[WCCF] AMD Zen 3 set to deliver >8% IPC with 200 MHz better clock. - Page 2 - Overclock.net - An Overclocking Community
Forum Jump: 

[WCCF] AMD Zen 3 set to deliver >8% IPC with 200 MHz better clock.

Reply
 
Thread Tools
post #11 of 67 (permalink) Old 10-13-2019, 12:10 AM
New to Overclock.net
 
DNMock's Avatar
 
Join Date: Jul 2014
Location: Dallas
Posts: 3,613
Rep: 171 (Unique: 125)
Zen 3? the 3950X hasn't even been released yet and I'm sure they will do a refresh on Zen 2 just like they did for Zen 1. It feels a wee bit early for talk about zen 3.


DNMock is offline  
Sponsored Links
Advertisement
 
post #12 of 67 (permalink) Old 10-13-2019, 12:17 AM
Waiting for 7nm EUV
 
tpi2007's Avatar
 
Join Date: Nov 2010
Posts: 11,469
Rep: 898 (Unique: 504)
Quote: Originally Posted by DNMock View Post
Zen 3? the 3950X hasn't even been released yet and I'm sure they will do a refresh on Zen 2 just like they did for Zen 1. It feels a wee bit early for talk about zen 3.

Nope. Zen 3 based Milan is on track.


https://www.anandtech.com/show/14568...me-milan-genoa

Quote:
But for the removal of doubt, Genoa is the Zen 4 design. Here's an exclusive: the one after that is another Italian City. I know you're shocked. We always knew that having the regular cadence in the road map was extremely important, and so the strategy was to have Rome be a part that could further accentuate our throughput leadership, and to get rid of as many asterisks as possible in terms of single thread performance, memory latency, and so on. That was always the strategy. Then Milan was designed to further erase any asterisks that remain, so in thinking about it, in the original strategy, Milan was where we expected to be back to IPC (or better) parity across all workloads. That was the thought process to the strategy.

Even with Intel's road map delays, we have continued to execute within a quarter of when we said we were going to execute - we've always come out within one quarter of where we planned it 3 years ago. Milan is right on track as well, but Intel is not, and so the relative competitiveness of Rome is actually better than we had originally had planned. We're seeing tremendous excitement in our customer base around it, and I think that's really a function of it being more competitive, but it's also a function of the fact that our partners now believe us. Now AMD delivered the original part and it works, and it's good and it's quality, we can deploy it at scale. We told customers two years ago that we we're going to come out with Rome in the middle of 2019 and now we’re telling them that we’re going to come out with Milan at a certain point in time, and they believe us. That helps a lot that belief, that the proof points, the predictable execution are very important.


tpi2007 is offline  
post #13 of 67 (permalink) Old 10-13-2019, 12:22 AM
sudo apt install sl
 
Join Date: Dec 2009
Posts: 5,889
Rep: 182 (Unique: 125)
Quote: Originally Posted by DNMock View Post
Zen 3? the 3950X hasn't even been released yet and I'm sure they will do a refresh on Zen 2 just like they did for Zen 1. It feels a wee bit early for talk about zen 3.
Zen 3 has been on AMD's roadmap since 2017 for a release date of 2020. Their recent roadmaps show it's on track.

Silent
(20 items)
CPU
Core i9 9900K... CoffeeTime! @ 5.1Ghz w/ 1.36v
Motherboard
Maximus VIII Formula
GPU
Radeon VII @ 2100Mhz/1250Mhz w/ 1218mV
RAM
TeamGroup Xtreem 32GB 3200Mhz CL15
Hard Drive
HP EX950 2TB
Hard Drive
Samsung 850 Evo 1TB
Hard Drive
Samsung 850 Evo 1TB
Power Supply
EVGA SuperNova 1200w P2
Cooling
EK Supremacy Full Copper Clean
Cooling
XSPC D5 Photon v2
Cooling
Black Ice Gen 2 GTX360 x2
Cooling
EK-Vector Radeon VII - Copper + Plexi
Case
Thermaltake Core X5 Tempered Glass Edition
Operating System
Clear Linux
Monitor
Acer XF270HUA
Keyboard
Cherry MX Board 6.0
Mouse
Logitech G600
Mouse
Alugraphics GamerArt
Audio
Definitive Technology Incline
Audio
SMSL M8A
▲ hide details ▲
WannaBeOCer is offline  
Sponsored Links
Advertisement
 
post #14 of 67 (permalink) Old 10-13-2019, 01:33 AM - Thread Starter
New to Overclock.net
 
guttheslayer's Avatar
 
Join Date: Apr 2015
Posts: 3,805
Rep: 111 (Unique: 65)
Quote: Originally Posted by DNMock View Post
Zen 3? the 3950X hasn't even been released yet and I'm sure they will do a refresh on Zen 2 just like they did for Zen 1. It feels a wee bit early for talk about zen 3.
You are not keeping up with the news. AMD presented their slide and it already show its Zen+, Zen 2 up to Zen 4.




You can clear see there is no Zen 2+. The fact that AMD include Zen+ inside means they are not omitting any "+" design at all, if there is any.

guttheslayer is offline  
post #15 of 67 (permalink) Old 10-13-2019, 07:46 AM
Spaghetti
 
Buris's Avatar
 
Join Date: May 2010
Location: Earth
Posts: 1,075
Rep: 42 (Unique: 34)
Quote: Originally Posted by epic1337 View Post
dont forget the 4thread per core rumor.
https://www.notebookcheck.net/Rumor-AMD-Zen-3-architecture-could-run-four-threads-per-core.435660.0.html

SMT4 (4-way SMT) coupled with 8core CCX means 8C/32T per CCX.
This won't happen for Zen 3, maybe Zen 4?

But, It's possible that with a die shrink of 15%, and the CCX now hosting 8-cores with shared L3, that AMD is now able to make chiplets that are small enough to cram 3 into a single CPU....

AKA: 24c/48t AM4 CPU.

MARS
(18 items)
CPU
Ryzen 9 3900x
Motherboard
Gigabyte X570 Aorus Master
GPU
EVGA 1080 Ti FTW3
RAM
G-Skill Sniper
Hard Drive
WD EMAZ001
Hard Drive
WD Black SN750
Power Supply
EVGA G3
Cooling
Noctua U12A
Case
NZXT H500
Operating System
Microsoft Windows 10
Operating System
Pop! OS
Monitor
BenQ XL2730z
Keyboard
Aukey Mechanical
Mouse
Mionix Castor
Mousepad
Aukey
Audio
PreSonus Eris E3.5
Audio
SennHeiser 598
Audio
Blue Yeti
▲ hide details ▲
Buris is offline  
post #16 of 67 (permalink) Old 10-13-2019, 08:05 AM - Thread Starter
New to Overclock.net
 
guttheslayer's Avatar
 
Join Date: Apr 2015
Posts: 3,805
Rep: 111 (Unique: 65)
Quote: Originally Posted by Buris View Post

But, It's possible that with a die shrink of 15%, and the CCX now hosting 8-cores with shared L3, that AMD is now able to make chiplets that are small enough to cram 3 into a single CPU....

AKA: 24c/48t AM4 CPU.
Already people have explain why its not possible, power delivery to feed 50% more cores (with the same socket pin as current Zen 2) and bandwidth limitation is the biggest reason why you wont see a core count increase at all.


Zen 3 is compatible with AM4 socket. This alone tell you the core count will stagnant because its reaching limit on current platform with Dual DDR channel to support 16C/32T.


Just look at the TR platform socket and you will see how many pins is need to support 24C and above. These high number of pins are not for show, they are there to delivery more power to the power-hungry cores.

guttheslayer is offline  
post #17 of 67 (permalink) Old 10-13-2019, 08:06 AM
Overclocking Enthusiast
 
ozlay's Avatar
 
Join Date: Aug 2009
Location: USA
Posts: 5,708
Rep: 280 (Unique: 222)
ZEN 4 on AM5 "more pins" with PCIe5.0 and DDR5. SMT4 and 5nm?

Edit: Forgot USB 4.0

Rikka
(9 items)
CPU
Intel I5-7600k
Motherboard
Asrock Z270M Extreme 4
GPU
Titan X
RAM
Gskill Royal
Hard Drive
970 Evo
Hard Drive
Optane 900p
Power Supply
Seasonic Prime Snow
Case
enthoo evolv matx tempered glass
Operating System
windows 10 pro
▲ hide details ▲



Last edited by ozlay; 10-13-2019 at 06:04 PM.
ozlay is offline  
post #18 of 67 (permalink) Old 10-13-2019, 08:13 AM
Spaghetti
 
Buris's Avatar
 
Join Date: May 2010
Location: Earth
Posts: 1,075
Rep: 42 (Unique: 34)
Quote: Originally Posted by guttheslayer View Post
Already people have explain why its not possible, power delivery to feed 50% more cores (with the same socket pin) and bandwidth limitation is the biggest reason why you wont see a core count increase at all.


Zen 3 is compatible with AM4 socket. This alone tell you the core count will stagnant becz its already impressive current platform with Dual DDR channel can support 16C/32T.


Just look at the TR platform socket and you will see how many pins is need to support 24C and above.
"It's impossible that we'll see 12 and 16 cores on AM4, 100% more cores power delivery ree!"- You, probably.

Dude, it's already been done, It'll happen again. Each core will use 10-15% less power, and the 3900 and 3900X already use way less power than what naysayers like you said they'd have to.3900X: Uses significantly less power than the 9900K, and yet most Z390 boards have significantly worse power delivery than an X570.Any X570 will be able to cruise with a 24-core part.

Bandwidth limitations on TR4? It's already been debunked. Try running the 32-core part in Windows Vs. Linux. It was originally automatically assumed that the issue was bandwidth, but we now know it's a windows scheduler issue, and has nothing to do with bandwidth. Not only has the Bandwidth argument been completely debunked, we are now seeing DDR4 modules that have been designed specifically for Zen 2 that run at 5000Mhz out of the box.

https://level1techs.com/article/unlo...uma-aware-apps

Do you think Zen 3 will somehow regress it's memory controller because you have decided/hoped it would? No. It's totally possible.

MARS
(18 items)
CPU
Ryzen 9 3900x
Motherboard
Gigabyte X570 Aorus Master
GPU
EVGA 1080 Ti FTW3
RAM
G-Skill Sniper
Hard Drive
WD EMAZ001
Hard Drive
WD Black SN750
Power Supply
EVGA G3
Cooling
Noctua U12A
Case
NZXT H500
Operating System
Microsoft Windows 10
Operating System
Pop! OS
Monitor
BenQ XL2730z
Keyboard
Aukey Mechanical
Mouse
Mionix Castor
Mousepad
Aukey
Audio
PreSonus Eris E3.5
Audio
SennHeiser 598
Audio
Blue Yeti
▲ hide details ▲
Buris is offline  
post #19 of 67 (permalink) Old 10-13-2019, 08:15 AM - Thread Starter
New to Overclock.net
 
guttheslayer's Avatar
 
Join Date: Apr 2015
Posts: 3,805
Rep: 111 (Unique: 65)
Quote: Originally Posted by ozlay View Post
ZEN 4 on AM5 "more pins" with PCIe5.0 and DDR5. SMT4 and 5nm?
It very possible Zen 4 will see the biggest leap on all front, considering 2.5D stack, more cores, PCI5.0, DDR5 and SMT4 (if it happen). There might even be a chance of DDR tri-channel?


The platform could also be forward compatible for next 5 years to come given all its latest connection standard

guttheslayer is offline  
post #20 of 67 (permalink) Old 10-13-2019, 08:18 AM - Thread Starter
New to Overclock.net
 
guttheslayer's Avatar
 
Join Date: Apr 2015
Posts: 3,805
Rep: 111 (Unique: 65)
Quote: Originally Posted by Buris View Post
"It's impossible that we'll see 12 and 16 cores on AM4, 100% more cores power delivery ree!"- You, probably.

Dude, it's already been done, It'll happen again. Each core will use 10-15% less power, and the 3900 and 3900X already use way less power than what naysayers like you said they'd have to.3900X: Uses significantly less power than the 9900K, and yet most Z390 boards have significantly worse power delivery than an X570.Any X570 will be able to cruise with a 24-core part.

Bandwidth limitations on TR4? It's already been debunked. Try running the 32-core part in Windows Vs. Linux. It was originally automatically assumed that the issue was bandwidth, but we now know it's a windows scheduler issue, and has nothing to do with bandwidth. Not only has the Bandwidth argument been completely debunked, we are now seeing DDR4 modules that have been designed specifically for Zen 2 that run at 5000Mhz out of the box.

https://level1techs.com/article/unlo...uma-aware-apps

Do you think Zen 3 will somehow regress it's memory controller because you have decided/hoped it would? No. It's totally possible.
Dude! Wake up.

This is downright fantasy. AMD already confirm Milan will feature 64C which is same as current Rome CPU. So it is expected to have no increase from to mainsteam. TR4 might see 64C due to same socket with EYPC, other than that, please stop dreaming oh my.


And I am so sorry, I was the one who predicted AMD could go to 16C/32T, and I bash those who disagree with me. Please go through my post history and check before you make such baseless judgement. Thanks.


Last edited by guttheslayer; 10-13-2019 at 08:27 AM.
guttheslayer is offline  
Reply

Quick Reply
Message:
Options

Register Now

In order to be able to post messages on the Overclock.net - An Overclocking Community forums, you must first register.
Please enter your desired user name, your email address and other required details in the form below.
User Name:
If you do not want to register, fill this field only and the name will be used as user name for your post.
Password
Please enter a password for your user account. Note that passwords are case-sensitive.
Password:
Confirm Password:
Email Address
Please enter a valid email address for yourself.
Email Address:

Log-in



Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
 
Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may post new threads
You may post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off