Speculation: Ryzen 3000 series

Page 193 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Asterox

Golden Member
May 15, 2012
1,042
1,837
136
Oh, it actually isn't. That was just from a misleading comment from someone who read the review before it was taken down...

3700X is also pulling quite a bit of power in this review?

Compared to what maybe R7 2700X, and this is problem for 7nm CPU? It is very expected, if CPU has much more AVX or FPU resources or performance.For example Handbrake test, 25W higher power consumption vs R7 2700X.Who cares about energy consumption(this is for Desktop), if the CPU performance is much faster.

 
Reactions: prtskg

IEC

Elite Member
Super Moderator
Jun 10, 2004
14,568
5,965
136
Watts = Joules/second

Multiply the average power consumption in Watts by the number of seconds ~= Joules of energy to complete the task.

Basic physics, the 3000 series will win every time in efficiency.
 

DrMrLordX

Lifer
Apr 27, 2000
22,496
12,368
136
.Who cares about energy consumption(this is for Desktop), if the CPU performance is much faster.

As you may have noticed, there's been quite a to-do over exactly how many VRM phases you need for which upcoming Matisse chip. If people are going to try and cheap out on boards then they may not get the same performance as what they see in reviews (due to throttling) and be a bit upset.

When a 65W TDP chip using the default settings in their motherboard can pull more than a 105W TDP 2700x . . . what that tells me is that AMD is getting more aggressive in how far they're willing to let the chips boost with standard PB (not PBO) + XFR.
 

majord

Senior member
Jul 26, 2015
509
710
136
As you may have noticed, there's been quite a to-do over exactly how many VRM phases you need for which upcoming Matisse chip. If people are going to try and cheap out on boards then they may not get the same performance as what they see in reviews (due to throttling) and be a bit upset.

When a 65W TDP chip using the default settings in their motherboard can pull more than a 105W TDP 2700x . . . what that tells me is that AMD is getting more aggressive in how far they're willing to let the chips boost with standard PB (not PBO) + XFR.


Hope they're not following intels lead here - will be very disaspointing..


Thing is, in the Demonstrations they were keen to point out the huge power consumption delta betwen an 'uncapped' (i.e as reviewed) 9900k, and the 3800x (I think it was?) whilst pulling off the same Cinebench scores.. so that gives some hope the default behavior will stick close to TDP
 
Reactions: thigobr

Thunder 57

Diamond Member
Aug 19, 2007
3,444
5,682
136
Compared to what maybe R7 2700X, and this is problem for 7nm CPU? It is very expected, if CPU has much more AVX or FPU resources or performance.For example Handbrake test, 25W higher power consumption vs R7 2700X.Who cares about energy consumption(this is for Desktop), if the CPU performance is much faster.

View attachment 8058

Those Handbrake numbers are very impressive. Looks like widening the AVX units to match Intel has... matched Intel If anything they probably bested Intel at least in this benchmark considering clock speeds.

It looks AMD has taken away one of Intel's last advantages. I would say AVX 512 is still one, for those few who use it at the moment, and their cache structure on their mainstream parts. That will have to change with higher core counts and larger L2 caches, as has already happened with Skylake-X.
 

Carfax83

Diamond Member
Nov 1, 2010
6,841
1,536
136
I was going to say how on Earth is the 8700K and the 8086K ahead of the 9700K in Handbrake, until I realized that the 9700K lacks HT! The 3900x just decimates the competition though, making the top Intel and Threadripper CPUs practically obsolete!

The gaming benchmarks seem to be what I would have predicted. I would have expected Intel to maintain their lead in certain games that are more memory latency intensive and have less parallelism ie Far Cry 5, while Zen 2 would be faster in games that are more parallel and cache sensitive like AC Origins.

Actually, you can put that one down for a milestone for AMD. This is the first time that AMD has EVER beaten Intel in an Assassin's Creed game benchmark. The most recent AC games like AC Origins and AC Odyssey are all very CPU intensive, and Intel has always had a significant lead over AMD in those titles.
 

majord

Senior member
Jul 26, 2015
509
710
136
I was going to say how on Earth is the 8700K and the 8086K ahead of the 9700K in Handbrake, until I realized that the 9700K lacks HT! The 3900x just decimates the competition though, making the top Intel and Threadripper CPUs practically obsolete!

The gaming benchmarks seem to be what I would have predicted. I would have expected Intel to maintain their lead in certain games that are more memory latency intensive and have less parallelism ie Far Cry 5, while Zen 2 would be faster in games that are more parallel and cache sensitive like AC Origins.

Actually, you can put that one down for a milestone for AMD. This is the first time that AMD has EVER beaten Intel in an Assassin's Creed game benchmark. The most recent AC games like AC Origins and AC Odyssey are all very CPU intensive, and Intel has always had a significant lead over AMD in those titles.

Link to game benches?( I still can't find them )
 
Reactions: piesquared

Arkaign

Lifer
Oct 27, 2006
20,736
1,379
126
I was going to say how on Earth is the 8700K and the 8086K ahead of the 9700K in Handbrake, until I realized that the 9700K lacks HT! The 3900x just decimates the competition though, making the top Intel and Threadripper CPUs practically obsolete!

The gaming benchmarks seem to be what I would have predicted. I would have expected Intel to maintain their lead in certain games that are more memory latency intensive and have less parallelism ie Far Cry 5, while Zen 2 would be faster in games that are more parallel and cache sensitive like AC Origins.

Actually, you can put that one down for a milestone for AMD. This is the first time that AMD has EVER beaten Intel in an Assassin's Creed game benchmark. The most recent AC games like AC Origins and AC Odyssey are all very CPU intensive, and Intel has always had a significant lead over AMD in those titles.

We have to take it with a grain of salt until we get more credible reviews. The Intel tests were with 2666, which is god awful, and the Zens with 3400, but given the pure incompetence of using different memory, we can't be sure they even configured the Zen2 with any care either.

Fwiw, 2666 is so bad that it would make a 9900k slower than an 8700k with 3400 in gaming.
 

Kedas

Senior member
Dec 6, 2018
355
339
136
Since the 12 core 3900X is already very close to intels 18 core for video rendering it looks like the 16 core (3950X) of AMD will beat the 18 core for video rendering or at least have the same performance and with less power. NICE!
(for less than half the price)
 

BigDaveX

Senior member
Jun 12, 2014
440
216
116
Those Handbrake numbers are very impressive. Looks like widening the AVX units to match Intel has... matched Intel If anything they probably bested Intel at least in this benchmark considering clock speeds.

It looks AMD has taken away one of Intel's last advantages. I would say AVX 512 is still one, for those few who use it at the moment, and their cache structure on their mainstream parts. That will have to change with higher core counts and larger L2 caches, as has already happened with Skylake-X.
It actually reminds me of what happened with the K7 to K8 jump, where the former architecture could be quite competitive with P4 Northwood, but was hamstrung by a slower system bus, weak cache system and lack of SSE2 support. K8 addressed all those issues, and all of a sudden AMD went from being somewhat behind to completely annihilating anything Intel had to offer.
 
Reactions: lightmanek

Abwx

Lifer
Apr 2, 2011
11,733
4,656
136
but given the pure incompetence of using different memory,

Fwiw, 2666 is so bad that it would make a 9900k slower than an 8700k with 3400 in gaming.

Those are the maximum frequencies officially supported by the CPUs, anything above is already overclocking and not guaranted by the manufacturer, so they tested at stock specifications.
 

Arkaign

Lifer
Oct 27, 2006
20,736
1,379
126
Those are the maximum frequencies officially supported by the CPUs, anything above is already overclocking and not guaranted by the manufacturer, so they tested at stock specifications.

I get it, but it's just dumb, because it's not representative of what people actually do with builds. Anyone with a Z370/390 Mobo from a reputable brand has 'DDR4-4000' plastered all over the box, and setting XMP is cake. Looking at builds online, it's dominated by speeds between 3200-3600, with a handful of exotics above that, and a handful of nerf kits below that. It's like the AXP 1700+ back in the day, or a Celeron 300A. *Someone* out there ran them stock, but everyone knew they were legendary overclockers, so ran them at 2500+ and 450Mhz respectively, or thereabouts.

Not a big deal, I expect most reviewers to do more real-world style testing. I want to see them head to head with the best Bdie at 3733+ all the way, at least for the higher SKUs. Maybe on the 3600, it'd be interesting to compare it with budget 3000/3200 kits.
 
Reactions: TheGiant and mikk

amrnuke

Golden Member
Apr 24, 2019
1,181
1,772
136
I get it, but it's just dumb, because it's not representative of what people actually do with builds. Anyone with a Z370/390 Mobo from a reputable brand has 'DDR4-4000' plastered all over the box, and setting XMP is cake. Looking at builds online, it's dominated by speeds between 3200-3600, with a handful of exotics above that, and a handful of nerf kits below that. It's like the AXP 1700+ back in the day, or a Celeron 300A. *Someone* out there ran them stock, but everyone knew they were legendary overclockers, so ran them at 2500+ and 450Mhz respectively, or thereabouts.

Not a big deal, I expect most reviewers to do more real-world style testing. I want to see them head to head with the best Bdie at 3733+ all the way, at least for the higher SKUs. Maybe on the 3600, it'd be interesting to compare it with budget 3000/3200 kits.
What will be interesting to see is how the systems stack up at same memory frequency, especially at the 3600-3733 range.
 
Reactions: Arkaign

Kedas

Senior member
Dec 6, 2018
355
339
136
When does the NDA end? (I mean time, in which time zone?)

7min past 7, Timezone +7 ?
 
Last edited:

lobz

Platinum Member
Feb 10, 2017
2,057
2,856
136
It actually reminds me of what happened with the K7 to K8 jump, where the former architecture could be quite competitive with P4 Northwood, but was hamstrung by a slower system bus, weak cache system and lack of SSE2 support. K8 addressed all those issues, and all of a sudden AMD went from being somewhat behind to completely annihilating anything Intel had to offer.
yeah, that time it was some encoding benchmarks that all intel partizans kept repeating, saying how gaming is for children. How times change
 
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/    |