Mid-Life Crisis

[ArchStanton]

I started a thread over at overclock.net before Avocado roped me in here (My middle-aged overclocking adventure | Overclock.net).  I'll probably make duplicate updates to both threads going forward.

 

RL delayed my plans for slowly easing into a stable overclock, but I think I'm ready to proceed now.  I was able to install an external temperature sensor.  Now I will have ambient temperature data as a point of reference during my tweaking.  I also moved the system to its new home.  I hope to get some baseline benchmarks done tonight. 

 

Edited January 19, 2022 by ArchStanton

[Avacado]

Setup is so clean. Looking forward to watching this OC journey!

[ArchStanton]

11 minutes ago, Avacado said:

Setup is so clean.

Thank you.  Cable management for two separate desktops and a 5.1 channel audio system brought me to my knees (literally and figuratively), but worth it in the end (to me).  The center/surround speakers are just included because I already had them, will likely spend 99% of its time in 2.1 channel mode.

[ArchStanton]

I'll try to describe the baseline test conditions.  DOCP off.  PBO off. Pretty much everything in the BIOS that looked like it might boost performance off.  The following background processes were running -> Samsung magician, Steam, Fan Control V101, Radeon Software, HWiNFO64 (sensors only).

 

 

[pio]

You should head over to the CPUz bench competition we're doing for fun.

Looks like a fantastic rig and setup!  Your overall cable management makes me jealous.  My "mid life crisis" seemed to manifest in me buying a bunch of old junk.  You went the opposite route.  Touche.  Any plans on going bigger on the monitors?  That seems to be about the only thing really "lacking" in your setup there, and even then that's just a personal preference anyway.  10/10 rig there, decent to great audio, great keyboard, great cable management, I mean you're really not leaving us much to critique here.

[ArchStanton]

48 minutes ago, pioneerisloud said:

Any plans on going bigger on the monitors? 

 Not for the time being.  The M28U is bigger than anything I've used in the past.  Still feels like "vast tracts of land" from my perspective. 

 

The smaller monitor is just a display for the old system under the left side of the table.  Keeping it saved me the hassle of moving all my family's files over to my fresh drives.  Keeping it accessible to them also saves me the aggravation of worrying about what they might inadvertently install on my "toy" when I'm not home.  As they are not what I would call "computer literate", and my toy stays locked when I'm not around to play with it. 

[ArchStanton]

Round 2

 

Same conditions as before except DOCP is now enabled.  I honestly expect DOCP to have the single largest performance impact of any of the options available in the BIOS, but I've yet to create my spreadsheet for tracking % increases from one iteration to the next.

 

[ArchStanton]

These initial tests have shown me that I don't actually have a good understanding of what DOCP is.  I had noticed that the CPU was limited to 3400mhz in the baseline test, but I didn't know that was in anyway tied to DOCP, which I had erroneously assumed was just an AMD version of XMP (I expected DOCP to only affect DRAM/fclk/etc rather than the maximum frequency of the CPU).

 

A summary of the changes in performance.  I'll add stability testing at a later stage in the process (likely after turning on the "base" PBO without curve optimizer).

 

Edited January 20, 2022 by ArchStanton

[Avacado]

5 hours ago, ArchStanton said:

These initial tests have shown me that I don't actually have a good understanding of what DOCP is.  I had noticed that the CPU was limited to 3400mhz in the baseline test, but I didn't know that was in anyway tied to DOCP, which I had erroneously assumed was just an AMD version of XMP (I expected DOCP to only affect DRAM/fclk/etc rather than the maximum frequency of the CPU).

 

A summary of the changes in performance.  I'll add stability testing at a later stage in the process (likely after turning on the "base" PBO without curve optimizer.

 

Pretty significant jump. I don't know anything about DOCP either. I had Threadrippers for a very short time till they were all sold. 

[J7SC_Orion]

Kudos for your clean methodical approach to do one step at a time  . There's still extra performance to be had with the 5950X / Asus X570 Dark Hero (I run of of those), but I am sure you will do more step-by-step updates, apart from pointing out the 'DynamicOC' feature on that mobo / CPU combo. It doesn't require sub-zero cooling, btw

 

 

[ArchStanton]

1 minute ago, J7SC_Orion said:

 pointing out the 'DynamicOC' feature on that mobo / CPU combo

Aye, DynamicOC is the reason I landed on the Dark Hero.  For the money, I think it should have included a heat spreading backplate (with thermal pads where appropriate), but it's a seller's market in just about everything ATM.  

[J7SC_Orion]

14 minutes ago, ArchStanton said:

Aye, DynamicOC is the reason I landed on the Dark Hero.  For the money, I think it should have included a heat spreading backplate (with thermal pads where appropriate), but it's a seller's market in just about everything ATM.  

 

...probably, though it seems sturdy and does run quite cool - cooler than the 'regular CH8 Hero Wifi right next to it (pic in my sig) with a similar cooling arrangement. Anyway, I really like the Dark Hero; it's a fun board to push, especially w/Dynamic OC with which to get top multi-core and single-core results at the same time with just a few simple steps.

[GanjaSMK]

Commenting on your picture in OP, I'm both jealous and inspired by your whole setup.  Speakers look pretty impressive, what are they? And, how loud have you pumped it yet?

[ArchStanton]

16 minutes ago, GanjaSMK said:

Speakers look pretty impressive, what are they? And, how loud have you pumped it yet?

The floor standing units are DCM Time Window 1A's.  I have an uncle who was a bit of an audiophile in his youth.  He gave them to me when I was 16 or 17 years of age.  They were the pinnacle of what could be bought in the mid 1970's.  I've had them with me for about 25 years now.  I'm sure there are many superior products on the market today, but these still win the "bang for the buck" trophy with many serious audio equipment reviewers who are old enough to remember them.  They aren't, and never have been the loudest thing around, but if I were to try and describe their performance with a single word, it would be "clear".  They do need help from a subwoofer, and there's one hidden under the right side of the table.

 

As to how much I have pumped them?  All I can say for certain is "louder than the wife allows"...

[GanjaSMK]

Man I can relate! 

I don't have them any more but in the late 1980's my dad had a great business connection (friend) of his build custom speakers that were taller than I was when he bought 'em.  And man-oh-man let me tell you.... I wish I never let go of 'em.  I'm still rocking some floor pounder realistic's from that era that I found in near-perfect condition (cosmetic issues only) at a thrift shop.

 

But, they're in storage right now as I'm on the road.  Ah I miss great big, booming, non-sound-bar-driven audio!   

Nothing wrong with the modern setups, I'm just one for nostalgia.  What are the top speakers, or are those part of those big ol bad boys?

[ArchStanton]

16 hours ago, GanjaSMK said:

What are the top speakers, or are those part of those big ol bad boys?

They are the audio equivalent of the Titanic's fourth smokestack.  The center channel under the M28U monitor and the small surrounds on top of the TW's are just there so that I don't have to store them elsewhere.  They are some generic Polk Audio units I picked up at a Circuit City while I was in college. At the time, I was in need a new receiver, as the old Kenwood my uncle gave me along with the TW's bit the dust.  I bought a 5.1 channel Onkyo unit, and the sales associate managed to talk me into the Polk Audio garbage so that I would be able to "take advantage" of the new receiver's increased channels.  I don't generally use them for a several reasons.  They sound like crap compared to the TW's, and most of my audio sources are stereo rather than surround sound mixed anyway.  The Onkyo died a few years ago and was replaced by a Yamaha.  The Yamaha has an "All Channel Stereo" mode, but I prefer "Straight Audio" 2.1 channel mode with the sub-woofer giving the TW's the low frequency assistance they need.  

[ArchStanton]

I had planned on further tweaking and testing last night, but I hit a snag.  Initially, I was going to rerun all of my original benchmarks with only 2 DIMM slots populated.  I wanted to see if asking a little less of the IMC had any effect on the performance at stock settings.  Also, I convinced myself that I will have more headroom running only two sticks for all the usual reasons plus the added benefit of better cooling of the RAM itself (some of my reading yesterday has led me to believe that keeping the RAM at 40 Celsius or lower is likely to pay dividends when trying to push a memory OC).

 

After removing the inner most stick from each channel (the MB manual indicates populating the 2nd and 4th slots is the preferred configuration when only 2 sticks are used) the system operated normally through all but my last benchmark (3DMark CPU profile).  When the test reached the 4-thread section, performance dropped down to comically low levels and stayed there through the duration of the test.  In fact, the system remained "gimped" after the test concluded.  Laggy in the extreme.

 

Here I failed to document the situation correctly.  I got flustered and focused on resolving the issue without bothering to track what I was doing.  I checked task manger to see if Windows was up to something in the background (not that I could see).  I restarted the system and initially things were normal.  I began to rerun the 3DMark bench and this time the system entered "borked" mode during the 8-thread section.  I restarted and looked things over in BIOS.  I booted into Windows and fired up CB R23.  I was blessed with a "mind-blowing" MC score of 3551...

 

I started thinking that maybe I had dislodged a stick of RAM while removing two of them.  Three RAM re-seats/restarts/borked benchmarks later I decided that wasn't the case.  It was "cool and dry" in my laboratory last night, and while I hadn't donned my anti-static wrist strap, I had been careful to ground myself right before touching anything in the system, but maybe not careful enough.  I say this because, removing the RAM from dimm B2 and replacing it with one of its fellows seems to have eliminated the issue.  I'm not convinced that this stick was defective from the factory because it had passed at least 4 iterations of memtest86 in this system prior to the creation of my original post.  I left the unit running AIDA64's generic "system stability" test overnight and things seemed correct when I left the house for work this morning.

 

I keep telling myself I'll need to be more "detached" when/if "hiccups" occur in the future .  I'll try to stay calm and do a more thorough job of diagnosing/documenting the issue.  All the benchmarks that completed correctly before the issue manifested showed at most a marginal improvement that I think could be explained by a temperature variation in the testing area (about 1.5 Celsius cooler than previous).  I hope to post those results later today (those that I didn't inadvertently erase in my haste to "fix my baby").  I think I'm likely to run a battery of longer stability tests prior to making any further tweaks.  I plan to include OCCT/testmem5, generic prime95, and maybe prime95 through corecycler as well.  I'm considering adding geekbench5 as another general-purpose performance metric.

 

c'est la vie 

[J7SC_Orion]

2 hours ago, ArchStanton said:

I had planned on further tweaking and testing last night, but I hit a snag.  Initially, I was going to rerun all of my original benchmarks with only 2 DIMM slots populated.  I wanted to see if asking a little less of the IMC had any effect on the performance at stock settings.  Also, I convinced myself that I will have more headroom running only two sticks for all the usual reasons plus the added benefit of better cooling of the RAM itself (some of my reading yesterday has led me to believe that keeping the RAM at 40 Celsius or lower is likely to pay dividends when trying to push a memory OC).

 

After removing the inner most stick from each channel (the MB manual indicates populating the 2nd and 4th slots is the preferred configuration when only 2 sticks are used) the system operated normally through all but my last benchmark (3DMark CPU profile).  When the test reached the 4-thread section, performance dropped down to comically low levels and stayed there through the duration of the test.  In fact, the system remained "gimped" after the test concluded.  Laggy in the extreme.

 

Here I failed to document the situation correctly.  I got flustered and focused on resolving the issue without bothering to track what I was doing.  I checked task manger to see if Windows was up to something in the background (not that I could see).  I restarted the system and initially things were normal.  I began to rerun the 3DMark bench and this time the system entered "borked" mode during the 8-thread section.  I restarted and looked things over in BIOS.  I booted into Windows and fired up CB R23.  I was blessed with a "mind-blowing" MC score of 3551...

 

I started thinking that maybe I had dislodged a stick of RAM while removing two of them.  Three RAM re-seats/restarts/borked benchmarks later I decided that wasn't the case.  It was "cool and dry" in my laboratory last night, and while I hadn't donned my anti-static wrist strap, I had been careful to ground myself right before touching anything in the system, but maybe not careful enough.  I say this because, removing the RAM from dimm B2 and replacing it with one of its fellows seems to have eliminated the issue.  I'm not convinced that this stick was defective from the factory because it had passed at least 4 iterations of memtest86 in this system prior to the creation of my original post.  I left the unit running AIDA64's generic "system stability" test overnight and things seemed correct when I left the house for work this morning.

 

I keep telling myself I'll need to be more "detached" when/if "hiccups" occur in the future .  I'll try to stay calm and do a more thorough job of diagnosing/documenting the issue.  All the benchmarks that completed correctly before the issue manifested showed at most a marginal improvement that I think could be explained by a temperature variation in the testing area (about 1.5 Celsius cooler than previous).  I hope to post those results later today (those that I didn't inadvertently erase in my haste to "fix my baby").  I think I'm likely to run a battery of longer stability tests prior to making any further tweaks.  I plan to include OCCT/testmem5, generic prime95, and maybe prime95 through corecycler as well.  I'm considering adding geekbench5 as another general-purpose performance metric.

 

c'est la vie 

 

FYI re. the RAM issue you describe: 4 sticks versus 2 sticks, and SR vs DR - all makes a big difference in Ryzen 5k

 

 

 

[ArchStanton]

2 minutes ago, J7SC_Orion said:

 

FYI re. the RAM issue you describe: 4 sticks versus 2 sticks, and SR vs DR - all makes a big difference in Ryzen 5k

 

Aye, I have seen both of the linked videos in the past, and both they and further reading elsewhere prompted me to investigate this avenue since 32GB is likely perfectly sufficient for my day-to-day needs, but having the option to compare to 64GB was something I wanted as well.

[ArchStanton]

Some results from before the levy broke last night.

 

 

[ArchStanton]

A 3DMark CPU Profile run completed after swapping RAM module located in dimm B2.

 

 

[ArchStanton]

Summary.  No real performance changes detected by my test regimen.  Likely just some thermal and run to run variance even for the memory latency, I think.

 

 

 

Edited January 22, 2022 by ArchStanton

[ArchStanton]

If I understand this correctly, my system is in "quad rank" mode with 2 modules of this RAM installed.  2 ranks per module * 2 modules installed = 4 ranks total.

 

 

[ArchStanton]

5 hours of Aida64's seemingly underwhelming "stability test" yielded the following (I used it because it was at hand and ready to run).  I think I'll make a few passes of their "cache & memory benchmark" to see if that slight reduction in memory latency might be real and then maybe some prime95 small fft with AVX for some additional assurance that things are like they should be.  I do find it interesting to note that my RAM exceeded 40 Celsius with only two modules installed while using my current fan control curve, but it's still running at the "DOCP" specified 1.45v.  I plan to see if there are gains to be had with a manual memory OC using 1.40v at some point in the future.

 

 

[ArchStanton]

5 run average memory latency in Aida64 Cache & Memory benchmark = 62.04ns.  So possibly a genuine 2.5 to 3 percent improvement over the 64GB results, but this did not seem to translate into the scores from my stable of other benchmarks.