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

#26
GoldenX
TheLaughingMan said:

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.
Yeah, we will see if the chiplet design solves it for good.
Posted on Reply
#27
Dave65
Was there any doubt:D
Posted on Reply
#28
kid41212003
I'm starting to regret buying the 8700K...
Posted on Reply
#29
GoldenX
kid41212003 said:

I'm starting to regret buying the 8700K...
Don't, it can serve very well for many years. By the moment it's really the time to change it, AM4 will be old.
Posted on Reply
#30
TristanX
7nm allow for twice less power consumption, over 16 nm, so you have it - 16C drawing the same power as single Ryzen 2700X. There won't be faster clocks than 4.3 GHz, because there is additional IO chip, and raising speed will increase power consumption beyond power delivery capacity. Some may try to OC, but traditionally, Ryzens do not OC well
Posted on Reply
#31
xorbe
TheLaughingMan said:

Is this a confirmation that these rumors are true or simply that they exist?
This is the big question. We can safely assume 16/32 exists, but will they offer it, and when.
Posted on Reply
#32
xkm1948
16c32t is great, but feeding that amount of cores with dual channel DDR4 with the IMC outside of compute cores? That memory sub system performance may not look good at all.
Posted on Reply
#33
Eric3988
Sounds very intriguing, let the price wars commence!
Posted on Reply
#35
bug
Even if this is totally legit, feeding 16c/32t needs at least a quad-channel RAM setup. It's not going to make its way into the mainstream.
Me, I'm just curious what I can get in the $200-300 range to make me ditch my current CPU.
Posted on Reply
#36
mtcn77
GorbazTheDragon said:

Having dabbled with a 5675c the last few months I think you have to add quite a lot of overhead for background tasks. Even if the individual tasks are very low load, the thread switching ends up taking a huge amount of time so having a large thread count is almost always better.
Good chip; wrong task. L4 helps in deep cache searches, emphasizing single-thread performance. You tried to multitask, that severs cache lines by half.
Posted on Reply
#37
Joss
bug said:

feeding 16c/32t needs at least a quad-channel RAM setup. It's not going to make its way into the mainstream
Maybe AMD is coming up with its own version of Intel's HEDT.
Posted on Reply
#38
GoldenX
Joss said:

Maybe AMD is coming up with its own version of Intel's HEDT.
Yeah, that's the Threadripper's idea, AM4 is designed with only dual channel in mind.
Posted on Reply
#39
XiGMAKiD
16-core AM4 is possible and almost everyone knows it, it's the clock speed that's a bit unbelievable
Posted on Reply
#40
Manu_PT
GorbazTheDragon said:

Having dabbled with a 5675c the last few months I think you have to add quite a lot of overhead for background tasks. Even if the individual tasks are very low load, the thread switching ends up taking a huge amount of time so having a large thread count is almost always better.
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.

yakk said:

Competition is good!

Otherwise we'd still have 4 core CPUs at $500+ just incredible.
Wich quad core was more than 500?? Cant remember

HimymCZe said:

considering even Ryzen 2000 have better IPC than i9,
even "just" 4,3Ghz Ryzen 3000 will EASILY beat i9. Every task. Every game. Every price.
All hail the new king.
Ryzen 2000 has better ipc than coffee lake? Lol ok.

kid41212003 said:

I'm starting to regret buying the 8700K...
If you mostly do gaming dont worry. Intel will still be faster than zen 2. You read it here first. If you are into multi threaded apps, 2700 would have been your best friend anyway.
Posted on Reply
#41
GoldenX
Manu_PT said:

Wich quad core was more than 500?? Cant remember
Not much, mostly older ones: The Core 2 Quad Extreme line, the Core i7 Extreme Edition line, the i7 870 and 880.
After that, mostly USD 300-350, for ten years.
Posted on Reply
#42
theeldest
I think we'll still see slightly higher clocks on the final skus. We already have direct comparisons from AMD on 14nm & 7nm with Vega & the Radeon VII. From vega64 to VII clocks went up 12.2% for base and 13% for boost. That's like an 1800x going from 3.6/4.0/4.1 to 4.0/4.5/4.6 (base/turbo/xfr). That's pretty reasonable.

And the fact that it's 16 cores doesn't really matter as it's more about whether you can get 2x 8-core chips to run at 4/4.6 which is much easier.
Posted on Reply
#43
theoneandonlymrk
xkm1948 said:

16c32t is great, but feeding that amount of cores with dual channel DDR4 with the IMC outside of compute cores? That memory sub system performance may not look good at all.
Its been a while since it was done and even then it was off chip not on chip(mcm) but it's quite possible it's fine you know, we'll see.
It's not like other chips and systems haven't done similar.
Posted on Reply
#44
lsevald
Personally I want something that performs on par or better than the 9900k, both in apps and games, for half the price :p. Can we please have a 8c/16t Ryzen 5 "3650x" AMD? Too bad if the R5's only get the lower binned dies; 4.0/4.8 boost according to rumor, while R9 go as high as 5.1GHz.
Posted on Reply
#45
biffzinker
xkm1948 said:

but feeding that amount of cores with dual channel DDR4 with the IMC outside of compute cores? That memory sub system performance may not look good at all.
A last Level 4 cache on the I/O die might be a way around that hold up.
Posted on Reply
#46
advanced3
While I believe the core count, I am not buying into the Clock speeds.
Posted on Reply
#47
bonehead123
Well, see, the $64MM question here is:

Now that they know that you know that they know you know, whatchaz gonna do 'bout it ?

hehehe......
Posted on Reply
#48
lexluthermiester
EarthDog said:

maybe bt knows something we don't...
He might. However, even if it's just a hypothesis, he's been right many time's before. I think it'll be safe to think that's if this info isn't spot-on, it'll be very close.

advanced3 said:

While I believe the core count, I am not buying into the Clock speeds.
Why not?
Posted on Reply
#49
GoldenX
advanced3 said:

While I believe the core count, I am not buying into the Clock speeds.
No one believed Zen's IPC gain over FX, even me, and yet here we are.
Posted on Reply
#50
Darmok N Jalad
I still don’t see why 5.0+ isn’t possible. Zen 1 was a first try at a new architecture, and Zen 1.5 wasn’t a massive departure as much as it was some tweaks and a new node. Zen 2 is AMD’s first real refinement, and it’s on 7nm to boot. It sure seems like current Zen offerings have an architectural limit preventing higher clocks. Just a wild guess, but that may be one of the things that the chiplet design solves, as the IO, IMC, and IF components may be what was so sensitive to faster clocks. Won’t be long now to find out. I hope AMD discusses it when Zen 2 arrives.
Posted on Reply
Add your own comment