The Ryzen "ThreadRipper"... 16 cores of awesome

Page 13 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

IEC

Elite Member
Super Moderator
Jun 10, 2004
14,362
5,033
136
Being 4 DIMMs this sould be fine! But you'll miss out on the dual rank performance gains, which are like 200-500 MHz higher RAM speed for several games and apps.

As if I needed an excuse to upgrade to 64GB! I'll look for a deal on Samsung B-die dual-rank. 3200 dual rank quad channel sounds like fun
 
Reactions: Drazick

ddogg

Golden Member
May 4, 2005
1,864
361
136
https://www.youtube.com/watch?v=TWFzWRoVNnE

Pretty decent video on Intel's upcoming x299 platform.
I think their 14-18 core chips have been delayed to 2018, so it seems it was just a knee jerk reaction to AMD threadripper announcement. Looks like AMD caught them with their pants down.
 

DeeJayBump

Member
Oct 9, 2008
60
63
91
https://www.youtube.com/watch?v=TWFzWRoVNnE

Pretty decent video on Intel's upcoming x299 platform.
I think their 14-18 core chips have been delayed to 2018, so it seems it was just a knee jerk reaction to AMD threadripper announcement. Looks like AMD caught them with their pants down.

1. Quite certain Intel had no intention of bringing the higher core counts of X299 to the desktop/HEDT consumer space AT ALL at least not for YEARS. So, IMO, it's not correct to say X299 has been delayed to 2018 onward, but probably more accurate to say it's being pulled FORWARD as a response to Ryzen, Threadripper and Epyc.

2. This is the AMD Threadripper thread, can we please keep the conversation primarily on Threadripper. There are plenty of other topics/threads/etc in these forums for Intel-centric discussion, IMO.
 

Soulkeeper

Diamond Member
Nov 23, 2001
6,713
142
106
Being 4 DIMMs this sould be fine! But you'll miss out on the dual rank performance gains, which are like 200-500 MHz higher RAM speed for several games and apps.


All those OC DIMMs are rated at JEDEC 2133 speeds or so, but come with an pretested headroom (the XMP setting). Why shouldn't ECC RAM have some headroom too, just not being marketed this way?

I think there are some faster registered kits out there.
I guess you're righ
Being 4 DIMMs this sould be fine! But you'll miss out on the dual rank performance gains, which are like 200-500 MHz higher RAM speed for several games and apps.


All those OC DIMMs are rated at JEDEC 2133 speeds or so, but come with an pretested headroom (the XMP setting). Why shouldn't ECC RAM have some headroom too, just not being marketed this way?

Yeah, that seems to be the case.
Not many competing products in the ecc unbuffered space either.
If there was a 3200 ecc kit I would have bought that.
 

Soulkeeper

Diamond Member
Nov 23, 2001
6,713
142
106
If you actually needed ECC, you wouldn't hesitate for a second, and it would be painless.

good point. I'm just trying to do unusual things for odd reasons I guess.
But the same applies to anyone with 64GB of mem on ryzen too, not just ecc, There is a tradeoff. Maybe i'm just feeling left out of the overclocking fun you other guys are having

Maybe I should get a fast 2x8 kit for this system and put the ecc stuff in a threadripper system.
 

SolMiester

Diamond Member
Dec 19, 2004
5,331
17
76
I would really love to upgrade my back office host to Ryzen, but with mainboards only supporting 6 sata ports and no raid 5, looks like i will have to stay with Intel
 

imported_jjj

Senior member
Feb 14, 2009
660
430
136
good point. I'm just trying to do unusual things for odd reasons I guess.
But the same applies to anyone with 64GB of mem on ryzen too, not just ecc, There is a tradeoff. Maybe i'm just feeling left out of the overclocking fun you other guys are having

Maybe I should get a fast 2x8 kit for this system and put the ecc stuff in a threadripper system.

That's not all that true. You can do 64GB at 3200MHz and you have better perf with dual rank kits than single rank.
2x16GB or 4x16GB should be in the same ballpark, perf wise with 2x8GB at 3600 -if you can get 3600 stable for 24/7 use.
The first link i could recall for some tests for single rank vs dual rank( some other german site and Hardware FR did some tests too a few months back) https://www.golem.de/news/ram-overc...-von-ddr4-3200-und-dual-rank-1704-127262.html
The perf gains do diminish at high memory clocks so even with 2x8GB you might be better off at 3200 and tight timings.
 
Last edited:

DisEnchantment

Golden Member
Mar 3, 2017
1,687
6,243
136
I am looking forward to this platform. I spoke to our IT guys about it and they seem interested.
Our IT got me HP Z840 with Intel E5 dual socket, but I think Thread Ripper, if it support ECC would be great.

I also have several i5 Machines around for tracing and logging. But for me i5s and i7s are not suitable for work.

My personal 1700X 3950MHz OC compares favorably to my official Dual Xeon E5-2609v4 in GHS compile time, (possibly due to higher clocks). I cannot produce benchmarks on my work machine, but I got the code home and tried compiling on my 1700X with VPN License server and I was not disappointed with my R7.
GHS Compiler is very well threaded (better threaded than gcc if you ask me). If you have 16 Cores the time is slashed to almost 1/4 compared to when you have 4 cores for example. In fact you have to make sure to use 14 cores when you have 16 otherwise it will freeze your machine till the compilation is over.
Also browsing git on repos with > 500,000 commits locks up the system, and that's when additional cores simply helps, the git client locks up a core trying to resolve commits and I can continue compile or do some thing else or browse another repo with millions of commits as long as I have extra free cores.

R7 does not disappoint.
The fact that I could match the HP Z840 Dual E5 with my €1,200 R7 machine is awesome.
Can't wait for R9.

I hope HP release some workstations based on this beast.
 
Last edited:

wildhorse2k

Member
May 12, 2017
180
83
71
All Ryzen cpu's has ECC capabilities. It's up to mobo makers to implement. There are a few that does and persons have verified it's working.

On which Ryzen motherboards does ECC work? Do we have evidence for this? I only found an article showing that ECC on Ryzen doesn't work properly (single bit errors get corrected but no halt due to uncorrectable errors).
 

wildhorse2k

Member
May 12, 2017
180
83
71

LOL have you actually read it? I see this:

This lack of settings severely hampers the overall ECC functionality, since a big part of it is that the motherboard should be able to log errors. Right now, no such logging capability exists.
...
If it does not, the hardware might be silently correcting single-bit errors and even detecting 'catastrophic' two-bit errors, but you will never know about it since there will be no log.
...
The first command (wmic memphysical get memoryerrorcorrection) essentially queries Windows as to whether it detects any form of memory error-correcting code functionality. Regrettably, 3 signifies "none". The other options could have been 2 (unknown), 4 (parity), 5 (single-bit ECC), or 6 (multi-bit ECC).
...
If the TotalWidth value is larger than the DataWidth value then ECC is enabled. Ideally, we would have liked to see 72 for the TotalWidth instead of 64, since that would have indicated that the OS detects a 72-bit memory bus width, which as we discussed in the first page is the norm for ECC memory. Once again, no such luck.
...
Just for the hell of it, we loaded up Windows Server 2016 in the slight hope that an operating system that was designed to be used on exclusively ECC-enabled hardware might produce different results. We tried all of the same Windows commands, as well as all the same applications, and they all reported the same information as Windows 10 Pro.
...
Based on just the above information, it would not be prudent to say that ECC is enabled on this platform
....
HOWEVER, things are not quite perfect. On that last line you will notice "1 UE". That is an uncorrected error (UE), otherwise known as a two-bit error or a hard error. Two-bit errors cannot be corrected by ECC memory. What is supposed to happen when they occur is that they should be detected, logged and ideally the system should be immediately halted. These are considered fatal errors and they can easily cause data corruption if the system is not quickly halted and/or rebooted. Regrettably, only 2 of the 3 steps happened. The hard error was detected and it was logged, but the system kept running. The only reason that it's the last line on that image is because we immediately took a screenshot just in case the system would halt, but that never happened.
....
In conclusion, what is currently available on the AM4 platform is an incomplete implementation of ECC.

That official BIOSTAR website proves nothing. What is interesting if somebody did investigate it on BIOSTAR and found it to be fully functional unlike on the X370 Taichi.
 

formulav8

Diamond Member
Sep 18, 2000
7,004
522
126
That was in March on a very early bios and he admitted lack of available settings at the time. And the soft single-bit errors were properly fixed as ECC is expected to do. Mobo makers are looking to beginning to implement in their newer bios. It would be nice if they rechecked in another article with the new bios's popping up this month.

Either way Ryzen cpu's DO support ECC. AMD will not validate a mainstream platform for a feature that would be rarely used and its up to the mobo makers to implement and validate. AMD does Not block it. I'd expect that to be a more prominent feature with the x399 platform. Especially since Intel will probably once again artificially block it.
 

wildhorse2k

Member
May 12, 2017
180
83
71
I expect ECC to be fully functional on Threadripper. It probably was never properly implemented on X370 as it was too time consuming and very few customers care about it on that platform. I doubt it will be widely supported there. We can consider ourselves lucky if one motherboard maker finishes it.

Personally I care more about high speed memory support than ECC.
 
Reactions: scannall

Soulkeeper

Diamond Member
Nov 23, 2001
6,713
142
106
That's not all that true. You can do 64GB at 3200MHz and you have better perf with dual rank kits than single rank.
2x16GB or 4x16GB should be in the same ballpark, perf wise with 2x8GB at 3600 -if you can get 3600 stable for 24/7 use.
The first link i could recall for some tests for single rank vs dual rank( some other german site and Hardware FR did some tests too a few months back) https://www.golem.de/news/ram-overc...-von-ddr4-3200-und-dual-rank-1704-127262.html
The perf gains do diminish at high memory clocks so even with 2x8GB you might be better off at 3200 and tight timings.

Thanks for that link. I heard there was a benefit, but I hadn't looked into it really. That makes me feel a bit better.
 

Crumpet

Senior member
Jan 15, 2017
745
539
96
As you're apparently not allowed to post anything involving AMD on the X299 thread (despite the Ryzen threads being absolutely full of Intel trolls getting away with murder)





Moderator callout. You have an issue with the moderation, make an MD thread.


esquared
Anandtech Forum Director
 
Last edited by a moderator:

itsmydamnation

Platinum Member
Feb 6, 2011
2,868
3,419
136
I would really love to upgrade my back office host to Ryzen, but with mainboards only supporting 6 sata ports and no raid 5, looks like i will have to stay with Intel
If you still use raid 5 you are doing life wrong.... Seriously raid6/z2 as minimum.
 
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/    |