Jump to content

Edit 2: Solved

Edit: Fixed the font size.

 

Hi, I've got a problem with my PC and can't quite figure out what causes it. Helpful advice is much appreciated, but read my post completely first please.

 

My hardware is the following:

-CPU: Intel Core i5-4670k

-MB: Gigabyte GA-Z87MX-D3H

-RAM: Corsair Vengeance 4x4GB (DDR3-1600 CL9 1,5V)

-GPU: EVGA Geforce GTX 780 ti 3GB

-Storage: SanDisk Extreme II 480GB

-PSU: Seasonic X-650

 

I can reliably reproduce the following error, which depends on my RAM configuration:

 

When all four DIMM-slots are populated the system doesn't post and continually powercycles probably applying default RAM settings to try to remove the error.

When any one of the RAM-DIMMs is removed so that only three (or less) slots are populated, the system POSTs and boots without issues.

 

The system has been running for half a year with all four RAM-modules before they were temporarily replaced with two of 8GB capacity each, 2133MHz clock, CL10-12-12-28 timings and 1,6V operating voltage for about two weeks. The afternoon I reverted to the four DIMMs the system POSTed and booted normally. On the next morning however said behaviour occured.

 

I tried the following to remedy the issue, unfortunately to no avail:

-cleaned and reseated RAM-DIMMs

-cleaned and reseated CPU

-cleared CMOS memory

-reflashed BIOS

-replaced BIOS battery

 

What fixed the issue for me was to replace the DIMMs with four different ones of similar specifications (4x4GB, 1333MHz, CL9, 1,5V).

I am not convinced that this is a permanent fix though, because the Vengeance DIMMs work in the computer I took my replacement RAM from.

Reduce > Reuse > Recycle

 

Build-log (way out of date)

Link to comment
https://linustechtips.com/topic/865773-weird-pc-ram-behaviour/
Share on other sites

Link to post
Share on other sites

36 minutes ago, kb5zue said:

When was the last time the BIOS was updated?  Is there any particular reason that you believe that everyone reading your post is blind?  Your post comes across as if:      YOU ARE YELLING!!!!!

I'm sorry for coming across as attention seeking. I couldn't figure out how to change the font size on mobile - I'm on PC now.

 

The BIOS firmware is from 13. August 2014 and was reflashed onto the chip as an attempt to fix the issue.

 

To me this looks like a power supply problem, but I don't understand why that would be the case.

Another option is that the CPU somehow can't reach the full address space of the RAM-modules but I reseated the CPU and RAM and also every other connector on the mainboard without success.

Reduce > Reuse > Recycle

 

Build-log (way out of date)

Link to comment
https://linustechtips.com/topic/865773-weird-pc-ram-behaviour/#findComment-10752055
Share on other sites

Link to post
Share on other sites

My guess is that the board doesn't like the amount of ram. Make sure that the board supports the amount you are putting in. My gigabyte board doesn't support RAM with the memory chips on both sides of the dimm. If it's not that then I suspect a faulty socket or bad memory. 

Have you tried turning it off and on again?

Link to comment
https://linustechtips.com/topic/865773-weird-pc-ram-behaviour/#findComment-10752116
Share on other sites

Link to post
Share on other sites

2 minutes ago, fixitnow said:

My guess is that the board doesn't like the amount of ram. Make sure that the board supports the amount you are putting in. My gigabyte board doesn't support RAM with the memory chips on both sides of the dimm. If it's not that then I suspect a faulty socket or bad memory. 

The Vengeance RAM-kit is on the campatibility list. I've had two kits installed which is compatible as well.

 

The only socket that could be faulty with the given behaviour is the CPU-socket. I inspected it and the CPU thoroughly and they seem fine. There is no excessive mounting preassure to cause lift-off.

But either way I don't think there is an issue with the sockets, since different RAM works fine - as does the Vengeance RAM in a different board.

Reduce > Reuse > Recycle

 

Build-log (way out of date)

Link to comment
https://linustechtips.com/topic/865773-weird-pc-ram-behaviour/#findComment-10752133
Share on other sites

Link to post
Share on other sites

  • 4 weeks later...

To give this thread some kind of closure:

I contacted the Gigabyte support to get further information:

 

Quote

 

Hi, I can reliably reproduce the following error, which depends on my RAM configuration:

 

When all four DIMM-slots are populated with Corsair #CMZ8GX3M2A1600C9 modules the system doesn't post and continually powercycles in short intervalls (ca. 8s). When /any/ one of the RAM-DIMMs is removed so that only /any/ three (or less) slots are populated, the system POSTs and boots without issues and recognizes all installed RAM.

 

The system has been running for half a year with all four Corsair RAM-modules (#CMZ8GX3M2A1600C9, 4x4GB, 1600MHz, CL9-9-9-24, 1.5V) before they were temporarily replaced with a Mushkin kit (#997124F, 2x8GB, 2133MHz, CL10-12-12-28, 1.6V) for about two weeks. The afternoon I reverted to the four Corsair DIMMs the system POSTed and booted normally. On the next morning however said faulty behaviour occured.

 

I tried the following to remedy the issue, unfortunately to no avail:

-cleaned and reseated RAM-DIMMs

-cleaned and reseated CPU

-reseated power connectors

-cleared CMOS memory

-reflashed BIOS (F7)

-replaced BIOS battery

 

Sanity-Check

The Vengeance DIMMs work fine under full load in the computer I took my replacement RAM from.

CPU- and DIMM-sockets are in very good condition and show little signs of wear.

The mainboard has no visible damage and is clean.

There is no excessive mounting pressure on the CPU to cause lift-off and the DIMMs don't mechanically interfere with anything.

All clocks and voltages are running as specified by Intel (i.e. Auto setting).

 

What fixed the issue for me was to replace the DIMMs with four different ones: -Geil (#F3-10600CL9D-8GBNT, 2x4GB, 1333MHz, CL9-9-9-24, 1.5V) -Crucial (#BLS2CP4G3D1609DS1S00CEU, 2x4GB, 1600MHz(running @ 1333MHz), CL9-9-9-24, 1.5V)

Replacing only two DIMMs didn't help.

 

I am now worried that there might be a power supply issue, caused by the Corsair DIMMs drawing too much power and degrading the DC/DC circuitry in the process. Is that likely?

 

How do I get the Corsair RAM to work again?

 

 

The answer came 3 days later suggesting to test each DIMM in each socket first and then if there weren't any problems to fill all slots one-by-one and test again with each module added.

 

Quote

 

Sehr geehrter Herr Karikiro,

Vielen Dank für Ihre Anfrage.

Wir freuen uns über Ihr Interesse an Produkten von GIGABYTE.

Den einfachsten Test haben Sie leider nicht gemacht.

Testen Sie erst mal mit nur einem Modul jede Bank einzeln auf Funktion.

Dann wenn alle Bänke OK sind, System startet immer mit einem RAM auf Bank1 -2 -3-4 testen Sie jedes Modul auf einer Bank auf Funktion.

Geht auch dies, kann folgender Fehler vorliegen.

Netzteil

CPU

Mit freundlichen Grüßen

Ihr GIGABYTE-Team

[...]

 

 

After testing that without success I went with a more bruteforce method: (almost) any combination possible.

Which lead me to the now fully functional solution:

The ver8.16 DIMMs must be inserted into the slots 1 & 2 in any order to function.

 

 

Reduce > Reuse > Recycle

 

Build-log (way out of date)

Link to comment
https://linustechtips.com/topic/865773-weird-pc-ram-behaviour/#findComment-10844039
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

×