Speculation: Ryzen 3000 series

Page 32 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

ub4ty

Senior member
Jun 21, 2017
749
898
96
AMD will launch server first (EPYC 2) and then Desktop Ryzen 3, so why did anyone expect for AMD to announce Desktop Ryzen 3 in CES 2019 when they havent officially launched EPYC 2 ??
Hmmm, is this the way Ryzen was initially launched? I'd expect them to detail Ryzen 3 and then Epyc 2. I'm sure there's no biggie if this is reversed
 

exquisitechar

Senior member
Apr 18, 2017
666
904
136
Hmmm, is this the way Ryzen was initially launched? I'd expect them to detail Ryzen 3 and then Epyc 2. I'm sure there's no biggie if this is reversed
It's been said by AMD that server comes first this time, forgot exactly when.
 

ub4ty

Senior member
Jun 21, 2017
749
898
96
The number of CCX's is given by the count. 16*16 implies that they are reporting it per CCX. This tells us two things. That AMD is still using the 4c CCX (I could see them going to 3 CCS's per chiplet but I figured the CCX would stay same size) and that AMD is doubling the L3 cache.

This is where tomorrow will answer. There has been suggestion by Adores source that they haven't seen an IO die for Ryzen 3. If that's the case it implies that that there is a different die for Ryzen. If that is the case, it might still only have 8MB per die.
I need to see something major on Ryzen 3 to make a move on it from 1st gen Ryzen.
7nm w/ power drop and clock pop isn't going to cut it with me.
PCIE 4.0 will get me interested but one more interesting change will make me sold.
I'd really like to see chiplet Ryzen 3. Have no expectations though for now... They either deliver what I'm looking for at a price I am ok with or they don't.

I'd love to learn more about Rome's I/O chip though and how it links to the chiplets.
Getting really comfy ahead of the talk for now...
 

rbk123

Senior member
Aug 22, 2006
745
348
136
Looks like 7nm Ryzen 3000 and 7nm Vega II for Q2
'soft launch' tomorrow.
https://wccftech.com/exclusive-follow-up-on-amds-agenda-for-ces-2019/

I wonder if this means that almost all specs will be given already...
That article waaayyyy over-emphasized the focus ("excitement") on being first to market with 7nm. Most don't care if it's 7nm or 700nm, it's the real life performance that matters and where the "excitement" needs to be placed. Hyper celebrating being first with 7nm implies a desire to get attention away from performance and onto something else. I hope I'm wrong and it's just the putz who wrote that article, but that's not a good sign.
 

DrMrLordX

Lifer
Apr 27, 2000
21,794
11,143
136
It's been said by AMD that server comes first this time, forgot exactly when.

Probably New Horizons. Technically AMD has already announced Rome, even though there are a lot of things we don't know yet about it. We just know it's coming, and that Rome chips will be the first ones to ship to those happy few who will get them.
 

ub4ty

Senior member
Jun 21, 2017
749
898
96
That article waaayyyy over-emphasized the focus ("excitement") on being first to market with 7nm. Most don't care if it's 7nm or 700nm, it's the real life performance that matters and where the "excitement" needs to be placed. Hyper celebrating being first with 7nm implies a desire to get attention away from performance and onto something else. I hope I'm wrong and it's just the putz who wrote that article, but that's not a good sign.
It's wccftech. They're literally the most bottom tier tech publication on the web.
As far as first to 7nm, that is the focus and everyone down to teenage enthusiast know this because its guaranteed to lower power and boost clocks. By how much, who knows...

Were a day away... The noise, click bait, and sensationalism hits peak right before announcements.
 

exquisitechar

Senior member
Apr 18, 2017
666
904
136
Looks like 7nm Ryzen 3000 and 7nm Vega II for Q2
'soft launch' tomorrow.
https://wccftech.com/exclusive-follow-up-on-amds-agenda-for-ces-2019/

I wonder if this means that almost all specs will be given already...
>As for how much detail AMD will go into, I can only guess, but my understanding is that they will not go into specific details at this time and are aiming for a more holistic approach to introducing their product (under promise, over deliver – the way it should be)
>From what I have heard this will be just a soft launch with the actual availability planned for sometime in Q2 2019.
Soft launch...more like no launch at all. Pointless article, aside from stating that there will be a consumer 7nm Vega. Remains to be seen if that's true or not.
 

FlanK3r

Senior member
Sep 15, 2009
313
38
91
My tip for tomorrow is:
AMD tell us Q of hard launch CPUs (propably Q2/Q3)
some specifications as number for cores, CCX, maybe up to MHz clock with XFR
live demo in Blender/Cinebench + some game (with Vega 7nm inside)
 

StinkyPinky

Diamond Member
Jul 6, 2002
6,825
870
126
I'm thinking Q2. Meets the yearly release schedule a lot of these types of companies have. Probably around April

Hardware Unboxed seem to think they won't announce anything to do with desktop ryzen but that makes zero sense to me. It's a long keynote, I doubt they are going to spend it talking about AI and 12nm laptop parts.
 

jpiniero

Lifer
Oct 1, 2010
14,826
5,442
136
Hardware Unboxed seem to think they won't announce anything to do with desktop ryzen but that makes zero sense to me. It's a long keynote, I doubt they are going to spend it talking about AI and 12nm laptop parts.

They could spend a bunch of time talking about Epyc 2. It'd be strange, being CES and all, but I don't think it's out of the question.

They could always do a couple demos and some teasing on Ryzen 3 without formally launching it.
 

Topweasel

Diamond Member
Oct 19, 2000
5,436
1,655
136
I'm thinking Q2. Meets the yearly release schedule a lot of these types of companies have. Probably around April

Hardware Unboxed seem to think they won't announce anything to do with desktop ryzen but that makes zero sense to me. It's a long keynote, I doubt they are going to spend it talking about AI and 12nm laptop parts.

Wasn't he one of the guys specifically trying to debunk Adore's leak? This is where a bunch of kids who haven't worked a real job do the internet version of road rage. Adored tried his best not to call anyone specifically out but has posted video's about tech journalism out for bad practices. They make a big deal so Adored covers his but. Adored supposedly has a source, that source actually matches up kind of well with the made for the fun of it spreadsheet (which may have been an attempt to keep people from wanting to figure out that source when the it went viral). Since then they seem to basically be taking the "If Adored says it, say the opposite". They want ammo to use against Jim, even if it means being wrong 100 times just to say that Jim was wrong once.
 
Mar 11, 2004
23,173
5,636
146
I'm thinking Q2. Meets the yearly release schedule a lot of these types of companies have. Probably around April

Hardware Unboxed seem to think they won't announce anything to do with desktop ryzen but that makes zero sense to me. It's a long keynote, I doubt they are going to spend it talking about AI and 12nm laptop parts.

Yeah, that wouldn't make much sense. EPYC Zen 2 has already been announced (and I feel like AMD said it was already being used by partners, same with Vega 20. I do expect them to do a big show (basically aping Intel and Nvidia) about their enterprise stuff (EPYC and Vega 20), so I'm not gonna be surprised if they talk that up quite a lot. And I really don't expect much detail (more just announcement) of Zen 2 based Ryzen 3000, and Navi (nothing much concrete, maybe core counts and some "up to" clock speed and performance claims on Ryzen compared to previous Ryzen, and Navi will basically just be "new architecture", and probably that it'll be using GDDR6). Would be disappointing to basically get nothing about either though.

Wasn't he one of the guys specifically trying to debunk Adore's leak? This is where a bunch of kids who haven't worked a real job do the internet version of road rage. Adored tried his best not to call anyone specifically out but has posted video's about tech journalism out for bad practices. They make a big deal so Adored covers his but. Adored supposedly has a source, that source actually matches up kind of well with the made for the fun of it spreadsheet (which may have been an attempt to keep people from wanting to figure out that source when the it went viral). Since then they seem to basically be taking the "If Adored says it, say the opposite". They want ammo to use against Jim, even if it means being wrong 100 times just to say that Jim was wrong once.

There's definitely something fishy going on. I have a hunch that AMD has been intentionally seeding misinformation (I doubt its even targeted at media, and more is to screw up say Intel and Nvidia getting info about what AMD is doing). Guess we'll see.

But, I still don't see how one can reconcile that AMD outright said Vega 20 is not consumer/gaming, that Navi is their first 7nm gaming chip, but that they're going to have some 7nm Vega II. I might be wrong, but I seem to recall AMD told their investors that outright when asked on one of their conference calls. They didn't give much more detail (just that Vega 20 was ahead of schedule and doing well; they didn't give any indication on Navi other than its in development on the way in the future and is 7nm). They haven't said anything about Vega II and I think its been off their roadmaps for over a year. Granted I don't know that they said anything about Polaris 30 (maybe they did, most of the stuff I saw pertained to 7nm and Navi and Vega 20, but I think if they'd have gotten any concreted details on it we'd have heard about it), so that leaves some room.

Plus how all the rumors were 16 core chip but now its suddenly just 12? Which I wouldn't be surprised if AMD did just the 12 core for now - I've even argued it might be smart to just tout higher clocking and reduced power 8 cores for now and then wait for like Computex or when there's an Intel product they'd like to show up, but maybe they want to stay ahead and know the 10 core is coming so they go for 12 now). And then either maybe Computex or when Intel announces or launches the 10 core they drop the 16 core bombshell.

There's a noticeable schism in the rumors and it seems like there's some undercurrent of rivalry or something at play. Caveat emptor! And make decisions based on what actually is there when you go to buy.
 

Mopetar

Diamond Member
Jan 31, 2011
8,004
6,445
136
Plus how all the rumors were 16 core chip but now its suddenly just 12? Which I wouldn't be surprised if AMD did just the 12 core for now

I think it all comes down to a matter of how IPC and clock speed stack up against Intel. If they've reached or come really close to parity, then a 12-core chip is already more than enough as it would have a ~50% advantage over the best Intel chip currently available.

The other side is how much they can produce. If they are using the same chiplets for Ryzen and Epyc, AMD gets more value from using full functional parts in Epyc or ThreadRipper. A 12-core Ryzen would presumably be from two salvaged dies with 6 functioning cores.

Holding the 16-core in reserve isn't a bad idea either, especially if they can't really produce enough of them right now anyways. If nothing else, they could always tease it. Don't even talk about the specifications for their top of the line Ryzen, but show off a benchmark just to make some big numbers.
 

cellarnoise

Senior member
Mar 22, 2017
728
399
136
I early on expected amd, and I do love the perpetual underdog to do well with intel's latest issues beyond their latest!

I hope they do well being first to a newer and smaller node.

I only expected from a smaller 14nm ryzen node refresh. A node change is normally a larger deal than the refinement cycles. So I was expecting refinement in memory latency, and maybe a few more cores (likely maybe 4 more - as how many can 2 channels can handle?)

Though I like the thought of a central I.O. with chiplets cranking out results with a relative fewer i.o. bandwidth.

As the Wall starts to come down, the tech has to get smarter to cache it
 

Vattila

Senior member
Oct 22, 2004
805
1,394
136
Wow! Cool to see how this thread has grown with discussion of all things related to Ryzen 3000. Now, with AMD's CES keynote just a few hours away, hopefully we will soon get answers to the poll, i.e. how 7nm Ryzen 3000 is designed (chiplet or monolithic) and some specifications on die configurations, core count and frequencies.

With the announcement of 12nm Ryzen Mobile APUs, I guess it is safe to presume that these chips will be coming to AM4 as well, replacing 2400G and 2200G, thus pretty much confirming one of the poll choices: 4-core monolithic APU (one 4-core CCX and an iGPU on SoC die).

With now two types of dies (IOC,chiplet) per CPU

Terminology: The dies in a disintegrated design are called "chiplets". For "Rome" we have 8 CPU chiplets and one IO chiplet.

Hyper celebrating being first with 7nm implies a desire to get attention away from performance and onto something else.

I agree that celebrating 7nm on its own, without meaningful gains in capability, frequency and power efficiency, makes little sense. But I suspect the 7nm generation signifies a lot of things for AMD — their whole business plan from 2014, when Lisa Su took over, is leading up to this. As Lisa Su alluded to in her recent interview with Fortune Magazine, they now envisage a strong multi-year product roadmap built on their 7nm leadership. They have tremendous opportunity.

But, I still don't see how one can reconcile that AMD outright said Vega 20 is not consumer/gaming, that Navi is their first 7nm gaming chip, but that they're going to have some 7nm Vega II.

As pointed out by others (perhaps elsewhere), note that Vega is a registered trademark and an internal codename for their architecture (by convention, codenames are quoted, as in "Vega"), while "Navi" is only an internal codename. They may use RX Vega II as a discrete GPU product name when bringing the "Navi" architecture to market.
 
Last edited:

DrMrLordX

Lifer
Apr 27, 2000
21,794
11,143
136
+ some game (with Vega 7nm inside)

Really?

But I suspect the 7nm generation signifies a lot of things for AMD — their whole business plan from 2014, when Lisa Su took over, is leading up to this.

It represents their shift away from GlobalFoundries (not completely, but it's the first major step). They have begun the exorcism of the legacy of Hector Ruiz.
 

Gideon

Golden Member
Nov 27, 2007
1,709
3,927
136

coercitiv

Diamond Member
Jan 24, 2014
6,384
12,802
136
Icelake with LPDDR4 will most certainly be considerably faster
Yeah, Intel finally pushing for fast DRAM with their mobile products was one fof the nice surprises from their CES announcement. Not only is this pushing iGPU performance up for both brands (AMD will benefit eventually from the industry widely adopting this new standard), we'll also have more fun comapring Geekbench scores vs. the dreaded Apple chips.
 

Zapetu

Member
Nov 6, 2018
94
165
66
While most of you have already found this, I will post this for those few that haven't spotted this live stream yet:


It should start about in half an hour. Let's hope for some concrete information about 7nm products.
 

NTMBK

Lifer
Nov 14, 2011
10,269
5,134
136
While most of you have already found this, I will post this for those few that haven't spotted this live stream yet:


It should start about in half an hour. Let's hope for some concrete information about 7nm products.

It'll be hilarious when they use the whole keynote to talk about their latest 28nm Bulldozer derivative for craptops.
 

vinhom

Member
Oct 14, 2016
25
5
81
Most recent AMD keynotes in past years we'd already have the slides available on WCCFTech or something by this point. I wonder what is going to happen.
 
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/    |