and clear results... and can also be ran in sustained mode, a feature which GB is sorely lacking (given it's mobile SoC focus you'd think throttling/anti-artificial boosting measurements would also be part of it's test suite but no... we're supposed to belive a 30s benchmarks is the be-all end-all of precise performance gauging lol)

And even in ideal scenarios, it's misleading at best. Take 5800X v 12600K results, 5800X typically scores 10xxx points, 12600K scores 11xxxx for DDR4 and 12xxx for DDR5. But taking a quick look at cumulative application performance on review outlets like TPU or ComputerBase, 5800X sits in front of 12600K by a few percent, not 10-20% behind, as GB would make newbies belive.

If anything, I feel like it doesn't get ENOUGH criticism. Still, I hope v6 might pleasantly surprise me one day.
 
  • Like
Reactions: inf64

v.strix

Junior Member
Aug 25, 2022
8
18
41
I am more interested on MT Performance. The AVX512 performance will tip the balance to the 7950X part(as in doubling the AES-XTS Performance of the 13900K)

Because I fully expect the High Clock 32T Zen4 CPUs to beat/match Slow Clock 32T Ice Lake Xeons on AVX512 and the 13900K with it's 32T even at High Speeds can't hold a candle to those slow xeons at AVX512 performance

View attachment 66671
Yeah, it ain't gonna double deleted there... ;)

Just look closer at the 5950X MT AES-XTX results. With 16 cores and 32 threads it barely reaches 2x of ST score! At that point slow clocks and AVXwhatnot doesn't even matter.

The slow and ugly Ice Lake Xeon, on the other hand, has tons of bandwidth on its 8 channels of memory. And that's exactly what this degenerate microbenchmark seems to be measuring here. :eek:

I guess, that shouldn't be one bit surprising when it's designed by self-proclaimed chimps. ;)

Please - no profanity in the tech subforums.
admin allisolm
 
Last edited by a moderator:
  • Like
Reactions: Drazick

gdansk

Platinum Member
Feb 8, 2011
2,890
4,363
136
View attachment 66786

Translation to English: RIP Sapphire Rapids

2 CCDs * 32.5W = 65W
Milan can fit 8 of them 8x32.5W) inside its TDP
Genoa can fit 12 of them (12x32.5W) inside its TDP

So, Genoa performance could be 12 CCD / 8 CCD * 74% efficiency increase
which is 1.5 * 1.74 = 2.61

Then Genoa performance will be +161% over Milan.
Do remember that they probably mean 65W TDP = 88W PPT. While for Milan/Genoa TDP = PPT.

Earlier rumors were 1100mhz for 2CU iGPU frequency.But as we see, 2200mhz is iGPU(RDNA2) frequency or 200mhz lower speed vs the fastest Rembrandt APU.



1.1 tflops. A bit behind in tflops HD 770 in Alder Lake but RDNA2 is probably better despite that.
But for media transcode I bet Intel will have better support for many years to come.
 

nicalandia

Diamond Member
Jan 10, 2019
3,331
5,282
136
Should be noted Zen4 doesn't actually have 512-bit width registers, it is actually just two 256-bit width registers.
View attachment 65358
It should be done like this.

Zen2-esque mode for AVX512 (FP0/1 + FP2/3 + FP4/5 :: 3-pipes for AVX512)
or
Zen3-esque mode for AVX256 (FP0 + FP1 + FP2 + FP3 + FP4 + FP5 :: 6-pipes for AVX256)

Zen4c cuts off the second FPU unit and runs AVX512 like AVX256 in Zen1.

Decoder = Full Decode of AVX512
NSQ = Splits AVX512 into AVX512_L0 to Scheduler0 and AVX512_HI to Scheduler1.
Zen4 FPU's Store0 and Store1 can operate simultaneous to store full 512-bit width.

There is no power penalty from AVX512. Since, it is just using the existing AVX256 units that were improved from Zen3. There is no penalty for mixing AVX256 and AVX512.

There is no 2x512-bit register.
There is no 12 FPU units.

Family 19h only has
3 FP Pipes of 256-bit width per PRF+Cluster.
2x128-bit Registers (6 FP pipes only)
2xStores (Last FP pipe in each FPU cluster)

This guy had it right all the time...

AMD Zen3 AVX256 and 500 KiB L2 74mm2 CCD
AMD Zen4 AVX512 and 1 MiB L2 71mm2 CCD

1661875232024.png
 

DrMrLordX

Lifer
Apr 27, 2000
22,027
11,606
136
So as it stands the best Intel is going to bring till 2024 is the 13900KS right? So 7950X3D will reign supreme till late 2024.

We don't know what the 13900k will be like or if Intel will bother with a 13900KS. But from the looks of things Raptor Lake may not meaningfully exceed Raphael as it is. It's a good thing that Raphael-X is coming anyway. AMD may not really need it.

Cheese had access to Raptor Lake and found no IPC increases in their own workloads https://chipsandcheese.com/2022/08/23/a-preview-of-raptor-lakes-improved-l2-caches/

That is a bit worrying. Especially since that may be the last meaningful desktop update from Intel until 2024.

I don't get the focus on the EXPO timings. This is nothing but XMP with AMD's branding. Were AMD CPU's previously handicapped when using (intel specific) XMP timings?

Basically, yes. Intel's memory controllers do not like the same timings as AMD's and vice versa. XMP 2.0 can do some pretty horrible things to timings and "hidden" voltages on AM4 systems. It's about time that AMD did something to attempt to address the problem.
 

FangBLade

Senior member
Apr 13, 2022
201
395
106
Some people still look at 13% IPC and says it is small, they forgot that Zen 4 in the end brings biggest performance jump, yes they choose clock path instead of IPC, but still performance jump is biggest in the Zen era, and power consumption will be much better than Intel so it is a clear win, can't wait to see 3d version, they have even better power efficiency, and with second generation 3dcache + better original cache hierarchy that will better cooperate with 3dcache + ddr5 + higher clocks = we should see significant perf jump in games, RPL won't catch up it even with best ddr5 and KS model.
 

AtenRa

Lifer
Feb 2, 2009
14,003
3,361
136
- "7600X will have to compete against i7 137xx" WAT... seriously... WHAT?

Current prices from newegg

Core i7 12700F = $315
GIGABYTE B660M AORUS PRO AX = $128
32GB DDR4 3600 CL16 = $96

Total for Alder Lake= $539

Core i7 13700K = $400
GIGABYTE B660M AORUS PRO AX = $128
32GB DDR4 3600 CL16 = $96

Total for Raptor Lake = $624

----------------------------------------
Ryzen 7600X = $300
B650 = $125 (lets assume they will be available at launch)
32GB DDR5 6000 CL40 = $220

Total for 6 core ZEN4 = $645

Im sorry im not giving this money for a six core in the end of 2022 start of 2023.
 

Saylick

Diamond Member
Sep 10, 2012
3,512
7,766
136
Might as well give the real source credit... That way, no misrepresentation of info like WCCFTech normally does due to typos and general lack of understanding.

I think the interesting thing about the two Bergamo SKUs is the core count vs TDP.

112 cores = 340W while 128 cores = 360W. Clocks are roughly the same.

That's 20W for an extra 16 cores, or roughly 1.25W per core. With 128 cores, that means the cores themselves use up: 128 cores x 1.25 W/core = 160W, leaving 200W for the IO?

Seems rather non-intuitive what's going on here. It surely can't be 200W for the IO, right?


1662009819127.png
1662009835557.png
 

Attachments

  • 1662009785832.png
    1662009785832.png
    5 MB · Views: 8
  • 1662009795384.png
    1662009795384.png
    5 MB · Views: 8
Jul 27, 2020
19,823
13,588
146
Zen 4 is reportedly a hot running chip.
AMD/TSMC better start working on integrated cooling channels or something. Otherwise, the future for high speed CPUs doesn't look good.

One possibility is to have an additional IHS (with a cutout for connection with the mobo socket) at the bottom of the die so heat has more surface area to dissipate from. But that will surely require a redesigned socket.
 

Saylick

Diamond Member
Sep 10, 2012
3,512
7,766
136
I have not seen that report(Zen4c Having similar IPC than Zen3)
Here, I'll snipped a quote from Charlie at Semiaccurate (behind paywall) to elaborate. You're right, I misread his reporting.
The layout for Bergamo is simple, take the sIOD from Genoa and put on eight Bergamo CCDs and you have a very different device from it’s bigger brother. Each CCD has 16 cores rather than the 8 cores of a Genoa CCD but the L3 cache is the same 32MB on both. Interestingly, Bergamo CCDs take a page from the Naples CCD and split each physical die into two logical CCXs each with 8C and 16MB of L3. It wasn’t explicitly stated in the AMD OEM briefings but it is highly likely that the CCXs will not be able to talk to each other directly but instead communicate by going back to the sIOD. This strongly implies a split xGMI3 link like Naples and since it worked decently there, it should perform well enough on Bergamo.

The cores are the same ISA so everything that Genoa has, Bergamo has, and the sIOD is the same so the platform is the same SP5 base as well.

...

Now we come to the interesting data, performance. AMD is making some amazingly bold claims about Bergamo, they outright state that it, “Targets up to 2X improvement over Milan 7763 across ALL key foundational workloads”, their caps. For reference that 7763 is the top Milan SKU with 64C at 2.45/3.5GHz and pulls 280W TDP.

...

Slightly more nuanced is that if the Zen4 core has an IPC gain over the Zen3 core, and it does, that will allow AMD to drop clocks vs Milan at the same performance level and save even more power in the real world. This can be clawed back by higher performance or seen as lowered TCO for customers.

...

That jaw stuff comes when you turn off threading or what AMD calls SMT Off mode. In this mode, Bergamo has 2x the cores as Milan but the same thread count, not exactly impressive. That said AMD claims it will beat a 64C/128T 7763 by *60%*! Let me repeat that, a Bergamo running in SMT off mode will have the same thread count as the current top Milan but be overall 60% faster.
 

Det0x

Golden Member
Sep 11, 2014
1,262
3,991
136
Want to see some gaming benchmarks, but here are some numbers in CPU-Z and GB5 from the leak above:

AMD Ryzen 7 7700X perforamnce in CPU-Z and Geekbench has been leaked

Starting off with CPU-z, the AMD Ryzen 7 7700X was running on the Gigabyte X670E AORUS Master motherboard and this is the retail variant. The CPU was running at 5.42 GHz all-core boost clocks & had an operating voltage of 1.152V with temperatures hovering around 71C. We don't know if this was under load or what type of cooling was used. The maximum multiplier reported for the chip is 55.5x which leads to a 5.5 GHz peak frequency limit.
1662610420622.png

1662610431415.png

With such a score, the CPU is 21% faster than Ryzen 7 5800X (8-core Zen3) and 20% faster than Ryzen 9 5950X (16-core Zen3) in single-core test. It is also 1% faster than Core i5-12600K (10-core Alder Lake), but it is outperformed by Core i7-12700K (12-core).

In terms of multi-threaded performance, the 7700X scores 1% higher than Core i9-12900K (16-core) and 8% higher than i7-12700K. This new Raphael CPU is also 28% faster than its predecessor (5800X).

There is also this new entry to Geekbench database with the same CPU but different motherboard (ASUS ROG Crosshair X670E Hero) and memory configuration (64GB of DDR5-6000):

AMD-Ryzen-7-7700X-Geekbench.png

Here, the CPU scores 2209 and 14459 points in single and multi-core Geekbench V5 tests respectively. In this case, 7700X ends up faster than Core i7-12700K by 16% in single-core and 2% in multi-core tests.

AMD Ryzen 7 7700X is officially set to launch on September 27th alongside three other Ryzen 7000 CPUs. According to the reports, AMD will lift the review embargo Ryzen 7000 as well as X670 motherboards a day sooner (September 26th).
 
Last edited: