Thursday, May 9th 2019

AMD Ryzen 9 3000 is a 16-core Socket AM4 Beast

AMD is giving finishing touches to its 3rd generation Ryzen socket AM4 processor family which is slated for a Computex 2019 unveiling, followed by a possible E3 market availability. Based on the "Matisse" multi-chip module that combines up to two 8-core "Zen 2" chiplets with a 14 nm I/O controller die, these processors see a 50-100 percent increase in core-counts over the current generation. The Ryzen 5 series now includes 8-core/16-thread parts, the Ryzen 7 series chips are 12-core/24-thread, while the newly created Ryzen 9 series (designed to rival Intel Core i9 LGA115x), will include 16-core/32-thread chips.

Thai PC enthusiast TUM_APISAK confirmed the existence of the Ryzen 9 series having landed himself with an engineering sample of the 16-core/32-thread chip that ticks at 3.30 GHz with 4.30 GHz Precision Boost frequency. The infamous Adored TV leaks that drew the skeleton of AMD's 3rd generation Ryzen roadmap, referenced two desktop Ryzen 9 parts, the Ryzen 9 3800X and Ryzen 9 3850X. The 3800X is supposed to be clocked at 3.90 GHz with 4.70 GHz boost, with a TDP rating of 125W, while the 3850X tops the charts at 4.30 GHz base and a staggering 5.10 GHz boost. The rated TDP has shot up to 135W. We can now imagine why some motherboard vendors are selective with BIOS updates on some of their lower-end boards. AMD is probably maximizing the clock-speed headroom of these chips out of the box, to preempt Intel's "Comet Lake" 10-core/20-thread processor.
Sources: TUM_Apisak, Tom's Hardware
Add your own comment

197 Comments on AMD Ryzen 9 3000 is a 16-core Socket AM4 Beast

#51
moproblems99
EarthDog, post: 4044991, member: 79836"
Nobody likes the increased power consumption, but, let's be clear here, heat and temperature are two different things. For example, which has the higher temperature... a lighter with a yellow flame or a bonfire with yellow flames? The answer.......they are the same temperature, but cleary a bonfire has more energy behind the temperature. I mean I have seen 5W ASICs on mining rigs burn my finger...
It's probably because that is what everyone shits on AMD about...But the heat, but drivers, but the temperatures. Intel CPU gobbling up 200W more power for 5% more performance over AMD: TAKE MY MONEY!

It really is funny to watch at times.

EarthDog, post: 4045008, member: 79836"
Does it? Are there other reviews showing different than this one?
https://www.techspot.com/article/1616-4ghz-ryzen-2nd-gen-vs-core-8th-gen/page3.html
Compare apples to apples (6c/12t) and look at the 1600X at 4 GHz versus the 8700K at 4 GHz....

now... their SMT IS more efficient than Intel and in heavily threaded benchmarks which use smt. IPC measurements are typically single threaded, and smt/ht cant be involved, otherwise its a multi-threaded benchmark which shows the difference between HT and SMT.
Why would they use the 2600X as their comparison when the 2700X was less than 5% different nearly all the time?

EDIT: Duh, thread count. My bad.

EDIT 2: 200W more power for 5% are just general approximates, not intended to be scientifically accurate.
Posted on Reply
#52
EarthDog
moproblems99, post: 4045139, member: 155919"
It's probably because that is what everyone shits on AMD about...But the heat, but drivers, but the temperatures. Intel CPU gobbling up 200W more power for 5% more performance over AMD: TAKE MY MONEY!
Maybe their gpus. Cpus arent really an issue.

I can top out a 2700x at 4.2 ghz all C/t with a subpar 240 aio... (same with most 9900k at 5ghz). It's not temps that hold these things back. World record for a screenshot of 2700x is 6 ghz... 9900k 7.6 ghz... the arch on this process just doesnt have it in them for higher clocks regardless of temps. BD was the opposite... 8ghz+ while Intel was 7ghz+.

I really hope that 16c/32t monstrosity is for TR.... this blurring of the lines and cores wars is getting old before it really starts, lol.
Posted on Reply
#53
TheLostSwede
TheLaughingMan, post: 4044918, member: 70334"
Is this a confirmation that these rumors are true or simply that they exist?
Yes
Posted on Reply
#54
moproblems99
EarthDog, post: 4045148, member: 79836"
It's not temps that hold these things back.
Not really talking about limits, thresholds, or holding back. More of in general.

Although, I don't believe those leaks for nothing. If Zen2 makes it over 4.6, I'd be shocked. Like you said, it's not thermals or power holding these back.
Posted on Reply
#55
EarthDog
In general, that's hilarious because outside of GPUs, I dont recall people bitching about drivers, temps, etc as it typically isnt a limit either..... :)
Posted on Reply
#56
moproblems99
EarthDog, post: 4045148, member: 79836"
I really hope that 16c/32t monstrosity is for TR.... this blurring of the lines and cores wars is getting old before it really starts, lol.
Honestly, I don't really care if they put 1000 cores on it if the price is reasonable. And by price, I mean total system price. If they keep boards to a reasonable price with necessary features and not just adding useless features and driving price up, what is the harm? People can just buy the cpu with the core count that fits them. You don't have to buy the highest numbered part to say you did.

Again, saying 'you' in general not 'you' as in EarthDog.

EarthDog, post: 4045153, member: 79836"
In general, that's hilarious because outside of GPUs, I dont recall people bitching about drivers, temps, etc as it typically isnt a limit either..... :)
That is my point, temperature and power are only a concern when 'their brand' is winning in those categories. Other wise, no one really cares.
Posted on Reply
#57
TheLostSwede
advanced3, post: 4045106, member: 186668"
While I believe the core count, I am not buying into the Clock speeds.
You should.

moproblems99, post: 4045152, member: 155919"
Not really talking about limits, thresholds, or holding back. More of in general.

Although, I don't believe those leaks for nothing. If Zen2 makes it over 4.6, I'd be shocked. Like you said, it's not thermals or power holding these back.
Prepare to be electrocuted.
Posted on Reply
#58
EarthDog
I'm sure, like Intel, it will be single/two cores... it will make 4.6 single dual/core without flinching. 5 Ghz, I bet we'll see a SKU with that as well.
Posted on Reply
#59
moproblems99
TheLostSwede, post: 4045157, member: 3382"
Prepare to be electrocuted.
Static fine?
Posted on Reply
#60
kings
Unless an 8-core comes up that can get close to 5.0Ghz in OC, it will not be easy to get rid of my beloved 5820K@4.8Ghz.
Posted on Reply
#61
Mamya3084
Is it possible that x570 adds quad channel memory support?
Posted on Reply
#63
TheLostSwede
Mamya3084, post: 4045178, member: 186201"
Is it possible that x570 adds quad channel memory support?
Why would it? The memory controller is in the CPU/SoC, not the chipset. Regardless, see the post above.

kings, post: 4045162, member: 180022"
Unless an 8-core comes up that can get close to 5.0Ghz in OC, it will not be easy to get rid of my beloved 5820K@4.8Ghz.
I think you will be pleasantly surprised.
Posted on Reply
#64
Mamya3084
I would have thought that some of the reserved pins could lead do additional quad channel support on x570. Not necessarily the chipset itself. Anyway, doesn't really matter.
Posted on Reply
#65
Crackong
kings, post: 4045162, member: 180022"
Unless an 8-core comes up that can get close to 5.0Ghz in OC, it will not be easy to get rid of my beloved 5820K@4.8Ghz.
There is an existing product : 9900k .
If you find the 9900k's isn't enough to convince you, there must be something other than 8 cores 5GHz capability in your mind.
Posted on Reply
#66
silapakorn
Maybe this time I'll go back to red team after over a decade of being on blue team.
Posted on Reply
#67
Caring1
3.3GHz - 4.3GHz 16 core is basically confirmed, the rest is still bullshit.
Posted on Reply
#68
TheLostSwede
Caring1, post: 4045198, member: 153156"
3.3GHz - 4.3GHz 16 core is basically confirmed, the rest is still bullshit.
Or not.
Posted on Reply
#69
GorbazTheDragon
Manu_PT, post: 4045080, member: 168799"
GamerNexus talked about that on one of his recent videos and clearly stated it does not work like that. Having stuff open in background while gaming doesnt benefit from 50 cores. Ram yes. Unlees you are encoding or rendering while playing games.
Ran 16GB DDR3 1866c9 on the 1150 system and 16GB 3200 16-17-17 on the 7700k so doesn't really match up.

The 4790k was noticeably better than the 5675c regardless, so the variables between those two are basically the same. Sure the broadwell chips are pretty rubbish clock wise but I really don't think you can chock it up to the ~10% difference in single core perf.
Posted on Reply
#70
cucker tarlson
TheLaughingMan, post: 4045007, member: 70334"
No. It has been shown dozens of times that Intel is still king mainly because of clock speed and IPC. Whenever the clock for clock tests are done, Intel loses almost its entire gaming performance lead and drops to low single digit % leads. Latency is an issue, but not nearly the game performance killer you think.
no,it's still +10% in many cases with ryzen 2600x having 200mhz advantage over 8400
https://www.purepc.pl/procesory/jaki_procesor_wybrac_test_amd_ryzen_5_2600_vs_intel_core_i5_8400?page=0,39
https://www.purepc.pl/procesory/jaki_procesor_wybrac_test_amd_ryzen_5_2600_vs_intel_core_i5_8400?page=0,36
https://www.purepc.pl/procesory/jaki_procesor_wybrac_test_amd_ryzen_5_2600_vs_intel_core_i5_8400?page=0,35
https://www.purepc.pl/procesory/jaki_procesor_wybrac_test_amd_ryzen_5_2600_vs_intel_core_i5_8400?page=0,37
https://www.purepc.pl/procesory/jaki_procesor_wybrac_test_amd_ryzen_5_2600_vs_intel_core_i5_8400?page=0,38
https://www.purepc.pl/procesory/jaki_procesor_wybrac_test_amd_ryzen_5_2600_vs_intel_core_i5_8400?page=0,43
Posted on Reply
#71
Gungar
More fake news? Btarunr is on fire this days...

DeathtoGnomes, post: 4044920, member: 151150"
Wonder if this still drop in existing motherboards without much hassle other than a BIOS update.
Well i did that with my 9900k with the Z370 maximus X hero, Oh boy i was wrong. I switched to Z390 aorus master, its night and day, more stable, the voltage distribution is A LOT BETTER and more efficient.
Posted on Reply
#72
ratirt
I think the arch for Ryzen is set. AMD is focusing on more frequency and latency increase and maybe some minor tweaks. They are pretty confident in what they are doing and they've got a plan which I believe more or less will be fulfilled. AMD is really back and you can see that with Intel's attempts to get the new node going, lowering prices, new processors releases scheduled. It is on. :)
Posted on Reply
#73
medi01
How to undermine a product 101: overhype it, to get dissapponted later.

16 cores is a given, 5Ghz not at all.

eidairaman1, post: 4044969, member: 40556"
My FX 8350 does 5.0 on air, older node to boot...
That's getting high clocks Prescott ways.
Posted on Reply
#74
Shatun_Bear
The nth time that fake AdoredTV numbers have been referenced.

He made the numbers up, as they come from a chart in his video all the way back December 2018! There is no way AMD knew the base/boost and price of every Ryzen 3K SKU back at that date. Also, he listed one of the SKUs as having a base clock merely 100mhz short of the boost clock of the Ryzen 2700x! I mean if that is not a red herring for his info, I don't know what is.
Posted on Reply
#75
TheLostSwede
Gungar, post: 4045250, member: 163163"
More fake news?
Nope.

Shatun_Bear, post: 4045266, member: 166032"
The nth time that fake AdoredTV numbers have been referenced.

He made the numbers up, as they come from a chart in his video all the way back December 2018! There is no way AMD knew the base/boost and price of every Ryzen 3K SKU back at that date. Also, he listed one of the SKUs as having a base clock merely 100mhz short of the boost clock of the Ryzen 2700x! I mean if that is not a red herring for his info, I don't know what is.
I hope you enjoy eating crow.
Posted on Reply
Add your own comment