• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

*** AMD "Zen" thread (inc AM4/APU discussion) ***

Soldato
Joined
26 May 2009
Posts
22,100
From the horse's mouth. No issues with Windows 10 SMT and Ryzen.

This is why I kept asking all those saying fixes and optimization was incoming to actually show me official word as I wanted to believe.

I like this bit:

We do not believe there is an issue with scheduling differences between the two versions of Windows. Any differences in performance can be more likely attributed to software architecture differences between these OSes.
Those two statements are in contrast, the difference in performance due to the software architecture is that Windows 10 (and 8) use CPU level drivers and 7 does not, which is why there is an issue with scheduling differences between the two (as 8/10 don't have a driver yet).

Bit worrying that AMD are letting a guy who apparently knows so little about their product make public statements lol.
 
Soldato
Joined
7 Feb 2015
Posts
2,864
Location
South West
Another problem with the PCper article, he talks about windows not recognising the threading, but it does and that was never the main issue. Windows 10 recognises there are 16 threads and how they are paired per core. What it is not recognising correctly for the majority of people is the cache configuration of the processor. instead it is setting it so each thread has its own L1, L2 and L3 cache.

This is how it is configuring the cores for people
Logical Processor to Cache Map:
*--------------- Data Cache 0, Level 1, 32 KB, Assoc 8, LineSize 64
*--------------- Instruction Cache 0, Level 1, 64 KB, Assoc 4, LineSize 64
*--------------- Unified Cache 0, Level 2, 512 KB, Assoc 8, LineSize 64
*--------------- Unified Cache 1, Level 3, 16 MB, Assoc 16, LineSize 64
-*-------------- Data Cache 1, Level 1, 32 KB, Assoc 8, LineSize 64
-*-------------- Instruction Cache 1, Level 1, 64 KB, Assoc 4, LineSize 64
-*-------------- Unified Cache 2, Level 2, 512 KB, Assoc 8, LineSize 64
-*-------------- Unified Cache 3, Level 3, 16 MB, Assoc 16, LineSize 64
--*------------- Data Cache 2, Level 1, 32 KB, Assoc 8, LineSize 64
--*------------- Instruction Cache 2, Level 1, 64 KB, Assoc 4, LineSize 64
--*------------- Unified Cache 4, Level 2, 512 KB, Assoc 8, LineSize 64
--*------------- Unified Cache 5, Level 3, 16 MB, Assoc 16, LineSize 64
---*------------ Data Cache 3, Level 1, 32 KB, Assoc 8, LineSize 64
---*------------ Instruction Cache 3, Level 1, 64 KB, Assoc 4, LineSize 64
---*------------ Unified Cache 6, Level 2, 512 KB, Assoc 8, LineSize 64
---*------------ Unified Cache 7, Level 3, 16 MB, Assoc 16, LineSize 64
----*----------- Data Cache 4, Level 1, 32 KB, Assoc 8, LineSize 64
----*----------- Instruction Cache 4, Level 1, 64 KB, Assoc 4, LineSize 64
----*----------- Unified Cache 8, Level 2, 512 KB, Assoc 8, LineSize 64
----*----------- Unified Cache 9, Level 3, 16 MB, Assoc 16, LineSize 64
-----*---------- Data Cache 5, Level 1, 32 KB, Assoc 8, LineSize 64
-----*---------- Instruction Cache 5, Level 1, 64 KB, Assoc 4, LineSize 64
-----*---------- Unified Cache 10, Level 2, 512 KB, Assoc 8, LineSize 64
-----*---------- Unified Cache 11, Level 3, 16 MB, Assoc 16, LineSize 64
------*--------- Data Cache 6, Level 1, 32 KB, Assoc 8, LineSize 64
------*--------- Instruction Cache 6, Level 1, 64 KB, Assoc 4, LineSize 64
------*--------- Unified Cache 12, Level 2, 512 KB, Assoc 8, LineSize 64
------*--------- Unified Cache 13, Level 3, 16 MB, Assoc 16, LineSize 64
-------*-------- Data Cache 7, Level 1, 32 KB, Assoc 8, LineSize 64
-------*-------- Instruction Cache 7, Level 1, 64 KB, Assoc 4, LineSize 64
-------*-------- Unified Cache 14, Level 2, 512 KB, Assoc 8, LineSize 64
-------*-------- Unified Cache 15, Level 3, 16 MB, Assoc 16, LineSize 64
--------*------- Data Cache 8, Level 1, 32 KB, Assoc 8, LineSize 64
--------*------- Instruction Cache 8, Level 1, 64 KB, Assoc 4, LineSize 64
--------*------- Unified Cache 16, Level 2, 512 KB, Assoc 8, LineSize 64
--------*------- Unified Cache 17, Level 3, 16 MB, Assoc 16, LineSize 64
---------*------ Data Cache 9, Level 1, 32 KB, Assoc 8, LineSize 64
---------*------ Instruction Cache 9, Level 1, 64 KB, Assoc 4, LineSize 64
---------*------ Unified Cache 18, Level 2, 512 KB, Assoc 8, LineSize 64
---------*------ Unified Cache 19, Level 3, 16 MB, Assoc 16, LineSize 64
----------*----- Data Cache 10, Level 1, 32 KB, Assoc 8, LineSize 64
----------*----- Instruction Cache 10, Level 1, 64 KB, Assoc 4, LineSize 64
----------*----- Unified Cache 20, Level 2, 512 KB, Assoc 8, LineSize 64
----------*----- Unified Cache 21, Level 3, 16 MB, Assoc 16, LineSize 64
-----------*---- Data Cache 11, Level 1, 32 KB, Assoc 8, LineSize 64
-----------*---- Instruction Cache 11, Level 1, 64 KB, Assoc 4, LineSize 64
-----------*---- Unified Cache 22, Level 2, 512 KB, Assoc 8, LineSize 64
-----------*---- Unified Cache 23, Level 3, 16 MB, Assoc 16, LineSize 64
------------*--- Data Cache 12, Level 1, 32 KB, Assoc 8, LineSize 64
------------*--- Instruction Cache 12, Level 1, 64 KB, Assoc 4, LineSize 64
------------*--- Unified Cache 24, Level 2, 512 KB, Assoc 8, LineSize 64
------------*--- Unified Cache 25, Level 3, 16 MB, Assoc 16, LineSize 64
-------------*-- Data Cache 13, Level 1, 32 KB, Assoc 8, LineSize 64
-------------*-- Instruction Cache 13, Level 1, 64 KB, Assoc 4, LineSize 64
-------------*-- Unified Cache 26, Level 2, 512 KB, Assoc 8, LineSize 64
-------------*-- Unified Cache 27, Level 3, 16 MB, Assoc 16, LineSize 64
--------------*- Data Cache 14, Level 1, 32 KB, Assoc 8, LineSize 64
--------------*- Instruction Cache 14, Level 1, 64 KB, Assoc 4, LineSize 64
--------------*- Unified Cache 28, Level 2, 512 KB, Assoc 8, LineSize 64
--------------*- Unified Cache 29, Level 3, 16 MB, Assoc 16, LineSize 64
---------------* Data Cache 15, Level 1, 32 KB, Assoc 8, LineSize 64
---------------* Instruction Cache 15, Level 1, 64 KB, Assoc 4, LineSize 64
---------------* Unified Cache 30, Level 2, 512 KB, Assoc 8, LineSize 64
---------------* Unified Cache 31, Level 3, 16 MB, Assoc 16, LineSize 64

This is how it should look roughly
Logical Processor to Cache Map:
**-------------- Data Cache 0, Level 1, 32 KB, Assoc 8, LineSize 64
**-------------- Instruction Cache 0, Level 1, 64 KB, Assoc 4, LineSize 64
**-------------- Unified Cache 0, Level 2, 512 KB, Assoc 8, LineSize 64
********-------- Unified Cache 1, Level 3, 8 MB, Assoc 16, LineSize 64
--**------------ Data Cache 1, Level 1, 32 KB, Assoc 8, LineSize 64
--**------------ Instruction Cache 1, Level 1, 64 KB, Assoc 4, LineSize 64
--**------------ Unified Cache 2, Level 2, 512 KB, Assoc 8, LineSize 64
----**---------- Data Cache 2, Level 1, 32 KB, Assoc 8, LineSize 64
----**---------- Instruction Cache 2, Level 1, 64 KB, Assoc 4, LineSize 64
----**---------- Unified Cache 3, Level 2, 512 KB, Assoc 8, LineSize 64
------**-------- Data Cache 3, Level 1, 32 KB, Assoc 8, LineSize 64
------**-------- Instruction Cache 3, Level 1, 64 KB, Assoc 4, LineSize 64
------**-------- Unified Cache 4, Level 2, 512 KB, Assoc 8, LineSize 64
--------**------ Data Cache 4, Level 1, 32 KB, Assoc 8, LineSize 64
--------**------ Instruction Cache 4, Level 1, 64 KB, Assoc 4, LineSize 64
--------**------ Unified Cache 5, Level 2, 512 KB, Assoc 8, LineSize 64
--------******** Unified Cache 5, Level 3, 8 MB, Assoc 16, LineSize 64
----------**---- Data Cache 5, Level 1, 32 KB, Assoc 8, LineSize 64
----------**---- Instruction Cache 6, Level 1, 64 KB, Assoc 4, LineSize 64
----------**---- Unified Cache 6, Level 2, 512 KB, Assoc 8, LineSize 64
------------**-- Data Cache 6, Level 1, 32 KB, Assoc 8, LineSize 64
------------**-- Instruction Cache 7, Level 1, 64 KB, Assoc 4, LineSize 64
------------**-- Unified Cache 7, Level 2, 512 KB, Assoc 8, LineSize 64
--------------** Data Cache 7, Level 1, 32 KB, Assoc 8, LineSize 64
--------------** Instruction Cache 8, Level 1, 64 KB, Assoc 4, LineSize 64
--------------** Unified Cache 8, Level 2, 512 KB, Assoc 8, LineSize 64
 
Last edited:
Associate
Joined
31 Dec 2010
Posts
2,427
Location
Sussex
I said that after all the updates it was closer to the GTX780 than the GTX680, that doesn't mean that it matches the GTX780.
Indeed, but it got rather close. Hard to find a comparison now, but this TPU review from Jan 2016 of an R9 390 has the R9 280X and the GTX780 almost the same:
yQOZlnS.png

Back to Ryzen, if the Ryzen 3 1300 and 1100 for example are only going to have speeds of 3.2/3.5 then from the test The Stilt did, they should a lot closer to the peak efficiency of the process. So their actual consumption should be lot lower than the rumoured 65W.

Well, unless of course they are the dregs of the dies and truly are what failed binning for anything higher and leak like mad.

But really, what the process really wants is mobile APUs running under 3GHz rather than desktop parts which need to compete at 4GHz or more.
 
Soldato
Joined
13 Mar 2008
Posts
9,638
Location
Ireland
From the horse's mouth. No issues with Windows 10 SMT and Ryzen.

This is why I kept asking all those saying fixes and optimization was incoming to actually show me official word as I wanted to believe.

It wasn't really about SMT though. When PCper investigated that, they ended up finding the real culprit, cross CCX latency because the Windows Scheduler in 10 kept moving work between them, even when it's not supposed to.

Video with their conclusion.
https://youtu.be/6laL-_hiAK0?t=26m23s

5d6bf457c6084db9bd19638c519f79c2.png


Other's have reported that before. Hardware.fr disabled one CCX, and ended up with a 20% performance increase in Battlefield 1, and around 3-8% in other games.


http://www.hardware.fr/articles/956-24/retour-sous-systeme-memoire-suite.html

57d46709a9eb4491aa4cd9a568a854e8.png



At the same time AMD'd rep stated this in their Update.
For the remaining outliers, AMD again sees multiple opportunities within the codebases of specific applications to improve how this software addresses the “Zen” architecture. We have already identified some simple changes that can improve a game’s understanding of the "Zen" core/cache topology, and we intend to provide a status update to the community when they are ready.
 
Soldato
Joined
18 Oct 2012
Posts
4,139
Location
Oxfordshire
I like this bit:


Those two statements are in contrast, the difference in performance due to the software architecture is that Windows 10 (and 8) use CPU level drivers and 7 does not, which is why there is an issue with scheduling differences between the two (as 8/10 don't have a driver yet).

Bit worrying that AMD are letting a guy who apparently knows so little about their product make public statements lol.

This +1. There has been evidence to show the controller is not working as it is in W7 with real time videos and clearly why there is a difference and yet it is now being stated officially different when the Reddit posts from a while back from AMD said otherwise?
 
Soldato
Joined
26 May 2009
Posts
22,100
The way I see it, in a dual Xeon/Opteron system the system knows that moving a thread from CPU 1 core 1 to CPU 1 core 2 is preferable to moving it to CPU 2 core 1. Ryzen needs that same logic applying to it's CCX.
 
Soldato
Joined
18 May 2010
Posts
22,302
Location
London
I was thinking about upgrading to a Ryzen. But having now reapplied my oc on my 3570k it's pretty rapid and I'm now probably gpu bound again as opposed to cpu bound. It's made a massive difference.

Ryzen is the only upgrade I'd consider from my 3570k. Doesn't feel right going to a 7700k which is just another 4 core CPU at the end of the day.

But gaming performance is not brilliant atm with Ryzen so will park it for the moment.
 

RSR

RSR

Soldato
Joined
17 Aug 2006
Posts
9,519
I'm still considering a upgrading from my 6850K, as its cheaper to buy board and CPU rather than a 6900K or 6950X. I'll probably pick one up at some point though.
 
Soldato
Joined
9 Nov 2009
Posts
24,769
Location
Planet Earth
I'm still considering a upgrading from my 6850K, as its cheaper to buy board and CPU rather than a 6900K or 6950X. I'll probably pick one up at some point though.

Or just wait until Skylake E is out?? By that time Ryzen will be in full swing,and Intel will be replacing BW-E,and prices of the Core i7 6900K will need to drop,as it won't have price or performance to justify the silly price!! :p
 

RSR

RSR

Soldato
Joined
17 Aug 2006
Posts
9,519
Or just wait until Skylake E is out?? By that time Ryzen will be in full swing,and Intel will be replacing BW-E,and prices of the Core i7 6900K will need to drop,as it won't have price or performance to justify the silly price!! :p

That's true but its new shiny shiny though :p and I've not been too impressed with X99.
 
Associate
Joined
8 Jul 2013
Posts
2,089
Location
Middle age travellers site
He is right regardless how you see it. Z270 is a dead socket, as is the X99.
The new Intel cpus coming, will need new motherboards in 5 (X299) and 8 months (Coffee Lake?).

At least AMD AM4 socket will be supported until 2020-2021 with new CPUs every year.
That is what he is talking about.

Unfortunately atm there is no sense on buying a 7700K or a 6900K. Maybe someone would argue about the 6800K which price is reasonable considering all other Intel offerings.
And given the shortage of the AMD motherboards, looks like I could have one by end of the week. I cannot not see myself been without PC until the end of the month. Except I am lucky and find motherboard elsewhere.

Also the above post by the AMD community whomever, is going to **** off many potential Ryzen buyers. Especially the "anecdotal evidence" regarding the Windows 7 better performance against Windows 10. If whole AMD cannot build a Windows 7 machine to run some game tests, then I lose faith completely to them as a company. Also there is no mentioning regarding Linux, which after the SMT topology kernel patch on 4.10 and 4.9.10 kernels, works as expected regarding the thread scheduling and performance.


Agree ...This is the main reason why my money is with AMD this time round and any systems i build for clients i will be trying to sway the AMD way this will be much easier once we have the 4c 6c line from them....

Intel have drip fed far too long keeping us on 4c with minimal increase in performance and with prices edging up year by year ...and this whole new socket every month... feels that way is a joke ...If Intel really had there way it would not surprise me if we had a separate Socket for every single CPU Intel makes..
Its companies like Intel that keep squeezing every little bit of cash out of us until we the buyers either go 2nd hand (but even then that would be pricey too) or just leave the hobby of pc's a go for cheaper alternatives like Arm, tablets ect

Thank the gods for AMD ...but the question is would AMD be same if the tables where turned ?

This applies to Nvidia too..

These companies do not seem to understand that they can quite easily be the death of themselves ..
 
Associate
Joined
8 Jul 2013
Posts
2,089
Location
Middle age travellers site
I like this bit:


Those two statements are in contrast, the difference in performance due to the software architecture is that Windows 10 (and 8) use CPU level drivers and 7 does not, which is why there is an issue with scheduling differences between the two (as 8/10 don't have a driver yet).

Bit worrying that AMD are letting a guy who apparently knows so little about their product make public statements lol.

Agree its a very strange statement
 
Soldato
Joined
4 Feb 2006
Posts
3,195

This video proves there is something wrong in some major games when all cores + SMT are enabled. Some of the latest games run noticeably better in the 4c4t mode.
Not sure why AMD dismissed this issue.

edit: conspiracy theory mode. They want to make their R5 Ryzen the main gaming chip so in their interest for the R7 to perform sub-par in games:eek:.
 
Soldato
Joined
28 May 2007
Posts
18,200
you have a strange definition of 'faster' circa 5Ghz oc vs circa 4.1Ghz oc is not 'slower' it may be a less parallel CPU and may be slower in some tasks but that's it.....

pretty sure you were debating me about 16 months ago when I said that slower hex core 5820k might be a better long term bet over the quad 6700k? Seem to have changed your tune now?

Pretty sure z270's going to be compatible with at least one more refresh of CPU's so a bit daft to label it a 'dead' socket right now...(z170 obviously doesn't look like it will be compatible with anything else)

Said it before but will say it again well done AMD, certainly exceeded my expectations re pricing if not performance (yet). Hope they can keep it up. Shame they appear to have launched a little bit prematurely... Still smooth launches have not been AMD's forte

You seem to have a strange definition of faster, but to be fair you do run around constantly telling people £360 is less than £200 so maybe that explains your trouble with the concept...

I'm not sure I've ever debated you TBH. Seem's futile. Maybe I've said to someone go for a 6700K as you can buy the system for less than 5820K or 7700K. But usually I'd say for with the 6 core over the quad if the price make sense.

Pretty sure Intel will cancel releasing another 6700K now. What's the point now Ryzen has released? and if they did who is going to buy it looking at the prices of the R5 quad cores.

It's OK you don't have to explain yourself to me. Buy whatever you CPU you want. Just don't fool yourself into thinking an i7 quad core is faster.
 
Associate
Joined
8 Jul 2013
Posts
2,089
Location
Middle age travellers site

This video proves there is something wrong in some major games when all cores + SMT are enabled. Some of the latest games run noticeably better in the 4c4t mode.
Not sure why AMD dismissed this issue.

edit: conspiracy theory mode. They want to make their R5 Ryzen the main gaming chip so in their interest for the R7 to perform sub-par in games:eek:.

Thx You make a good point re the R5 but isn't the fault with the CCX so i assume even the 6 core will still have the issue as it will still be using the CCX the r3 i imagine would not but that all depends on how what AMD disable the cores on them

All this is very strange it must be the swapping of the CCX and what is more strange is AMD stating that there is nothing wrong it just does not make sense..

So is it ..

1. Each current / future game's must be patched to run on Ryzen correctly

2. Windows Driver for Ryzen scheduler or optimizing or both

3. CCX is at fault and the above option's 1 & 2 will not fix this apart from the end user disabling a whole CCX to play games and for production use re-enable the CCX
 
Last edited:
Back
Top Bottom