Speculation: Ryzen 3000 series

Page 185 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

TheGiant

Senior member
Jun 12, 2017
748
353
106
Don't get me wrong, Zen is great, but the A64 - P4 gap was significant. And you mentioned the A64, not Phenom. Phenom II held it's own against Core 2, but by then, Nehalem was around.

And I always go where the performance (or performance/$) is. I upgraded from a 3570k. No fanboy here, though I generally go AMD if everything else is about equal. I like the longevity their sockets typically have. And, they're the underdog.
exactly
I remember changing p4 NW 2,4@3,0 to a-xp TB @2300MHz on nforce2- that CPU was so much better for min fps in gaming and general x87 computing (fortran at that time)
p4 without sse2 optimized code was a hand calculator compared to the athlon and with a64 things got even worse
at that time unreal tournament botmach was a CPU test good for gamers- the top a64 attacked 70fps and top prescott p4 (not the emergency editions) were in mid 30

now I think this r3x got too much hype
as I see it
IPC- looks like +3% vs skylake, which is fine, for me its a tie, so 4,635GHz CFL=4,5GHz r3x
what really matters is (for me) is how to get to performance of 9900K@5GHz and the voltage/power/mobo to get there - 4.85GHz r3x needed and above
we all know how voltage/power curve works
and I dont believe that TDP at all
 

Thunder 57

Diamond Member
Aug 19, 2007
3,423
5,659
136
exactly
I remember changing p4 NW 2,4@3,0 to a-xp TB @2300MHz on nforce2- that CPU was so much better for min fps in gaming and general x87 computing (fortran at that time)
p4 without sse2 optimized code was a hand calculator compared to the athlon and with a64 things got even worse
at that time unreal tournament botmach was a CPU test good for gamers- the top a64 attacked 70fps and top prescott p4 (not the emergency editions) were in mid 30

now I think this r3x got too much hype
as I see it
IPC- looks like +3% vs skylake, which is fine, for me its a tie, so 4,635GHz CFL=4,5GHz r3x
what really matters is (for me) is how to get to performance of 9900K@5GHz and the voltage/power/mobo to get there - 4.85GHz r3x needed and above
we all know how voltage/power curve works
and I dont believe that TDP at all

P4 was abysmal at unoptimized code or x87.
 
Last edited:

Carfax83

Diamond Member
Nov 1, 2010
6,841
1,536
136
Don't get me wrong, Zen is great, but the A64 - P4 gap was significant.

Indeed it was, but my reasoning for saying what I said is because you can win a battle, but lose the War. A64 was AMD winning an important battle, but losing the War in the long term. A64 didn't provide a strong enough foundation for AMD to iterate on it and compete with Intel's post Netburst architectures. Zen on the other hand, looks like it does provide a strong foundation for AMD to continue to improve upon, much like Intel did with Core.

And you mentioned the A64, not Phenom. Phenom II held it's own against Core 2, but by then, Nehalem was around.

Yes, but the gist of my argument was how A64 failed as a foundation for future micro-architectures, which is the exact opposite of Intel's Core.
 

phillyman36

Golden Member
Jun 28, 2004
1,789
201
106
Now do i wake up at midnight,3am or 11am my time(EST) to try to get a preorder lol. Hoping we start finding some website placement holders.
 
Reactions: lightmanek

Arkaign

Lifer
Oct 27, 2006
20,736
1,379
126
Yeah Athlon 64 (technically beginning with socket 940 Opteron) was one of the most significant moments in CPU history. Utter and complete domination in every aspect right up to Conroe literally years later. If they hadn't botched Phenom I, and PhII was what they had to match Conroe, then it would have been an even longer period of strength.

This time around, we're looking at relatively equal per core strength by all reasonable accounts (APISAK up there is notoriously reliable), only with more cores, and more perf/$. VERY good stuff, but not mind boggling like AMD64 was 16 years ago.

Desktop is sadly not what it once was either, though I think AMD's real magic will be putting 7nm Zen2 to work in laptops. That will be a massive sea change. The total marketplace for $250+ desktop CPUs is utterly minescule by comparison. But Zen2 should be capable of blowing the doors off Intel mobile processors until they can kick their process tech into anything competitive. Because of how it's measured, Intel 10nm is basically TSMC 7nm. If they can't solve it, then they're simply incapable of keeping up.

In any case, it's great to see Intel's feet held to the fire, they've been lazy, greedy, uninspired, and even incompetent. Being an undisputed market leader ever since Conroe in 2006 has been absolutely terrible from a culture and leadership standpoint, and much of that has to do with how Corporatism works, most especially with publicly traded companies. When you're competing essentially against yourself, you have every incentive to stagnate, profiteer, and lose the plot. Hubris syndrome is real, and examples abound throughout business history, and nobody ever seems to get that it can happen to them as well.

AOL, MCI, 3DFX, Blackberry, Nokia, etc, etc. No matter how gargantuan or successful a company is, greed and lack of vision can blow it for them eventually. And market leaders are particularly susceptible to such forces. Apple is beginning to look a lot like mid 80s IBM. Fat, arrogant, uninspired, and greedy. Intel has practically worn that description as a mantra for 10+ years as well. But let's be clear, nothing says AMD won't go the same route as well. Good companies can thrive in the pack, it takes exceptional ones to lead AND not fall prey to the destructive power of being on top. It's only too easy to start marking prices up, making arbitrary decisions, or cutting R&D to fatten the pockets of the elite. When CEOs do a bad job, they get paid tens or even hundreds of millions as an exit package, so it's not like they even have much incentive to chase anything deeper than immediate share prices. But, I trust Lisa Su more than your typical BS artist who makes their way to such positions. May AMDs leadership be long and successful, for all of our sakes. I think even Intel will be better off for having someone remind them what competence and inspiration can look like in action.
 

BigDaveX

Senior member
Jun 12, 2014
440
216
116
Indeed it was, but my reasoning for saying what I said is because you can win a battle, but lose the War. A64 was AMD winning an important battle, but losing the War in the long term. A64 didn't provide a strong enough foundation for AMD to iterate on it and compete with Intel's post Netburst architectures. Zen on the other hand, looks like it does provide a strong foundation for AMD to continue to improve upon, much like Intel did with Core.

Yes, but the gist of my argument was how A64 failed as a foundation for future micro-architectures, which is the exact opposite of Intel's Core.

I'm not sure that the Athlon 64 really was a failure in that regard - Phenom's problem wasn't that it was derived from the K8 core, but rather than it was released late and at lower clockspeeds than expected (not to mention the TLB glitch). Had Phenom I been going up against Conroe and Phenom II against Penryn, as was probably the original plan, it would have been a much more favourable situation for AMD.

If anything, AMD's real mistake with Phenom was producing a native quad core chip a generation earlier than they really needed to, given that desktop applications were still some ways off taking advantage of such a design, and the Xeons of the day were hobbled by atrocious memory latency courtesy of using FB-DIMMs.
 
Reactions: trollspotter

Hans de Vries

Senior member
May 2, 2008
346
1,175
136
www.chip-architect.com
It s not too high, if you look at Haswell scores it s obvious that use of AVX/AVX2 and 256bit FMA quite skew the result in favour of HW comparatively to Zen 1, the doubbling of the exe width in Zen 2 definitly null this advantage, not counting that it has about 10% better IPC than SKL, guess that people are still not used to see AMD leading in this area.

While we are not 100% sure yet about this Ryzen 5 3600 clocks there is some great humor in here. Just seeing the Ryzen 5 3600 as AMD's lowest new Zen2 CPU there at the very top-spot of a benchmark which was by design highly disadvantages for AMD. Talk about back firing: The first 120 entries or so are all Intel, except for this little entry level Zen 2 CPU which suddenly pops up at the number 1 position.

 
Last edited:
Reactions: Carfax83 and guachi

Asterox

Golden Member
May 15, 2012
1,042
1,837
136
If true then PassMark will urgently need to 'fix that bug' to keep intel on top.
I don't know how to read it though they all have low clocks for single thread.

No need for that, "Intel will apply the same Cinebench solution for Passmark=the test is not(it was in the past) relevant in real day to day practice."

Passmark vs Cinebench R20 hm.Who is more useful classic useless benchmark, or benchmark based on Maxon Cinema 4D.

 

inf64

Diamond Member
Mar 11, 2011
3,884
4,690
136
Just compared the passmark rating icelake gets (see here), it scored ~2625 at around 3.85Ghz (or 3.9Ghz). Worst case scenario for AMD in this benchmark:
1) Zen 2 (R5 3600) runs with XFR/PBO and its clock is ~4.4Ghz (maximum) - scoring 2979
2) Icelake runs at ~3.85Ghz (median speed reported during Geekbench runs)

Icelake @ 4.4Ghz would score 3000 pts, basically equaling Zen2.

Likely scenario:
1) r5 3600 runs at 4.3Ghz in ST Passmark test (median scores for a lot of Geekbench entries)
2) Icelake runs at 3.8-3.9Ghz range in ST Passmark test

Difference is around 3% between these two scenarios so more or less margin of error. Note that Icelake desktop parts (8+ cores) will likely have more cache and faster memory so the difference will be in Icelake's favor, but I expect sub 6% advantage here.
 

phillyman36

Golden Member
Jun 28, 2004
1,789
201
106
Are any of you going to use/enable the PBO on your 3900x (or any Ryzen cpu for that matter) when you get one? I'm not familiar with overclocking on AMD and how it works.
 

PotatoWithEarsOnSide

Senior member
Feb 23, 2017
664
701
106
Are any of you going to use/enable the PBO on your 3900x (or any Ryzen cpu for that matter) when you get one? I'm not familiar with overclocking on AMD and how it works.
If you've got an X CPU and an appropriate motherboard, your AMD CPU will take care of its own clocks.
PBO is a one-click outside of warranty overclock. You don't have to use it, and very few actually will.
 

phillyman36

Golden Member
Jun 28, 2004
1,789
201
106
If you've got an X CPU and an appropriate motherboard, your AMD CPU will take care of its own clocks.
PBO is a one-click outside of warranty overclock. You don't have to use it, and very few actually will.

Thanks. Right now my eyes are set on the 3900x and the Asus Crosshair VIII Hero.or the Strix -E. Leaning towards the Hero.
 

Kenmitch

Diamond Member
Oct 10, 1999
8,505
2,250
136
No need for that, "Intel will apply the same Cinebench solution for Passmark=the test is not(it was in the past) relevant in real day to day practice."

In the end that's the attitude a spoiled rich kid would have.

I'd imagine Intel views the 7th as a dentist appointment for multiple root canals.

The reviews will be interesting and can't come soon enough.
 
Reactions: DarthKyrie

Kenmitch

Diamond Member
Oct 10, 1999
8,505
2,250
136
Makes no difference if next Sunday the top 6 are all AMD.
In all honesty, how often do people scroll down the full list of scores on a benchmark? In reality we just look at the top handful.

It's going to be all about the gaming results in the end it looks like.

I just skim the results and head down to where the real action is.....The comment sections! It's most of the time more amusing than the reviews.
 
Reactions: lightmanek

Kedas

Senior member
Dec 6, 2018
355
339
136
The summary will be simple for games it's a wash depending on your favorite games one more than the other for a few %
AMD will be about equal or better in everything else and on top of that with a better price performance.

Reviewer go home we have all we needed.
I do want to see the score difference between an B450 and X570 though. (if there is any)
 

mopardude87

Diamond Member
Oct 22, 2018
3,348
1,575
96
Dang i just found out the asking price for the 32 threaded 3950X. What a absolute beast of a chip. Going to be very exciting to see how that one boosts. I am also eyeballing the 3700x as i think this one is going to be the fan favorite. About a $100 difference in price between a 3700x and its prob reasonably usable stock cooler and something like a 9900 non k and a D14/D15 cooler. Could be wrong about the 3700x and it may require such a cooler too but something tells me for stock the supplied stock cooler is going to be fine.

This may be the first time since C2D that i have actually been super excited about processors.
 
Reactions: lightmanek

moinmoin

Diamond Member
Jun 1, 2017
5,203
8,365
136
Yes, but the gist of my argument was how A64 failed as a foundation for future micro-architectures, which is the exact opposite of Intel's Core.
Athlon didn't really fail as a foundation, AMD just didn't have the resources to continue what it excelled at: hand optimized die circuits. The K10 (ending with Phenom II) still had superior IPC than the Construction/Cat cores that succeeded it, but the latters' die circuits were completely done by computers. The Construction/Cat cores just were bad follow ons to the K10 line, but AMD learned well with the Zen cores, and the switch to computer generated die circuits gives them the flexibility to do the semi-custom business as well as the near yearly cadence on Zen iterations.

Note that Icelake desktop parts (8+ cores) will likely have more cache and faster memory so the difference will be in Icelake's favor, but I expect sub 6% advantage here.
Is anybody really expecting Icelake desktop parts with 8+ cores? Intel's 10nm still seems too much of a mess to allow for any bigger dies or higher frequencies.
 

Kenmitch

Diamond Member
Oct 10, 1999
8,505
2,250
136
The saying " The king is dead, long live the king" keeps popping up in my mind.

Funny!

Out of the ashes a new king shall arise
The mighty one will stand dumbfounded

Needs a couple more versus, but it's a start
 

lobz

Platinum Member
Feb 10, 2017
2,057
2,856
136
Just compared the passmark rating icelake gets (see here), it scored ~2625 at around 3.85Ghz (or 3.9Ghz). Worst case scenario for AMD in this benchmark:
1) Zen 2 (R5 3600) runs with XFR/PBO and its clock is ~4.4Ghz (maximum) - scoring 2979
2) Icelake runs at ~3.85Ghz (median speed reported during Geekbench runs)

Icelake @ 4.4Ghz would score 3000 pts, basically equaling Zen2.

Likely scenario:
1) r5 3600 runs at 4.3Ghz in ST Passmark test (median scores for a lot of Geekbench entries)
2) Icelake runs at 3.8-3.9Ghz range in ST Passmark test

Difference is around 3% between these two scenarios so more or less margin of error. Note that Icelake desktop parts (8+ cores) will likely have more cache and faster memory so the difference will be in Icelake's favor, but I expect sub 6% advantage here.
let me paint you the realistic picture/scenario: when 10nm will have good enough yields to mass produce a 8C processor, it will have to compete with Zen 3 either right away or likely the next quarter
 
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/    |