Jump to content

Memtest Errors

Shisimona
 Share

Earlier today, I was playing around with OCCT. I did the Memory Test and errors were popped up instantly. I thought, this is odd as my system is running fine. Have no issues at all.

So before I went to work, I ran Memtest. It only did 4 passes as I have the free version and it came up with almost 1500 errors.

How do I have so many errors but yet my system runs perfectly fine. No BSOD, no shut downs, no restarts, no games crashing.

System is:
Ryzen 9 5900x

MSI MEG X570S Max Ace
G.Skill Trident Neo 32GB 2x16 GB 3200MHz 14-14-14-34
Seasonic Focus+ Gold 850 Watts
Asrock Phantom Gaming 6800XT 16GB
Samsung 980 Pro 500GB NVME OS Drive
WD Green 2TB Storage Drive
Crucial MX500 1GB Game Drive

Games I play are: Cyberpunk 2077, Saints Row 2022, Doom Eternal, Red Dead Redemption 2, Far Cry 6, and WoW. I can play hours with zero issues.

Everything is stock except I have my ram at DOCP which is 3200mhz 14-14-14-34. This is the same ram I used on my Asus Prime Pro X470 and Ryzen 5 2600x before I upgraded at the end of July.

Is there something else I can test the memory with?

Link to comment
Share on other sites

Link to post
Share on other sites

Memtest is rarely wrong. 

Try increasing the memory timings to something like 16-16-16-36  and run memtest again. 

make sure the ram runs with 1.35v or whatever voltage G-skill recommends.

If you still have problems, run them at 3000 Mhz and test again - if they don't get errors at 3000 it means your timings are too tight or the sticks can't handle 3200 mhz anymore for some reason.

Worth trying to move the sticks in the other two slots if the motherboard lets you just in case one of the slot is causing problem (oxidized contacts or some other reason)

 

Link to comment
Share on other sites

Link to post
Share on other sites

1 hour ago, Shisimona said:

Earlier today, I was playing around with OCCT. I did the Memory Test and errors were popped up instantly. I thought, this is odd as my system is running fine. Have no issues at all.

So before I went to work, I ran Memtest. It only did 4 passes as I have the free version and it came up with almost 1500 errors.

How do I have so many errors but yet my system runs perfectly fine. No BSOD, no shut downs, no restarts, no games crashing.

System is:
Ryzen 9 5900x

MSI MEG X570S Max Ace
G.Skill Trident Neo 32GB 2x16 GB 3200MHz 14-14-14-34
Seasonic Focus+ Gold 850 Watts
Asrock Phantom Gaming 6800XT 16GB
Samsung 980 Pro 500GB NVME OS Drive
WD Green 2TB Storage Drive
Crucial MX500 1GB Game Drive

Games I play are: Cyberpunk 2077, Saints Row 2022, Doom Eternal, Red Dead Redemption 2, Far Cry 6, and WoW. I can play hours with zero issues.

Everything is stock except I have my ram at DOCP which is 3200mhz 14-14-14-34. This is the same ram I used on my Asus Prime Pro X470 and Ryzen 5 2600x before I upgraded at the end of July.

Is there something else I can test the memory with?

Try setting timings manually. My buddy had same issue with the same kit on 5800x and B550 MB. Download Aida64 and go to motherboard->SPD. Find the 3200@14-14-14-34 1.35v. And write down all timings. Go to bios and set timings, dram voltage and frequency manually. Remember to fold the page out so you're able to see what timings are what.

image.png.4e4cfe362a20544264ca2cdb1014b5bf.png

CPU: i7 9700K@5ghz all core (silence + summer)

cache x47, AVX offset 0

Motherboard: Asus Maximus XI Hero

RAM: Patriot Viper Steel 2x8gb@3733mhz @jedec cl15

PSU: EVGA SuperNova 750 G3

Mon:  LG 27GL850-B, 144Hz Nano IPS - G-Sync compatible/Freesync

GPU: EVGA 1080ti SC Black Edition w/ G12 bracket

Sound: Odac + Fiio E09K

Case: Fractal Design R6 TG Blackout

Storage: MP510 960gb and 860 Evo 500gb

Cooling: CPU: Kraken X62 - GPU: Kraken X42(RIP) -> X53 (From RMA)

Link to comment
Share on other sites

Link to post
Share on other sites

So I ran at 3000 MHZ at 14-14-14-34 and it passed.

For tight timing, I bought this kit because it was advertised for 3200MHz 14-14-14-34 with XMP/DCOP.

But what i'm really trying to figure out, with so many errors, why wasn't my system crashing/BSOD.

Link to comment
Share on other sites

Link to post
Share on other sites

Well it could be the error was in just one chip  in one stick, for example 1 bit corrupt in memory range 32500-32768 MB and maybe it just happened you rarely used so many applications for Windows to put actual data in that range and corrupt it.

It could be the error only showed up when doing a certain pattern of requests... for example could be that chip on the ram stick fails only when the memory controller doesn't wait some amount of time until the chip is read. (for example that CL14 is cas latency, how much the ram chip needs to prepare all the internal connections and begin putting data on its contacts - the memory controller may request some data from that rank, then request some data from another rank (or the other channel) and it just so happened that doing so more than than that minimum time has passed so most of the time, even though you specified 14, maybe the memory controller took 15 or 16 to actually begin retrieving the requested data, and therefore you don't get errors (because the chip manages to setup everything by the time that value reaches 15)

It's just wild guesses... but as an easy fix I'd suggest just relaxing those timings.. the performance differences aren't really that big

Link to comment
Share on other sites

Link to post
Share on other sites

I played with the timings today. Anything running at 3200MHz or higher I would get errors. I even put the timings at 20-20-20-50 at 3200MHz and still would get errors. I even bumped the voltage high as 1.4, still errors. 3000MHz at 14-14-14-34 I get no errors.

I do believe G.Skill has lifetime warranty on their memory, so it might be worth to RMA them. 

Link to comment
Share on other sites

Link to post
Share on other sites

Here's an update.

When I was in the BIOS last night, I noticed the BIOS was from 2021. I updated the BIOS to the latest.

Ran Memtest overnight at the XMP/DCOP rating, and ran without zero errors. I then ran it a 2nd time to make sure it wasn't a fluke. Zero errors again. I only have the free version of Memtest, so I can only do 4 passes per run. I ran OCCT also, as thats how I found out about these errors and ran for an hour with zero errors. As before the BIOS update, the errors would have happen within seconds of starting the test.

I'm guessing the 2021 BIOS didn't like my ram and was not fully stable running at those XMP/DCOP settings.

Link to comment
Share on other sites

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share


×