AMD: Ram OC Article

Printed From: ASRock.com
Category: Technical Support
Forum Name: AMD Motherboards
Forum Description: Question about ASRock AMD motherboards
URL: https://forum.asrock.com/forum_posts.asp?TID=5588
Printed Date: 23 Dec 2024 at 11:14pm
Software Version: Web Wiz Forums 12.04 - http://www.webwizforums.com

Topic: AMD: Ram OC Article

Posted By: AlbinoRhino
Subject: AMD: Ram OC Article
Date Posted: 15 Jul 2017 at 12:48am

Here's an interesting read on when Single Rank is better than Dual Rank, when Gear Down Mode should/shouldn't be enabled, and Tighter lower timings vs looser higher timings:

https://community.amd.com/community/gaming/blog/2017/07/14/memory-oc-showdown-frequency-vs-memory-timings" rel="nofollow - https://community.amd.com/community/gaming/blog/2017/07/14/memory-oc-showdown-frequency-vs-memory-timings

Replies:

Posted By: wardog
Date Posted: 15 Jul 2017 at 2:35am

" rel="nofollow - Finally. What I have been preaching here.

The ancient debate: frequency or timings?

Last, but not least, Sami set out to find whether it was tighter timings or higher clockspeeds that mattered most on the AMD Ryzen??processor. Sami pushed this combination of hardware up to DDR4-3520, DDR4-3466 with tighter timings, and DDR4-3200 with the tightest timings that could be achieved while maintaining stability with Memtest.

The verdict: tighter timings won. DDR4-3200 with aggressive timing adjustments outperformed the looser timings needed to hit DDR4-3520, while 3466 clearly split the difference with the right balance of timings and frequency.

...................

............

Conclusion #3: Chasing the highest possible clockspeed required timings so relaxed that real world performance suffered versus lower frequencies with tighter timings. This is a fine balance, however, so testing on your platform is always helpful.

Posted By: AlbinoRhino
Date Posted: 15 Jul 2017 at 3:01am

" rel="nofollow -

wardog wrote:

[URL=][/URL]Finally. What I have been preaching here.

<h1>The ancient debate: frequency or timings?</h1><p style="min-height: 8pt; padding: 0px;">?

Last,
but not least, Sami set out to find whether it was tighter timings or
higher clockspeeds that mattered most on the AMD Ryzen??processor. Sami
pushed this combination of hardware up to DDR4-3520, DDR4-3466 with
tighter timings, and DDR4-3200 with the tightest timings that could be
achieved while maintaining stability with Memtest.

<p style="min-height: 8pt; padding: 0px;">?

The
verdict: tighter timings won. DDR4-3200 with aggressive timing
adjustments outperformed the looser timings needed to hit DDR4-3520,
while 3466 clearly split the difference with the right balance of
timings and frequency.

...................

............

Conclusion #3: Chasing the highest possible clockspeed
required timings so relaxed that real world performance suffered versus
lower frequencies with tighter timings. This is a fine balance, however,
so testing on your platform is always helpful.

This is why I won't compromise my timings at higher clock speeds. I'd rather have tighter timings at a lower clock than loose timings at a higher clock. Now if I could find some literature on PROC Ohms I'd be happy. How high is reasonable on Ram Ohms?

Posted By: wardog
Date Posted: 15 Jul 2017 at 5:08am

" rel="nofollow - There will be no "reasonable' value.

procODT is derived from many variables(mem to CPU traces, CPU, Memory, "noise' levels over traces, voltages etc) and thus EACH Users computer will be different.

Consider procODT to be a tolerance level, in laymans terms.

Two old skool, and now outdated with the settings/options, for reference ONLY.

http://www.bit-tech.net/reviews/tech/memory/the_secrets_of_pc_memory_part_3/8/

http://archive.techarp.com/showFreeBOG9847.html?lang=0

Those two pages have great background/info on procODT. But again, since both are rather dated now, the #'s themselves are ....... not up to date. Fortunately, their desc of the function IS applicable to today.

HTH

Posted By: Drain_Bamaged
Date Posted: 15 Jul 2017 at 5:19am

A adjustment to procODT was absolutely necessary for my RAM.

Posted By: AlbinoRhino
Date Posted: 15 Jul 2017 at 9:05am

Drain_Bamaged wrote:

<span style=": rgb43, 43, 43;">A adjustment to procODT was absolutely necessary for my RAM.</span>

Yeah I'm at 53.3 Ohms@3066MHz, but am looking for a ballpark Ohm for 3200MHz.

I have no idea how high you can go with it still being considered safe. I tried 60 Ohms for 3200, but haven't tried higher.

At 2933 I was at 48 Ohms.