Register a SA Forums Account here!
JOINING THE SA FORUMS WILL REMOVE THIS BIG AD, THE ANNOYING UNDERLINED ADS, AND STUPID INTERSTITIAL ADS!!!

You can: log in, read the tech support FAQ, or request your lost password. This dumb message (and those ads) will appear on every screen until you register! Get rid of this crap by registering your own SA Forums Account and joining roughly 150,000 Goons, for the one-time price of $9.95! We charge money because it costs us money per month for bills, and since we don't believe in showing ads to our users, we try to make the money back through forum registrations.
 
  • Post
  • Reply
Combat Pretzel
Jun 23, 2004

No, seriously... what kurds?!
So that 12nm spin is going to be OG Zen, not anything architecturally improved, other than erratas?

Adbot
ADBOT LOVES YOU

Watermelon Daiquiri
Jul 10, 2010
I TRIED TO BAIT THE TXPOL THREAD WITH THE WORLD'S WORST POSSIBLE TAKE AND ALL I GOT WAS THIS STUPID AVATAR.
Yeah, semi-accurate is 'legit' insofar as its been around and tried its best; it has set itself up as an 'industry' news site, hence the price.

Malcolm XML posted:

hell yeah bois up uP UP!!!!


nodes are marketing names at this point this is very likely "10nm" FEOL with "14nm" BEOL or somesuch like tsmc 16nm used older BEOL tech

Heck intel is basically hosed with 10nm, 14+++ is gonna be better than 10 for everything but extreme density/power needs (apparently SemiAccurate has the scoop on whats wrong but i am not spending 1k on that)

And I was just about to make this point lol. It's almost certainly that either they decided the process improvements were advanced enough for 14+++++++++ or whatever to merit a new marketing name, they decided to implement triple/quad patterning for FE and kept double for BE 1/2, or both.

wargames
Mar 16, 2008

official yospos cat censor
12nm isn't even on the list of technology nodes https://en.wikichip.org/wiki/WikiChip

EoRaptor
Sep 13, 2003



Combat Pretzel posted:

So that 12nm spin is going to be OG Zen, not anything architecturally improved, other than erratas?

It depends on whether the gate assembly is different that the current 'node'. It's no longer just shrinking the size of transistors, but changing how they are assembled and connected. For instance, a change from 14nm LPP to 12nm LPU would involve a complete redesign of the chip layout.

It's also a question of if this new node/process is even going to be used by AMD, and which part of AMD (GPU vs CPU) would even choose to adopt it.

Watermelon Daiquiri
Jul 10, 2010
I TRIED TO BAIT THE TXPOL THREAD WITH THE WORLD'S WORST POSSIBLE TAKE AND ALL I GOT WAS THIS STUPID AVATAR.
I believe that link explicitly says that AMD is using it for both ryzen and vega, and thats the entire point of the presentation the link is reporting on

eames
May 9, 2009

https://www.computerbase.de/2017-09/globalfoundries-12lp/

12LP stands for "leading performance" and is a optimized version of 14nm with 15% higher density and "10% higher performance".

GF still claims that they're skipping 10nm by going from 14 to 7nm because 12LP is technically 14nm.

Risk production for both 7nm and 12LP starts in the 1H/2018 so the author doesn't expect products before Q4/2018.

This article makes it sound like GF is having similar troubles with 7nm as Intel with their 10nm. :shrug:

Watermelon Daiquiri
Jul 10, 2010
I TRIED TO BAIT THE TXPOL THREAD WITH THE WORLD'S WORST POSSIBLE TAKE AND ALL I GOT WAS THIS STUPID AVATAR.
Well, yeah, both intel-10 and gf-7 are banking on EUV working out, which so far it isnt doing so hot

wargames
Mar 16, 2008

official yospos cat censor

Watermelon Daiquiri posted:

Well, yeah, both intel-10 and gf-7 are banking on EUV working out, which so far it isnt doing so hot

glofo 7nm isn't just them its part of that whole glofo/ibm/someother dudes all going yeah gently caress intel let's make 7nm happen.

Combat Pretzel
Jun 23, 2004

No, seriously... what kurds?!
If it's Q4/18, it better be Zen 2.

Bloody Antlers
Mar 27, 2010

by Jeffrey of YOSPOS
Has AMD signaled whether they plan on eventually providing current generation Ryzen-based products with support for faster than DDR4 3200 RAM via an AGESA update or motherboard chipset revision? Or is there a known limiting factor that definitely puts a ceiling on this generation at DDR4 3200, such as Infinity Fabric losing its poo poo if the CAS latency of the RAM is too high? I don't think DDR4 3200 is lacking or anything, it's just that after seeing the performance increases in benchmarks when going from DDR4 2800 to DDR4 3200, I can't help but wonder how Ryzen would do at even higher memory clocks.

Going up from CAS 14 to CAS 16 you can find kits of DDR4 3600 16-16-16-36 at the same voltage. Go from CAS 14 to 18, and you find DDR4 4000 18-19-19-39. The highest clocked DDR4 I see on newegg atm is DDR4 4400 CAS 19 timings 19-19-19-39.

I'm hoping the memory compatibility issues that we've seen patched in bios updates are just the result of AMD not having the time to roll out bios profiles for the variety of assorted DRAM chips on the market yet. If Ryzen can benefit from additional memory bandwidth at the expense of additional latency, well.. gently caress that would be sweet.

EmpyreanFlux
Mar 1, 2013

The AUDACITY! The IMPUDENCE! The unabated NERVE!

eames posted:

https://www.computerbase.de/2017-09/globalfoundries-12lp/

12LP stands for "leading performance" and is a optimized version of 14nm with 15% higher density and "10% higher performance".

GF still claims that they're skipping 10nm by going from 14 to 7nm because 12LP is technically 14nm.

Risk production for both 7nm and 12LP starts in the 1H/2018 so the author doesn't expect products before Q4/2018.

This article makes it sound like GF is having similar troubles with 7nm as Intel with their 10nm. :shrug:

Eh, it could also mean AMD is dropping 7nm DUV because EUV is expected soon enough that it doesn't make sense to do a 7nm DUV and EUV Zen2. It could also mean there would be some degree of segmentation in what exactly is getting 12nm LP and 7nm DUV as well, so 12nm desktop products but 7nm mobile and server because GloFo simply doesn't have the capacity to do a blanket 7nm production for all AMDs products in 2018.

That's very different from having process problems.

Prescription Combs
Apr 20, 2005
   6

Bloody Antlers posted:

Has AMD signaled whether they plan on eventually providing current generation Ryzen-based products with support for faster than DDR4 3200 RAM via an AGESA update or motherboard chipset revision? Or is there a known limiting factor that definitely puts a ceiling on this generation at DDR4 3200, such as Infinity Fabric losing its poo poo if the CAS latency of the RAM is too high? I don't think DDR4 3200 is lacking or anything, it's just that after seeing the performance increases in benchmarks when going from DDR4 2800 to DDR4 3200, I can't help but wonder how Ryzen would do at even higher memory clocks.

Going up from CAS 14 to CAS 16 you can find kits of DDR4 3600 16-16-16-36 at the same voltage. Go from CAS 14 to 18, and you find DDR4 4000 18-19-19-39. The highest clocked DDR4 I see on newegg atm is DDR4 4400 CAS 19 timings 19-19-19-39.

I'm hoping the memory compatibility issues that we've seen patched in bios updates are just the result of AMD not having the time to roll out bios profiles for the variety of assorted DRAM chips on the market yet. If Ryzen can benefit from additional memory bandwidth at the expense of additional latency, well.. gently caress that would be sweet.

Recent bios updates got my non samsung b-die g.skill ram to run at 3200 w/out any issues.

Risky Bisquick
Jan 18, 2008

PLEASE LET ME WRITE YOUR VICTIM IMPACT STATEMENT SO I CAN FURTHER DEMONSTRATE THE CALAMITY THAT IS OUR JUSTICE SYSTEM.



Buglord

FaustianQ posted:

Eh, it could also mean AMD is dropping 7nm DUV because EUV is expected soon enough that it doesn't make sense to do a 7nm DUV and EUV Zen2. It could also mean there would be some degree of segmentation in what exactly is getting 12nm LP and 7nm DUV as well, so 12nm desktop products but 7nm mobile and server because GloFo simply doesn't have the capacity to do a blanket 7nm production for all AMDs products in 2018.

That's very different from having process problems.

e: can't read, you are right though. AMD demand strong

Malcolm XML
Aug 8, 2009

I always knew it would end like this.

Watermelon Daiquiri posted:

Well, yeah, both intel-10 and gf-7 are banking on EUV working out, which so far it isnt doing so hot

p sure intel 10 is still multi patterning only

Malcolm XML
Aug 8, 2009

I always knew it would end like this.
also we're getting into nodes that need multi-patterned EUV which is gonna be hella fun

EmpyreanFlux
Mar 1, 2013

The AUDACITY! The IMPUDENCE! The unabated NERVE!

eames posted:

https://www.computerbase.de/2017-09/globalfoundries-12lp/

12LP stands for "leading performance" and is a optimized version of 14nm with 15% higher density and "10% higher performance".

Also don't forget this part is versus 16nm TSMC, not 14nm LPP. I thought 14nm LPP was already denser than 16nmFF+, so uh, is this just a rebranding of a process?

Watermelon Daiquiri
Jul 10, 2010
I TRIED TO BAIT THE TXPOL THREAD WITH THE WORLD'S WORST POSSIBLE TAKE AND ALL I GOT WAS THIS STUPID AVATAR.

wargames posted:

glofo 7nm isn't just them its part of that whole glofo/ibm/someother dudes all going yeah gently caress intel let's make 7nm happen.

Really? I thought that was just for '14', and they went on their separate ways (granted, with the lessons learned from 14)

Malcolm XML posted:

p sure intel 10 is still multi patterning only

It's very really a possibility I constructed a fake memory based on samsungs and everyone else's roadmaps of 'multi patterning sub-14 then euv sub-14+'

wargames
Mar 16, 2008

official yospos cat censor

Watermelon Daiquiri posted:

Really? I thought that was just for '14', and they went on their separate ways (granted, with the lessons learned from 14)


It's very really a possibility I constructed a fake memory based on samsungs and everyone else's roadmaps of 'multi patterning sub-14 then euv sub-14+'

https://arstechnica.com/gadgets/2015/07/ibm-unveils-industrys-first-7nm-chip-moving-beyond-silicon/

from ars

quote:

IBM, working with GlobalFoundries, Samsung, SUNY, and various equipment suppliers, has produced the world's first 7nm chip with functional transistors

New Zealand can eat me
Aug 29, 2008

:matters:


Corsair has finally confirmed that they will implement support for the 20C offset in Corsair Link soon

Sneeze Party
Apr 26, 2002

These are, by far, the most brilliant photographs that I have ever seen, and you are a GOD AMONG MEN.
Toilet Rascal
Just built my first Ryzen setup. I went with the MSI B350M "Pro Gaming" board, some Corsair RGB DDR4 (it's all Microcenter had in stock when I got there as far as 3000mhz+ 16gb kits), and the 1600x, with my first ever M2. I had purchased a 1070 a few months ago.

The RAM I have running at 2966, but I think it'll hit 3200 if I play around with it some more. I'm thinking about OC'ing the 1600x... but it seems like, if I do that, then the processor runs at 100% power all the time? Is that functionality baked into the AM4 platform, or am I missing something?

This is a step up from my 2500k at 4.2ghz. I decided to take the plunge even though benchmarks didn't show that much of a speed improvement... but it's definitely faster. My fps in BF1 (and everything else) are way more consistently high now than they used to be.

Sneeze Party fucked around with this message at 21:09 on Sep 24, 2017

SamDabbers
May 26, 2003



What you want is P-state overclocking, which leaves all the power management and idle-throttling capabilities enabled, but lets you tweak the top-end speed and voltage. It should be under an "AMD CBS" submenu in your bios, and there are easy to find guides on various sites.

BF1 is actually pretty decently multithreaded and can take advantage of Ryzen's ample cores. Enjoy your buttery smooth multitasking too!

Edit; not apropos of the above: I had to RMA my 1700 for the "performance marginality" segfault problem. AMD's RMA department took a few days to start the process once I opened the support ticket. They had me test a few times with various BIOS settings and send them a picture of the inside of my case to make sure I didn't put it together incorrectly, and then issued a prepaid FedEx label. The new chip arrived about a week after I shipped out the bad one, and it's a good sample which goes to 3.8GHz at stock voltage and doesn't segfault under heavy load. Thanks AMD! Ryzen satisfies, and so did AMD support.

SamDabbers fucked around with this message at 21:26 on Sep 24, 2017

Sneeze Party
Apr 26, 2002

These are, by far, the most brilliant photographs that I have ever seen, and you are a GOD AMONG MEN.
Toilet Rascal
What I've found so far is that MSI B350 motherboards don't support p-state overclocking. Bummer.

EmpyreanFlux
Mar 1, 2013

The AUDACITY! The IMPUDENCE! The unabated NERVE!

SamDabbers posted:

Edit; not apropos of the above: I had to RMA my 1700 for the "performance marginality" segfault problem. AMD's RMA department took a few days to start the process once I opened the support ticket. They had me test a few times with various BIOS settings and send them a picture of the inside of my case to make sure I didn't put it together incorrectly, and then issued a prepaid FedEx label. The new chip arrived about a week after I shipped out the bad one, and it's a good sample which goes to 3.8GHz at stock voltage and doesn't segfault under heavy load. Thanks AMD! Ryzen satisfies, and so did AMD support.

:staredog: Holy poo poo, that's a top ten percenter.

bobfather
Sep 20, 2001

I will analyze your nervous system for beer money
Ryzen chips underclock and undervolt themselves automatically, even without p-states. Some motherboards support p-states in addition to the on-board power management features, but they're rare and the users of those boards don't report good results from using p-states.

If you overclock Ryzen, it will still step down when there's no work to do. You can see it happen in something like hwinfo.

kirtar
Sep 11, 2011

Strum in a harmonizing quartet
I want to cause a revolution

What can I do? My savage
nature is beyond wild

FaustianQ posted:

:staredog: Holy poo poo, that's a top ten percenter.

:catstare: I have to use 1.375V to hit 3.8 stable.

Scarecow
May 20, 2008

3200mhz RAM is literally the Devil. Literally.
Lipstick Apathy
So I got my Threadripper build finally finished (for now) and im running at:

4.1ghz 1.385v
3200Mhz 14-13-13-28-42
2 Sli 1080tis @ 2076Mhz core

measured from the Vcore pinout my Zenith board has and only hitting over 60c when I run prime95 ( cpu hits 310W :staredog: )

Also getting some sick mad 3Dmark scores in Firestrike like this one :
30K
https://www.3dmark.com/3dm/22326827

still need to work on getting Pstate clocking to work

Scarecow fucked around with this message at 01:25 on Sep 25, 2017

MaxxBot
Oct 6, 2003

you could have clapped

you should have clapped!!
That's a hell of a build and good overclocking numbers, will probably do something similar with Threadripper 2 and Big Volta sans SLI.

EmpyreanFlux
Mar 1, 2013

The AUDACITY! The IMPUDENCE! The unabated NERVE!
310W for 16 cores running @ 4.1GHhz seems...low?

Scarecow
May 20, 2008

3200mhz RAM is literally the Devil. Literally.
Lipstick Apathy

FaustianQ posted:

310W for 16 cores running @ 4.1GHhz seems...low?

I'm going off by what HWinfo says my cpu power draw is so :shrug:

EmpyreanFlux
Mar 1, 2013

The AUDACITY! The IMPUDENCE! The unabated NERVE!

Scarecow posted:

I'm going off by what HWinfo says my cpu power draw is so :shrug:

It's just that Skylake-X seems to draw way, way more.

Scarecow
May 20, 2008

3200mhz RAM is literally the Devil. Literally.
Lipstick Apathy

FaustianQ posted:

It's just that Skylake-X seems to draw way, way more.

It could be because intel includes vrm power draw while amd does not?
I'll look into it tonight after work

AARP LARPer
Feb 19, 2005

THE DARK SIDE OF SCIENCE BREEDS A WEAPON OF WAR

Buglord

I'm jealous! That thing looks like it should be in the Reanimator's lab; mine looks like a bird's nest.

Paul MaudDib
May 3, 2006

TEAM NVIDIA:
FORUM POLICE
https://zippy.gfycat.com/GraciousActiveCoral.webm

This is why we can't have nice things :sigh:

EmpyreanFlux
Mar 1, 2013

The AUDACITY! The IMPUDENCE! The unabated NERVE!
I'm loving dead holy poo poo.

GRINDCORE MEGGIDO
Feb 28, 1985


Hahahahaaa

Palladium
May 8, 2012

Very Good
✔️✔️✔️✔️
"PC building is easy" -Reddit

AARP LARPer
Feb 19, 2005

THE DARK SIDE OF SCIENCE BREEDS A WEAPON OF WAR

Buglord
lmao. it's probably mayonnaise too.

SwissArmyDruid
Feb 14, 2014

by sebmojo
What kills me is that the dead board is also an AMD board.

Oh well, socket AM3+, no big loss at this point.

3peat
May 6, 2010


Is it legal for them to sell it with no warranty or return? Over here everything sold by retailers must have a minimum of 2yrs warranty, and if you order online you get a 15 days window to return it without justification (both being EU regulations), which applies to open box items too. I thought America would have something similar.

Adbot
ADBOT LOVES YOU

underage at the vape shop
May 11, 2011

by Cyrano4747

3peat posted:

I thought America would have something similar.

america is garbage

  • 1
  • 2
  • 3
  • 4
  • 5
  • Post
  • Reply