So my PC is starting to show its age by occasionally doing stupid things to me. The latest one is this odd HDD "hitch" that seems to occur randomly, usually at the most inconvenient of times (like hours into a game when I haven't saved) and I'm having trouble tracking down a cause. I've already eliminated software and the drive itself as the source of the problem, so I am now looking at the mobo as the source. Does anyone know of a good (preferably free) utility that will help me diagnose a potential mobo problem?

By "hitch", do you mean random rebooting? If so, sounds like it could be an overheating problem.

You might want to try Motherboard Monitor: http://www.pcworld.com/downloads/file/fid,7309-order,1-page,1-c,alldownloads/description.html

Not exactly what I was looking for, that utility just reports system temperatures and I already have plenty of things that can do that for me. I'm looking for something that can monitor the activity of the motherboard components (the IDE controller, various ports and slots, etc.) and possibly catch whatever is happening in the act and tell me what is causing this.

No suggestions for the utility but . . . my XP system started blue screening on occasion recently. The change to colder temps affected my RAM. Pulled it out, cleaned the edge connectors and . . . fixed. . . =)

What sort of games are you playing?

If you're playing old games or games on an emulator, could it be that your hard drive is going to sleep / spinning down? The symptoms of such a "hitch" would be that your game stutters or pauses for a while, and saving your game would take a few extra seconds, or in some games might even report failure.

Have a browse around your computer's power options and see if disabling it, or increasing the value, helps. If that does help, you might want to consider leaving it disabled, or buying a new hard drive. Unfortunately motherboard problems are almost impossible to detect.

The only real evidence a motherboard gives that it's faulty are:
1) Power rail fluctuations (which are actually more likely to be a PSU problem), which you can usually check in your computer's BIOS.
2) Port failures (e.g. a flaky IDE or USB connector), which you can check by swapping devices to another port.

The best way I can think of to troubleshoot a motherboard problem is:
* Rule out the video card by running a 3D benchmark
* Rule out the CPU by running a tool like prime95
* Rule out the memory by running memtest86
* Rule out the power supply by borrowing another computer and rigging your new computer to its power (if the cables are long enough, you don't have to unscrew either PSU)
* Rule out the motherboard's BIOS by making sure it's up to date
* Rule out peripherals by running with them disconnected (you can run without a hard drive by using a LiveCD or flash drive OS)
* Rule out the OS by booting a different one (e.g. a linux LiveCD), and see if the problem is replicated, or whether it's unique to your OS. Note that a LiveCD won't really help troubleshoot hard disk problems, though.
* If the problem persists across all of these changes, then a motherboard problem is possible.

I was almost certain that my last PC problem was a motherboard failure, but it turned out to be the power supply.

No, its definitely much more than just the drive dozing off. I already have my machine configured so that the only thing that goes to sleep is my monitor. Also, games themselves are not a cause/factor at all, the problem occurs at really odd times, no matter what I am currently doing. It actually happened really badly earlier today while I was running a hard drive scan off of a bootable utility disk, which basically eliminated the possibility of any of the software or OS being the cause.

When the problem occurs, I can hear the drive actually "click" like it does when parks itself at power down, but none of the drive scans I have run report any problems with either of the hard drives in my system (even the scan from earlier today when I heard the drive "hitching") or any of the data on them for that matter, which would seem to indicate the drives are not at fault. I thought it might be the power supply crapping out, but a quick swap out with a different PSU confirmed that was not the issue at all.

The only thing I am left with is the mobo and I'm afraid I might be looking at a failing IDE controller, but I can't confirm that at all. None of the system logs in Windows or Linux show any errors at the time of the HDD incident, so I'm hoping that a third party app can help me out with this.

Now that I have mentioned Linux, I should also mention that in the course of trying to identify the problem, I have removed Linux from the system, so any suggestions anyone has will have to be a Windows-only app or something that is bootable. One of the first things I tried was checking all the connections and clearing dust. No joy. Of course you edited while I was typing. As you can see from my response, I have pretty much done all of that stuff, at least the pertinent ones.

This may sound like a dumb suggestion, buuuutt.... have you visually inspected the mobo for, say, bulging capacitors? Many a mobo have been done in by those, but they can also be replaced fairly easily.

Actually, no, I hadn't even considered that. I'll have to check that out tomorrow, I just got the case all back together after today's adventures in troubleshooting land, so I don't really feel like breaking it down again at the moment.

I have narrowed the occurrences of the problem down to something related to swap file read/writes... at least it seems like that when it occurs while within Windows. So I did a wipe of the swap file, defragged the drive I had it on, then put the swap back. I'm going to try running the machine through its paces tomorrow to see if I can replicate it again.

The "click of death" was my first thought, but after running multiple chkdsk's and other drive verification tools, I have not found a single data or hardware problem, hence why I am so stumped by this. So far, the removal/replacement of the swap file seems to have corrected the issue; I have stressed the system as much as I can for the past 24 hours and it hasn't happened at all, but I am going to continue to monitor. Fortunately, it is not the OS drive that this is happening to, its just a data drive, but it does have a lot of stuff I don't want to lose (music, game downloads, pictures, etc). UNfortunately, my external backup drive appears to have crapped out, but I'm reasonably certain its just a problem with the enclosure (the power cable looks damaged).

Looks pretty good, but it locked up as soon as I launched it. I'll keep trying it.
Not exactly what I need. It does provide a lot of detailed info about what actual components are in my system, but nothing about their current state.

EDIT - Dammit! It happened again! It was just once and it didn't repeat itself this time, so I immediately quit everything I was doing, fired up Sandra (fantastic app BTW, thanks for that Judas. Here, have a carrot!) and ran a burn-in test for over an hour. Nothing happened! No glitches, hitches, burp or farts. In fact, the performance results were pretty good, considering how crappy this machine is. I can't make heads or tails of this!

Thanks for the carrot :D

Well, looks like after all that, it is actually the "click of death". The drive in question is definitely failing, slowly but surely. Time to make a run to the computer parts store!

On that note, since I now have the opportunity to finally move this machine off of EIDE drives and onto SATA drives, anyone have any general advice/recommendations? I'm currently looking at getting a pair of 500GB drives like this:
http://www.microcenter.com/single_product_results.phtml?product_id=0340804

I've never actually owned a Samsung drive (always had Seagate or WD), anybody got any experience with them?

Good catch, this mobo is so old, it has SATA150 (1.5Gb/s) connectors only. I think I'm going to have a hard time finding those in any stores near me. I assume a 3.0Gb/s drive will not work with those connectors (i.e. throttle down to the connection speed)?

Getting a pair of 1TB drives would be awesome, but I'm trying to keep this under $100, if possible.

After a bit of confusing Googling, I pretty much found that same info. A couple of sites mention the drives might need a jumper change to restrict their speed, but otherwise they should work.

I don't like having a single drive just because of situations like the one I'm in now. My computer currently has a pair of identical IDE drive, one that has the OS and all my software installed (in separate partitions) and one for all personal data and swap file. If I have all that on one drive and it shits the bed, I'm screwed, but the way it is now, I'm only half screwed and at least have a chance to be fully not screwed. My best option for keeping things like that is to get a pair of 500GB drives (ooh, maybe I'll try tackling RAID again!).