More system help needed
DanielH

I have 2 (maybe related problems)

My system had 4 gigs of memory with option to upgrade to 16 (4 slots). I removed the 2 2-gig cards and installed 2 4-gig cards. This is when the whole system started acting up.

1. Recall last post about this issue.

Since then I removed both hard drives I had and installed a new one. Still the same problem.

The old hard drive had Windows 10 upgraded from windows 7. The new one is a clean installation of Windows 10.

I never got this fixed and have just been muddling through. I cannot access the boot manager. I cannot do a system recovery from a CD because the boot manager is broken.

I can select "Boot from CD/DVD", but then I get the error "BOOTMGR is missing"

Sometimes the system will boot into the OS. Sometimes, I have to reboot 4 to 5 times until it does.

2. Sometimes, the system will go to the dreaded blue screen with a system component error. I thought that it might be overheating. I replaced the heatsink/fan with a better one. While I was talking to the salesman about the issue he said it might be the memory. I did recently replace them. He said I should try a program called memtest86. I ran that and it found no errors in the memory.

Chris Katko

memtest should find errors in anything that accesses RAM. (CPU, RAM)

You may have a broken motherboard. The hard drive controller, the lines to the connector, all kinds of things like that can die. Or it might be a bad power supply.

Also, did you measure if it's overheating or not? Depending on level of overheating, that won't affect a cold boot at all unless the heatsink is completely gone/misaligned. But it could affect midboot or desktop. But this doesn't really sound like an overheating issue.

[edit]

Ahhh crap:

DanielH said:

My system had 4 gigs of memory with option to upgrade to 16 (4 slots). I removed the 2 2-gig cards and installed 2 4-gig cards. This is when the whole system started acting up.

Does your motherboard have all of the case support screws correctly installed? That is... did you... by any chance... FLEX the motherboard when you put the new RAM in? I literally lost my motherboard that way. It killed off one of the memory channels. I took the RAM out because I was installing a new heatsink. The RAM side of the motherboard didn't have enough supports because I was lazy years ago and didn't buy more. It flexed. My computer subtly died more and more.

Motherboard failures are a PITA to diagnose.

Bob Keane
DanielH said:

My system had 4 gigs of memory with option to upgrade to 16 (4 slots). I removed the 2 2-gig cards and installed 2 4-gig cards. This is when the whole system started acting up.

What happens when you put the old memory back? And have you tried a Knoppix boot disk to see if the boot manager works?

Edgar Reynaldo

Second live disc, reseat memory, then try old memories again

Hard Rock

Is your ram compatible with your motherboard? They are very finicky. My motherboard supported 2x sticks of my ram, but when I added a second pair it did not support 4 x of them, even though I could mix and match any pair without issues! Drove me crazy!

I ended up buying a whole different brand of ram to solve the issue.

Steve Terry

Power supply... mine went belly up and had to replace it. 0 BSODs since.

Chris Katko

It's gotta be PSU, or motherboard related. (or "maybe" heat)

If memtest comes back fine, it's testing your complete CPU-to-RAM interface. So all of those should be fine.

Derezo

You can see your voltages under low use from the BIOS. It should list them in the power management as 3.3v, 12v and 5v. If they are off by + or - 5% you should replace the power supply right away. It's possible to do damage to your equipment if it is off significantly. There is some room for tolerance though, so if your 3.3v reads 3.4 don't panic ;D I'm not sure where to draw the line exactly, but somewhere around 3%.

These voltages may change under load though, so even if they read correctly here the problem may still be power supply related, especially if it happens more when you're using an auxiliary device like your DVD-ROM.

Chris Katko

PSU diagnosing is actually a PITA. Well, not as bad as thermals.

Why? Because EVERY DAMN APP gets "read my data" wrong because "my data" is actually supposed to be scaled by PropretaryAlgorithm(TM) which changes per-chip, and even then, thermals are only valid in certain ranges.

With 5 different motherboard apps, I get FIVE different thermal ratings. And they let you offset them... too bad you don't know what the offset is supposed to be. (And, with my AMD FX-8370, the low range offset is different from the high range offset. So I can either be accurate at low, or high, not both.)

That said, voltages are more likely to be exact, and, if in doubt, you can always get a simple multi-meter and confirm their output (assuming the voltages aren't dropping in short spikes that you can't see).

The point about voltages is, when in BIOS, you'll normally see fine voltages. But the second your computer boots up and gets under load, they may drop. And then you need other a motherboard monitoring app, or, a multimeter.

Thread #617189. Printed from Allegro.cc