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

Page 382 - 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
800
1,364
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

TESKATLIPOKA

Platinum Member
May 1, 2020
2,373
2,868
136
The Anandtech review mentioned the IGPU could "possibly be used in light gaming." Has anyone found a review showing any benchmarks in that type of scenario. I am wondering if light gaming means solitaire or some kind of actual game. It says it even has ray tracing. Like could it run Battlefield 5 at resolution 320x240 and ray trace or is it just on paper and not actually usable?
Here is the average IGP performance from TPU. Link

Performance is very weak, but that was to be expected from a 2CU IGP.
 

Abwx

Lifer
Apr 2, 2011
11,057
3,716
136
Highest power draw in Blender of any CPU for the 7950X - double that of the 5950X for a mere 36% reduction in render time.

What are you talking about, the CPU hoover around 190W in Blender or whatever else.


Never late when it comes to vilify AMD, be it with dubbious numbers that do not hold when compared to other sites...
 

tamz_msc

Diamond Member
Jan 5, 2017
3,828
3,661
136
A step back in efficiency for sure, but still significantly more efficient than Alder Lake. Raptor Lake likely will be behind as well.
With E-core spam and PL2=PL1 being default behaviour, Intel has a chance of being more efficient, if not outright faster than the 7950X.
 
Reactions: Kaluan
Jul 27, 2020
16,827
10,781
106
View attachment 68117

Power efficiency in Blender. Looks like a regression vs Zen 3, but still more efficient than Intel chips overall.
So the additional performance benefit is not without considerable cost in electric bills. That 5950X looks mighty tempting there. All alone on its hill of efficiency.

Looks like Zen 3 is still the undisputed efficiency champion for 24/7 DC workloads, if other CPU heavy calculations mirror these results.
 

moinmoin

Diamond Member
Jun 1, 2017
4,975
7,736
136
Reactions: Tlh97 and Saylick

eek2121

Platinum Member
Aug 2, 2005
2,934
4,035
136
With E-core spam and PL2=PL1 being default behaviour, Intel has a chance of being more efficient, if not outright faster than the 7950X.

I think you don’t understand that Intel cannot break the laws of physics. Those cores don’t magically generate their own power to run on. There will possibly be improvement over Alder Lake, but Intel will very likely trail Zen 4, to say nothing about Zen 3 in terms of efficiency.
 

tamz_msc

Diamond Member
Jan 5, 2017
3,828
3,661
136
What are you talking about, the CPU hoover around 190W in Blender or whatever else.


Never late when it comes to vilify AMD, be it with dubbious numbers that do not hold when compared to other sites...
That's Prime95 where it hits 95 W after TDC limit. It thermal throttled in P95, hence the low power consumption. Check Gamers Nexus for actual physical measurement of power draw at the EPS cable.
 

eek2121

Platinum Member
Aug 2, 2005
2,934
4,035
136
That's Prime95 where it hits 95 W after TDC limit. It thermal throttled in P95, hence the low power consumption. Check Gamers Nexus for actual physical measurement of power draw at the EPS cable.

The chip is designed to hit 95C. In GN's testing it did so in Blender as well. According to AMD this is normal.

IF GN was hitting 95C with a 360mm AiO, I'm curious as to how high this chip would boost with a chiller or dual rads.

EDIT: Looking forward to picking up the 7950X. It looks like some of my workloads will be significantly improved.

I am kind of tempted to see if we get a 7950X3D, however.
 

Qwertilot

Golden Member
Nov 28, 2013
1,604
257
126
So the additional performance benefit is not without considerable cost in electric bills. That 5950X looks mighty tempting there. All alone on its hill of efficiency.

Looks like Zen 3 is still the undisputed efficiency champion for 24/7 DC workloads, if other CPU heavy calculations mirror these results.

Zen4 will be more efficient I presume, even quite a lot more so. Just need to wait for the versions limited to lower power draws. Like with the recent GPU's, some people need/want all the power they can get. Technology not keeping up so the top things have a higher power draw.

99% of consumers will be much better off with the non X versions, who knows when they'll release them. As someone said above, there'll be an awful lot of demand over in servers.
 

Hitman928

Diamond Member
Apr 15, 2012
5,392
8,281
136
"byhrough Zencally" sounds intriguing.


Need to know how it fares in ECO mode.


According to Computerbase, when set to the same TDP limit as a 5950x, the 7950x only loses 5% of its average multicore performance.



It also looks like Zen 4 continues the trend that undervolting is the new overclocking. With an undervolt they showed both less power consumption and higher clocks.





 

Abwx

Lifer
Apr 2, 2011
11,057
3,716
136
That's Prime95 where it hits 95 W after TDC limit. It thermal throttled in P95, hence the low power consumption. Check Gamers Nexus for actual physical measurement of power draw at the EPS cable.


In prime 95 it use 196W and 189W in Cinebench.

Measurement at the main for full system with Cinebench are displayed, it s 329W for the 7950X and 342/368W for the 12900K/KS respectively, so much for your "higher TDP" of all CPUs.

 

LightningZ71

Golden Member
Mar 10, 2017
1,631
1,903
136
So the additional performance benefit is not without considerable cost in electric bills. That 5950X looks mighty tempting there. All alone on its hill of efficiency.

Looks like Zen 3 is still the undisputed efficiency champion for 24/7 DC workloads, if other CPU heavy calculations mirror these results.
I'm not sold on your conclusion. How much power does a given ZEN4 part require to match the similar Zen3 part? How much work can the comparable Zen4 part do when its using the same power a ZEN3 part is using at maximum performance?

Judging from the few benches on eco mode that are out, it appears that ZEN4 can do more work per unit of energy than ZEN3 can until you push it to the absolute maximum performance that the silicon can achieve.
 

moinmoin

Diamond Member
Jun 1, 2017
4,975
7,736
136
Reviews seem to be noting that the 7950x runs at 95C (no throttling) even with 360mm AIOs and that this is intended by AMD.
While expected this makes it harder to account for thermal throttling. If temperature is essentially always at 95C anyway there needs to be some additional value to monitor, like thermal pressure or whatever to call it what can eventually lead to thermal throttling.
IF GN was hitting 95C with a 360mm AiO, I'm curious as to how high this chip would boost with a chiller or dual rads.
Yeah, reviews of different cooling systems for Zen 4 will be very interesting.
 

tamz_msc

Diamond Member
Jan 5, 2017
3,828
3,661
136
I think you don’t understand that Intel cannot break the laws of physics. Those cores don’t magically generate their own power to run on. There will possibly be improvement over Alder Lake, but Intel will very likely trail Zen 4, to say nothing about Zen 3 in terms of efficiency.
We'll see in a few weeks.
 

tamz_msc

Diamond Member
Jan 5, 2017
3,828
3,661
136
The chip is designed to hit 95C. In GN's testing it did so in Blender as well. According to AMD this is normal.

IF GN was hitting 95C with a 360mm AiO, I'm curious as to how high this chip would boost with a chiller or dual rads.

EDIT: Looking forward to picking up the 7950X. It looks like some of my workloads will be significantly improved.

I am kind of tempted to see if we get a 7950X3D, however.
In prime 95 it use 196W and 189W in Cinebench.

Measurement at the main for full system with Cinebench are displayed, it s 329W for the 7950X and 342/368W for the 12900K/KS respectively, so much for your "higher TDP" of all CPUs.

People here don't carefully read or parse the information that they like to post. 95 Watts is not the same as 95 Celsius. It (7950X) clearly throttles in Prime 95 as stated by Computerbase because of the TDC limit. As per Gamers Nexus physical power draw measurement, which is clearly better than reading off numbers from HWInfo, the 7950X is drawing 7 more watts than the 12900K.
 

exquisitechar

Senior member
Apr 18, 2017
657
872
136
Not many surprises. The details about the AVX-512 implementation and performance are nice. Overall, like I said after Raphael was first unveiled, I'm not impressed and this is the least exciting Zen generation on desktops after Zen+. Aside from the power efficiency in general and the MT performance of the 7950X, there's very little to be excited about when ADL exists and is almost a year old. AMD is coasting by hard on their node advantage and a fresh design is badly needed for next time. Hopefully, Zen 5 delivers.

Raptor Lake is going to crush Raphael in MT performance in the mid-range, have the ST performance crown by a small margin, and probably a big advantage in gaming performance with tuned memory. AMD needs to get V-Cache variants out as soon as possible.
According to Computerbase, when set to the same TDP limit as a 5950x, the 7950x only loses 5% of its average multicore performance.

View attachment 68126
Wow, the 7950X is unnecessarily, stupidly inefficient at stock. IMO, they should have saved the 230 W PPT for when they increase the core count. Barely better performance compared to 142 W and I already see people who think Zen 4 is inefficient and a hot mess, is it really worth it?
 
Reactions: Tlh97 and tamz_msc

Abwx

Lifer
Apr 2, 2011
11,057
3,716
136
People here don't carefully read or parse the information that they like to post. 95 Watts is not the same as 95 Celsius. It (7950X) clearly throttles in Prime 95 as stated by Computerbase because of the TDC limit. As per Gamers Nexus physical power draw measurement, which is clearly better than reading off numbers from HWInfo, the 7950X is drawing 7 more watts than the 12900K.

Computerbase measure power on the 230V main as well, and the 12900K/KS both drain more power, they are a more serious site than Gamernexus with way better methodology.

As for temp It s up to the manufacturer to decide what is a safe temp, 7950X limit was fixed at 95°C or so, on the other hand ADL is allowed to reach 102°C as displayed by Computerbase, so what is the fuss all about..?.
 
Last edited:

nicalandia

Diamond Member
Jan 10, 2019
3,330
5,281
136
I'm not impressed and this is the least exciting Zen generation on desktops after Zen+.
What? Nearly 30% ST performance Boost and Nearly 50% MT performance boost and you call this "Least Exciting"?

This is single handedly the most performance AMD has been able to extract since Bulldozer and you are unimpressed? Get Out Of Here...!
 
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/    |