Weird issue with old Core 2 Duo and 965 board.

[Joepublic2]

I'm working on my brother's computer which was acting weird (probably some spyware/virus or something) and noticed something strange when I ran memtest with 4 DIMMs installed (normally there's just 2; upgrading the RAM because he wants to install win7).

Here's it with 2 1GB DIMMs (it passes memtest btw and the hardware works fine and benches where you'd expect it to):

Here's it with 4 1GB DIMMs installed:

Any ideas what's going on here? It takes about 20-30 seconds to post but doesn't throw any errors, it just runs really, really slowly as might be expected. It'll even try to boot windows but it takes 10-20 minutes for the xp boot animation to come up. It does the same thing with 4 different pairs of DIMMs, too, even installed in different configurations. The power supply is brand new and the waveform for all the pots looks good under load; I think I'm going to recap the board sometime this week and see if that fixes it in the meantime.

 

[VirtualLarry]

First of all, use Memtest86+, the plus is important. I've found that the non-plus version is just plain buggy.

 

[Joepublic2]

First of all, use Memtest86+, the plus is important. I've found that the non-plus version is just plain buggy.

Okay; I'll try that tommorow.

 

[Joepublic2]

Does the same thing in Memtest86+ but I have an extra clue now:

It's not reading SPD info from one of the DIMM slots. The slot it's not reading changes at random when I have 4 DIMMs in there so I think the slots are at least physically okay.

 

[Joepublic2]

Well, a friend that owns a computer shop I used to work at hooked me up with an old supermicro p35 board he hasn't been able to sell locally. I snagged a q9300 with a stock HSF off ebay to put into it (27$, it would have cost at least 25$ to recap this gigabyte board) and 2x2GB of ddr2 to go with the 2x2GB I already had. Should have sprung for the DS3 with polymer caps originally over the S3 with electrolytic caps, but my bro got a free upgrade out of it.