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

Page 472 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Toggle sidebar Toggle sidebar
Vattila

Vattila

Senior member
Oct 22, 2004
799
1,351
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
Sort by date Sort by votes
T

Tuna-Fish

Golden Member
Mar 4, 2011
1,355
1,549
136
gdansk said:
On paper Zen 4 *does* looks weaker/simpler than Golden Cove but its unquantifiable characteristics must be excellent as it ends up with similar performance at similar clock rates. Not in every workload but when averaged out they're pretty close.
Click to expand...

Cache latencies are notably lower. That's probably all it is.

zen4_latency_cycles.png


Latency in cycles is lower for all levels of cache. That 1 cycle of L1 latency probably has a deceptively high impact. L1 has >95% hit rate on most well behaved loads, one extra cycle on every access adds up.

Geddagod said:
Zen 4s reorder buffer performs surprisingly well compared to GLC despite the size difference
Click to expand...
When your average latencies are lower, you don't need as big of a ROB to keep the units occupied.

itsmydamnation said:
separate integer and FP resources. approximately 0 workloads stress both at the same time , you know how like SPEC has separate FP and INT benchmarks.
Click to expand...
That's not really right. Even heavy vector loads typically have a bunch of scalar integer "bookkeeping", like loop counters, list indexes, etc stuff they need to update.

Granted, with the 2 dedicated scalar ports, GLC now probably has enough for most cases.
 
Last edited:
  • Like
Reactions: Tlh97, Henry swagger, lightmanek and 7 others
Upvote 0 Downvote
E

eek2121

Platinum Member
Aug 2, 2005
2,930
4,026
136
They will probably reveal pricing once the KS part drops.

Kaluan said:
Target audience you mean? Yeah I suppose.
7000 non-X, 7600X/7700X budget builds, upcoming APUs etc

But I doubt you can't run a 7950X on them if you really wanted to.
Click to expand...

Indeed, these boards will very likely be able to run all chips. Expect IO to be gimped, however. Likely no PCIE 5. Limited PCIE 4, if any.
 
  • Like
Reactions: Kaluan
Upvote 0 Downvote
I

itsmydamnation

Platinum Member
Feb 6, 2011
2,776
3,161
136
Tuna-Fish said:
That's not really right. Even heavy vector loads typically have a bunch of scalar integer "bookkeeping", like loop counters, list indexes, etc stuff they need to update.

Granted, with the 2 dedicated scalar ports, GLC now probably has enough for most cases.
Click to expand...
Thats why I used the word stressed . its not that there isn't a bunch of flow control , loop counter etc going on. If this was really such a big advantage AMD would have much higher dispatch then they do :) . Does anyone know if Zen3/4 has higher retirement /cycle then dispatch like Zen2 did ?
 
Upvote 0 Downvote
A

Abwx

Lifer
Apr 2, 2011
10,960
3,474
136
All links must be accompanied by your own commentary.
itsmydamnation said:
Thats why I used the word stressed . its not that there isn't a bunch of flow control , loop counter etc going on. If this was really such a big advantage AMD would have much higher dispatch then they do :) . Does anyone know if Zen3/4 has higher retirement /cycle then dispatch like Zen2 did ?
Click to expand...

Page 471 - Discussion - Speculation: Zen 4 (EPYC 4 "Genoa", Ryzen 7000, etc.)

Page 471 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
forums.anandtech.com forums.anandtech.com
 
Upvote 1 Downvote
A

Abwx

Lifer
Apr 2, 2011
10,960
3,474
136
Last edited:
  • Like
Reactions: lightmanek
Upvote 0 Downvote
J

JustViewing

Member
Aug 17, 2022
135
232
76
itsmydamnation said:
1. separate integer and FP resources. approximately 0 workloads stress both at the same time , you know how like SPEC has separate FP and INT benchmarks.
Click to expand...

Not necessarily, not all FP workload are steaming AVX. Most of the time they will be mix of both. Having said that, compilers may optimize for Intel's combined Int/FP architecture by keeping FP and code sections separately to help with scheduler. Unlike in the past, modern schedulers are very flexible with its deep out of order buffers.
 
  • Like
Reactions: Geddagod
Upvote 0 Downvote
M

Mopetar

Diamond Member
Jan 31, 2011
7,848
6,005
136
Now that we're starting to see cheaper Zen 4 CPUs, I think we're going to see less expensive motherboards. It's considerably easier to talk the person who just bought a 7950X into a premium board, but the guy who's in the market for a 7600 or a 7700 is t looking to spend nearly as much.

It's the same every time a new platform launches. Companies lead with high-end products to get higher margins out of enthusiasts that are willing to spend more than the average consumer. Over time the mainstream and low-end options trickle out, but there's not a lot of financial incentive to lead with a low-cost, low-margin product.
 
  • Like
Reactions: lightmanek
Upvote 0 Downvote
Asterox

Asterox

Golden Member
May 15, 2012
1,026
1,775
136
Hans Gruber said:
Those AM5 motherboards need to come way down in price. Otherwise nobody will be buying AM5 CPU's.
Click to expand...

Harness your horses, much cheaper A620 AM5 motherboards are arriving in February. :grinning: AM5 socket was launched a few months ago, not say a year ago.

https://twitter.com/x/status/1613805358669234178

If we directly compare prices, and if we compare AM5/B650 technical details=you can't expect the same price it's not realistic at all.So these are motherboards from the same Gigabyte series, and to be honest the B450 model is a bit too expensive in year 2023.

Are you a human?

www.newegg.com www.newegg.com

Are you a human?

www.newegg.com www.newegg.com

Are you a human?

www.newegg.com www.newegg.com
 
  • Like
Reactions: Mopetar
Upvote 1 Downvote
biostud

biostud

Lifer
Feb 27, 2003
18,251
4,765
136
When there were a difference between functionality between X and B boards I can understand why you would choose an X board. But the X670E and B650E are based on the same chip (x2 for the X670E) , it is just the number of pcie lanes, sata and USB connectors that differ. How many users really need that?
 
  • Like
Reactions: Racan and Exist50
Upvote 0 Downvote
Det0x

Det0x

Golden Member
Sep 11, 2014
1,031
2,963
136
Last edited:
  • Like
  • Love
Reactions: Tlh97, DAPUNISHER, IEC and 6 others
Upvote 0 Downvote
J

JoeRambo

Golden Member
Jun 13, 2013
1,814
2,105
136
Tuna-Fish said:
Latency in cycles is lower for all levels of cache. That 1 cycle of L1 latency probably has a deceptively high impact. L1 has >95% hit rate on most well behaved loads, one extra cycle on every access adds up.
Click to expand...

Intel has advantages with size tho, 50% more L1 for just 1 more cycle of latency is good tradeoff. And 100% more L2 with 14 vs 16 or 17 cycles is damn good tradeoff as well.
Combined that creates a situation, when workload does not allow AMD to make use of their excellent L3, they get beaten badly by wider Intel's machinery:

5Ghz vs 5Ghz:
1673783277780.png
1673783243357.png
 
  • Like
Reactions: pcp7
Upvote 0 Downvote
K

Kaluan

Senior member
Jan 4, 2022
500
1,071
96
Hans Gruber said:
Those AM5 motherboards need to come way down in price. Otherwise nobody will be buying AM5 CPU's.
Click to expand...
Okay, this has got to be trolling by some at this point. I refuse to believe everyone muttering this has never at least glanced at DDR5 B660/760 motherboard prices. They're just as expensive if not more so than low end B650 offerings. And likely even more skimp on features.

Do people really look at 12th or 13th gen performance numbers (which are almost exclusively sporting fast/non-JEDEC DDR5) then look at a ("D4") super-basic $120 B660 boards and think "gee, I can have all that but for $50 less than those expensive B650 boards!" ?

Just looked at MSI's B760M Mortar WIFI and their B650M Mortar WIFI and the AM5 one is actually cheaper.

So I'm really bewildered about WTH people are on about.

1. ALL DDR5 MOBO OPTIONS COME WITH A PREMIUM, be it AMD or be it Intel.
2. You're not really buying "premium" 13th gen performance if you go DDR4.
3. If you're not building a Intel system based on future proofing and are aware you'll get less performance w/ DDR4 then go for it. But you should probably know AM4 is still a thing too. Often cheaper as well.

Once A620 mobos drop, I hope people can stop with that misguided talking point, AM5 may have the cheapest DDR5 boards on the market.
As far as I can tell, Intel H610 DDR5 boards don't even exist (and I doubt they reliably can run 200W+ CPUs either, but that's another subject).
 
  • Like
Reactions: Tlh97, DAPUNISHER, uzzi38 and 12 others
Upvote 0 Downvote
You must log in or register to reply here.

TRENDING THREADS

COMPANY
RESOURCES
FOLLOW
Top Bottom