TR Forums

Ok, so I ordered a "bare-bones" kit from Tiger Direct consisting of a Gigabyte mother board, AMD Phenom II x6 1100t six-core CPU, 500 GB Western Digital HDD, 4 GB Corsair PC10666 DDR3 RAM, and a cheap case. Oh and an SATA DVD drive from an old system. Assembled it and tried to install a brand new copy of Windows 7 (64 bit), but it kept shutting off during the installation without completing it. I tried several times with no avail. I decided to try another OS, so I used an old copy of XP I had lying around, and again it didn't make it through the install (I even got the dreaded BSOD). By now I'm a bit perturbed, but I try W7 again just for kicks. And it worked...sorta. It made it through the install, then I installed the driver disc that came with the MoBo, and it worked, but everything was running really slow and I had the hourglass/spinning circle thing for about 15 minutes after start-up. Installed drivers for audio interface, and same thing... finally finished after about 20 minutes, restarted and 15 minutes of hourglass before I could do anything. BTW I'm using this PC for music production.

I'm far from being an expert in this stuff, but it seems to me that maybe the hard drive could be faulty? Could that explain why everything is taking so long to install, including reboot time? If not, what else could be causing this?

After I installed my audio drivers (for an external USB audio interface), I tried installing FL Studio (an audio sequencer program) and after installation and reboot (so about 30 minutes later) the desktop icon was some weird rectangle (instead of the FL icon) and when I clicked on it, it said the specified path does not exist (or something like that). Then I was locked up... mouse cursor would move, but opening folders, start menu, etc. wouldn't work. Basically, something is not right.

Seeing as I ordered this as a kit, I don't have tech support from the manufacturer, so I have to determine what component is at issue here so I know what to exchange. Any ideas? I spoke to a buddy about it, and he thinks it's probably the HDD, which was my guess, but I'm certainly not sure. If I can't figure it out I'll just take it to a pro, but after dropping several hundred $ on a PC, I really don't wanna pay for anything right now.

Any ideas/suggestions?

Are you able to get into the BIOS? There should be a diagnostic screen. Are all temps and everything looking good from there? A higher than normal CPU temperature could indicate the CPU cooler isn't properly seated.

My guess: RAM. A lot of these issues can be caused by bad RAM. Some of them have happened to me.



You best way to confirm this suspicion? Grab a copy of Memtest 86+ Burn it to CD. Boot it with the CD in the drive and let it do its things for a few hours. If come back to a screen disturbingly filled with red entries in the bottom half, you have an issue with your RAM.

If memtest86+ doesn't report any errors, and you suspect it may be the hard drive, you can test it with hddscan to make sure, http://hddscan.com/

Hey guys, thanks for the quick replies. Let me first state that I'm a complete noob with this stuff. This is my first excursion in to building a PC, so I'm not very techy with this sort of stuff. So with these 2 applications you guys suggested (memtest86+ and hddscan), and I able to run these on a PC that isn't really functioning? Like w/ that memtest, do I just boot from that disc?

As far as the BIOS goes, I'm not really even sure how to do that. I know, it's pretty sad. I just figured everything would, ya know, WORK. What would I have to do to determine if it's running too hot? Like, step by step.

I really appreciate any and help, really I do. This is my first new PC in years, and I just want it to work. Thanks so much.

Also, the reason I suspected the the HDD was that it took numerous attempts at installing W7, and I couldn't even get through an XP install. Could an improperly seated heat-sink cause that? Again, I'm pretty new to this stuff, but common sense would tell me that something is preventing the PC from writing and accessing data. Does that make sense? Anyway, I'm reinstalling W7 right now but the 32 bit version instead. I'll then try those apps and check the BIOS.

futilerecordings wrote:

Hey guys, thanks for the quick replies. Let me first state that I'm a complete noob with this stuff.... and I just want it to work. Thanks so much.

Nothing wrong with being a n00b. Everyone was there at some time. My only thoughts on those lines are that things going wrong are part of the learning experience. Persevere, and you get to be old and cranky and talk about your scars

So the way Memtest works is like this

From a working computer:
1. You download memtest86+ (I prefer the latest, stable ISO)
2. Burn the ISO to a CD on the working computer.
3. Power on the PC in question and insert the CD.
4. Reboot.

What ends up happening is a very lightweight program is loaded of the CD on reboot (instead of windows). It sits in the smallest amount of memory it can, and then does nothing but write and read patterns of data to the rest of your RAM. If it doesn't get an expected pattern back, it says there's an error with your memory.

The reason I suggested memory to be a possible culprit is that memory errors are actually a fairly common occurrence (especially for built from scratch units), and pretty much anything your computer does requires passing data to and from the RAM. Reading from a DVD and then installing to an HDD requires data processed in RAM. Loading windows on start up requires data being read into RAM. Processing an audio sample requires some sort of read/write to the RAM.

The symptoms are here too (though they're fairly similar for a bad HDD too. so that remains a possibility).

- If the instructions or the data is written to RAM, and then read from RAM, and doesnt touch the part of RAM that is bad, then everything goes fine.

- If the instructions or data is written to and read from RAM, and involves the bad part, then you get an error. You get some sort of data that wasn't expected.

-- If it's small, non-critical data, and the program is written well, it may be able to recover and try again, randomly writing and reading data to a good section of RAM this time. This is why things slow down.
-- If its a big error, or critical piece of data, sometimes this error really $#@$ things up. Then your PC crashes, blue screens or hangs, etc. etc.

Ok cool, so I burnt that memscan disc from an ISO, so I'll be giving that a try shortly. The HDDScan utility, does that work in a similar manner, or do you need to have an OS/working system?

Ugh... got halfway through memtest and it shut itself down... again. Had to reboot about 5 times in order for it to allow me to boot from disc again. WTF.

futilerecordings wrote:

Ugh... got halfway through memtest and it shut itself down... again. Had to reboot about 5 times in order for it to allow me to boot from disc again. WTF.

So that suggests it's something more fundamental, like overheating or power problems. You bought a retail CPU so you're using the stock heatsink, correct? Did you have any issues when attaching it?
I assume that "cheap" case came with an even cheaper (ie free with the case) power supply? You plugged both power cables into the motherboard, right?

Well it finished the memtest and it during the test it kept saying "no problems detected". However, it also said it would display the test results when I log back on, but it didn't do that... of course, I can't log-on. I don't even have a working OS. So mad.

futilerecordings wrote:

Ugh... got halfway through memtest and it shut itself down... again. Had to reboot about 5 times in order for it to allow me to boot from disc again. WTF.

UberGerbil wrote:

futilerecordings wrote:

Ugh... got halfway through memtest and it shut itself down... again. Had to reboot about 5 times in order for it to allow me to boot from disc again. WTF.

So that suggests it's something more fundamental, like overheating or power problems. You bought a retail CPU so you're using the stock heatsink, correct? Did you have any issues when attaching it?
I assume that "cheap" case came with an even cheaper (ie free with the case) power supply? You plugged both power cables into the motherboard, right?

No issues installing it. Yes, I'm using the stock heat sink/fan that came w/ the CPU. I guess I plugged both cables in to the mobo. I mean, if I didn't, would it even power up?

futilerecordings wrote:

Well it finished the memtest and it during the test it kept saying "no problems detected". However, it also said it would display the test results when I log back on, but it didn't do that... of course, I can't log-on. I don't even have a working OS. So mad.

Memtest just displays the results at the end of the run; it doesn't save them anywhere for future viewing. If it was just a memory error, you'd see those errors at the end of the run; but If the machine shutdown/rebooted (for other reasons) during a run, you'd have no record of what was going on just prior to the failure.

futilerecordings wrote:

No issues installing it. Yes, I'm using the stock heat sink/fan that came w/ the CPU. I guess I plugged both cables in to the mobo. I mean, if I didn't, would it even power up?

I don't know; I don't go around testing motherboards without plugging them in properly and I doubt anybody else has done systematic tests of that. But I wouldn't be surprised if got at least partway through the boot process (though you might get a motherboard warning): the second power cable is mostly necessary for suppling the juice to the PCIe slots: modern CPUs and chipsets don't pull a lot of power on their own, and it's not like the boot process is going to load the CPU the way some intensive loads (games, Prime, etc) might. So it might at least boot. You should double-check all your cables just in case.

When you first boot the machine and go into the BIOS (this requires a keypress like Del or F1, as indicated on the screen), there should be a screen that shows temperature readings for the CPU (and other things, like the chipset). If you give us the precise model of your motherboard, we can walk you through the process (or point you to the right page in the manual).

futilerecordings wrote:

Ugh... got halfway through memtest and it shut itself down... again. Had to reboot about 5 times in order for it to allow me to boot from disc again. WTF.

This looks like a heat issue... the machine is running okay, and then as it heats up you eventually run into issues. As some others have mentioned, check that you have your heat sink seated properly, that all of your fans are working, etc.

cphite wrote:

This looks like a heat issue... the machine is running okay, and then as it heats up you eventually run into issues. As some others have mentioned, check that you have your heat sink seated properly, that all of your fans are working, etc.

^^this is VERY LIKELY


Ah, the joys of the first build . . . well, even subsequent builds can be troublesome . . .

My buddy built his first PC last year (with my help), and had loads of problems. Two things we tracked it down to were the case grounding stand-offs and the power supply. For some reason, the case didn't have enough grounding capacityfor the mobo, and the PSU didn't like DDR2 mobos (higher voltage than DDR3 maybe?). A different case and different PSU fixed it, and now he games just happily

Anyway, the way we got his system work was systematically testing all of the critical components idividually(yes, a binary chart). Using my working system, we tested all of his components singularly until we had identified the problem. My recommendation to you is to try the same - start with a known working system of similar components and start testing away. And, only use one stick of RAM for the testing.

But, in all honesty, in order for you to more quickly cut to the chase, to me this really does sound like the same issue we were having, and what I truly suspect is that the "kit" purchased came with a third-rate power supply. I would swap that first (no need to be tidy about it, just get it running - should only take 15 mins for an ugly swap). We ended up being able to strip two systems and swap mobos in just 15 mins after 1000 tries at it (and that's all stand-offs and PSU cables being in place, and basic essential things like a single HDD, single RAM stick, and CPU fan - nothing else matters to get it up and running, so let that DVD cable dangle and audio sit aside until you get it all worked out).

In summary, check to make sure all your stand-offs are screwed in, and the mobo is attached to those stand-offs. Then swap power supplies with a working system and check the voltages in BIOS. I think the PSU is the culprit. You should know by the time the system POSTs if it's going to work or not.

Which PSU and case are you using?

This is the case/PSU: http://www.tigerdirect.com/applications ... No=7200803

Regarding it being a heat issue... perhaps you misread my post, or I didn't explain myself properly, but it was never running properly. Like I said, it took 5 attempts to even get the OS installed, so it's not as if it was running for a while, then stopped. Once I finally got it installed, the smallest little task (such as opening a folder) would take literally 30 minutes. Am I right in thinking that, if there were an issue with the HDD such that it was having trouble writing and retrieving data, it would cause the OS install to take an exceptionally long time and accessing info would also take long?

Fortunately Tiger Direct is being cool... they're sending me a replacement HDD, so HOPEFULLY that is the problem and it will be solved. If not, at least it narrows it down. (Just re-read something... I realized when I was doing the memtest, it didn't stop halfway through like I thought. It finished the test and, upon completion, automatically restarts the PC. My mistake. So it actually completed the memtest and came back with no problems).

I really appreciate all the advice guys... I guess I'm a bit stubborn with this sort of stuff. Maybe if I understood these things better, but it doesn't make sense to me that a power or heat issue could cause a PC to run incredibly slowly with no error messages, etc. Given the fact that I can't really do anything in Windows, is there a way to find out if it's running too hot or if the power isn't right in BIOS? If so, how?

Anyway, again I really appreciate the help.

paulsz28 wrote:

cphite wrote:

This looks like a heat issue... the machine is running okay, and then as it heats up you eventually run into issues. As some others have mentioned, check that you have your heat sink seated properly, that all of your fans are working, etc.

^^this is VERY LIKELY



But, in all honesty, in order for you to more quickly cut to the chase, to me this really does sound like the same issue we were having, and what I truly suspect is that the "kit" purchased came with a third-rate power supply. I would swap that first (no need to be tidy about it, just get it running - should only take 15 mins for an ugly swap). We ended up being able to strip two systems and swap mobos in just 15 mins after 1000 tries at it (and that's all stand-offs and PSU cables being in place, and basic essential things like a single HDD, single RAM stick, and CPU fan - nothing else matters to get it up and running, so let that DVD cable dangle and audio sit aside until you get it all worked out).

In summary, check to make sure all your stand-offs are screwed in, and the mobo is attached to those stand-offs. Then swap power supplies with a working system and check the voltages in BIOS. I think the PSU is the culprit. You should know by the time the system POSTs if it's going to work or not.

So is it feasible to take the PSU out of one of my old PCs and use it in this one? What sort of specs should I be looking at to know if it's compatible?

Memtest86+ does not reboot at the end of the test, it loops continuously until you shut the PC down. If the PC rebooted during the Memtest run, there's a hardware problem somewhere. DO NOT even bother wasting your time trying to reinstall the OS until you've got the hardware issues sorted; you'll just end up with a corrupted Windows install and more frustration.

In decreasing order of probability, I'd say the culprit is likely to be:

1. RAM
2. Motherboard
3. Power Supply

If the power supply is a no-name or generic brand, bump that up from #3 to #1.

It could also be an improperly seated CPU heatsink. Go into the BIOS, find the page where it shows you the readings from the temperature sensors, and tell us what your CPU temperature is.

The OP mentioned earlier that he wasn't sure how to get into the BIOS to check temps.
So I figured I would pop in to give a quick tip for getting in the BIOS for a quick check.
Most often at the initial post screen when the machine turns on it should say on the screen how to get into the BIOS ( only for a few seconds )
Most motherboards require you to hit the delete key or F12 key or something such to get to the BIOS. from there, it's just a matter of searching through the BIOS screens a bit 'til you find some sort of diagnostic area.

At least if it's just heat then it's something a CPU re-seating or some such that needs to be done. If temps are fine hardware mentioned previously is very likely the problem.

futilerecordings wrote:

So is it feasible to take the PSU out of one of my old PCs and use it in this one? What sort of specs should I be looking at to know if it's compatible?

Answer: A BIG Maybe. You would need the check the Newegg Power Calculator to determine a general range of PSUs to use (note: this is NOT the be-all-end-all power calculator - adjust by 10%-20% as you feel comfortable http://images10.newegg.com/BizIntell/to ... index.html). Based in THIS calculator, you only need a ~350W PSU, but I suspect that ANY system running an 1100T needs a PSU capable of 500W absolute minimum, 600W sensible minimum. AND, the rail voltages matter a lot. You would need at least a dual rail PSU, that can support ~20A per rail (more is better, of course).

I don't see in your original post anything regarding a graphics card (GPU - graphics processing unit). Did you choose a mobo with integrated graphics, and if not what discrete GPU (video add-in card) are you running?

just brew it! wrote:

It could also be an improperly seated CPU heatsink. Go into the BIOS, find the page where it shows you the readings from the temperature sensors, and tell us what your CPU temperature is.

^^this too

After reading this post a little more, I think an obvious check-and-see task is to make sure you have enough (and not too much) thermal paste on your CPU/HSF (heat sink fan). Too much, oddly, can cause overheating issues, which causes your CPU/mobo to shut down when it detects too-high temperatures.

For AMD processors and coolers, the Arctic Silver company (you do at least have Arctic Silver 5, don't you?) recommends to REMOVE the factory thermal paste (with proper removal agents - see Tech Report "How to Build a PC" article http://techreport.com/articles.x/13671) and apply a half-rice-grain-sized drop of their thermal "goo." I've used AC5 and this measurement on EVERY AMD processor I've had since 6 years ago, and mine run very cool. I've in the past caused over-heating issues myself by having TOO MUCH thermal paste. Remember, with this "goo," less is more - half-rice-grain-sized is ideal - it'll spread by itself once it heats up, no worries about manually spreading via a toothpick and such.

Let us know. We've all had problems with multiple builds, and all the problems are common - ALL PROBLEMS CAN BE SOLVED! Let us know, and good luck - WE'RE HERE TO HELP!!

I would also check the alignment of standoffs and the Mobo.. I remember one build that I did where I was swapping out an old mobo for a new one, and I didn't take all the standoffs off the case. These were metal ones, and the new mobo didn't line up perfectly with one of the standoffs. Silly me I thought what was the worst that could happen and fired up the machine. For 3 months I kept getting random restarts and lockups because the stand-off was causing an intermittent short.

Also with regards to the heat issue, a machine doesn't have to be running for a long time in order to heat up the insides of your case. Think convection oven, and if a fan isn't working, the air can get hot much quicker than you think. It could also be why memtest completed, but as soon as the machine booted up you started having issues

@ Just brew it: Memtest86 definitely reboots at the end, it specifically states during the test that it will finish upon completion and display the results once you log on. I also tried several times to remove the CPU & heatsink and replace them... I'm pretty sure they're seated properly.

@ paulsz28: Just using the built-in graphics card in the mobo at this point.

@ druidcent: What exactly are "standoffs"?

To everyone else who replied, I really appreciate at. Like I said, I have a new HDD coming, so that will certainly narrow down the issue. One thing I didn't think about until someone mentioned it in this thread... The original processor that was included in this kit was the 1050t, but I upgraded to the 1100t. Paulsz28 mentioned that "I suspect that ANY system running an 1100T needs a PSU capable of 500W absolute minimum, 600W sensible minimum" so that just might be the issue.

I was on a budget, so I had to cut corners somewhere. I planned on purchasing higher-end graphics card later, as well as a better PSU (and more RAM of course). So even if the new HDD corrects the problem, I'll still be getting a new PSU very soon. Thanks for the help everyone.

paulsz28 wrote:

For AMD processors and coolers, the Arctic Silver company (you do at least have Arctic Silver 5, don't you?)

If he's got an AMD retail boxed CPU the stock cooler and thermal paste should be adequate unless he's planning to overclock or wants a larger, lower RPM fan to reduce system noise.

To the OP: Did this "kit" come with the CPU already installed on the motherboard? If not, was the CPU in an AMD retail box with a heatsink, or did it come in plain packaging with a separate aftermarket heatsink? Just trying to get a better idea of where you started from to try and figure out what might have gone wrong.

druidcent wrote:

Also with regards to the heat issue, a machine doesn't have to be running for a long time in order to heat up the insides of your case. Think convection oven, and if a fan isn't working, the air can get hot much quicker than you think. It could also be why memtest completed, but as soon as the machine booted up you started having issues

If the CPU heatsink isn't properly latched down or has no thermal paste on it, the CPU can overheat *very* quickly -- a matter of minutes or even seconds depending on how badly the heatsink is attached, and how much CPU load you've got.

futilerecordings wrote:

@ Just brew it: Memtest86 definitely reboots at the end, it specifically states during the test that it will finish upon completion and display the results once you log on. I also tried several times to remove the CPU & heatsink and replace them... I'm pretty sure they're seated properly.

That's not the same Memtest people are talking about here then. You really should be using Memtest86+. One pass will give you reasonable confidence that there are no major issues; ideally you should let it run 2-3 passes, or overnight.

If the memory is in fact good, then I guess I would add the hard drive back into the list of potential culprits.

I thought about this last...but I think that it is the most likely cause, so I'll put it first...then comment on your other questions.

For everyone's reference, is this the combo package that you have? If so, I think that I found your problem. That motherboard, the Gigabyte M68MT, does not support 125W Phenom II chips, only 95W chips. There is a 95W 1055T, but not a 95W 1100T.

Rest of the comments.

@ Just brew it: Memtest86 definitely reboots at the end, it specifically states during the test that it will finish upon completion and display the results once you log on. I also tried several times to remove the CPU & heatsink and replace them... I'm pretty sure they're seated properly.

Memtest definitely does not reboot. You might want to do the classic "is this disk burned right" test: re-burn it on a slower speed setting. Sometimes old or low-quality CDs don't burn well at default (typically highest speed for the burner) settings. It might be worth a try anyway.

Removing and replacing the heatsink can mess up the layer of thermal compound...which is not great in the long term. But it is unlikely to cause major problems in what you are doing. You might double-check that the fan on the cooler is connected to the CPU fan header on the motherboard. Some motherboards put other fan headers near the socket and the BIOS tends not to like to not see a CPU fan attached to that specific connection when temperatures start to rise during normal operation. I don't think that this is a likely problem...but it is worth looking. You would, by far, not be the first person to use fan1 instead of cpufan.

If you want to check CPU temperatures without an OS installed, you might try a live Linux distribution CD. There are several temperature monitoring programs under Linux. The Knoppix live DVD (not the CD version) has GKRellM, a monitoring program that can give you CPU temperature (and a host of other things).

@ paulsz28: Just using the built-in graphics card in the mobo at this point.

Then you almost certainly don't have a total wattage problem from the power supply...but it still could be a flaky supply with unstable voltages. A 1100t system, with an additional graphics card under load could need 500+ W. What you are running, no chance of a power (wattage) issue. Again, that doesn't rule out the power supply only that rating is an issue.

You asked earlier if you could move a power supply from an older system. Yes. It is pretty straightforward...provided you have all of the needed connections. It sort of depends on how old the particular power supply is. The most important thing is that you need to look at the connections that go to the motherboard. It would be nice to have a SATA power connector for the DVD at least...if you do not have an LP4/SATA adapter cable. Powering the hard drive is actually not a requirement at this point since everything that you are running is running from the DVD drive.

@ druidcent: What exactly are "standoffs"?

Standoffs (link to a picture) are the little hexagonal-shaped bolts with a threaded head that you screw into the case so that the motherboard does not ground itself electrically to the case. What they are saying is that there are more holes in most cases than are needed for standoffs for any given motherboard since patterns can vary based on board size and manufacturer. You should make certain that the only standoffs in the case are the ones that you need for your motherboard. Otherwise, you could be touching an active electrical component with the metal of the standoff and grounding it to the metal of the case.

The other pictures in the article which had the above picture talks about standoff installation.

[SDG]Mantis wrote:

For everyone's reference, is this the combo package that you have? If so, I think that I found your problem. That motherboard, the Gigabyte M68MT, does not support 125W Phenom II chips, only 95W chips. There is a 95W 1055T, but not a 95W 1100T.

Depends on the rev of the board. According to Gigabyte's product page, the rev 3 M68MT supports the 1100T.

[SDG]Mantis wrote:

Memtest definitely does not reboot. You might want to do the classic "is this disk burned right" test: re-burn it on a slower speed setting. Sometimes old or low-quality CDs don't burn well at default (typically highest speed for the burner) settings. It might be worth a try anyway.

As I noted in my followup, I believe he is running a different memory diagnostic than we have been assuming.

One other thought -- if it is indeed a DiabloTek PSU, the PSU moves up on the list of potential suspects. IIRC they're one of those third-tier PSU vendors who tend to exaggerate their wattage specs. (And this is why I tend to avoid cases which include a PSU -- you never know what kind of junk you're gonna get.)

You guys are awesome... thanks so much for all the help.

@ Mantis: That is most definitely the kit I purchased, except I upgraded to the 1100t. It is, however, the Rev3 MoBo, which just brew it suggests is fine. Here's a link to that kit:
http://www.tigerdirect.com/applications ... CatId=5294

Regarding the memtest, I just had a look at the file I ran, it's called memtest86 4.0. Do you really think there could be that much of a difference? I assure you that this particular version said that it would reboot upon completion and post results upon logging-in. However, during the testing it has the status on-screen indicating that no problems have been detected. Anyway, I'm about to try that latest version... will it still run if you don't have a HDD installed? I took it out to ship it back to Tiger Direct.

The CPU did come in the AMD box with the heat sink and fan. I applied thermal grease of course.

Yeah so evidently the memtest86 doesn't run without a HDD installed. Told you guys I'm a noob
Anyway, so I have a new HDD coming... if that doesn't solve and it is indeed the PSU, could you guys recommend one that won't break the bank?

I spoke to AMD tech support just to do some verification... all the components are supported by the 1100t, including the mobo (Rev3).

futilerecordings wrote:

Yeah so evidently the memtest86 doesn't run without a HDD installed. Told you guys I'm a noob

Umm... what?

I *always* run it without a HDD installed. It is pretty much the first thing I do with new motherboards after installing the CPU and RAM, before I even hook up the HDD and install an OS.

You need to download the zipped bootable ISO, unzip it, and burn the image to a CD. The system you want to test boots directly from the CD.

just brew it! wrote:

futilerecordings wrote:

Yeah so evidently the memtest86 doesn't run without a HDD installed. Told you guys I'm a noob

Umm... what?

I *always* run it without a HDD installed. It is pretty much the first thing I do with new motherboards after installing the CPU and RAM, before I even hook up the HDD and install an OS.

You need to download the zipped bootable ISO, unzip it, and burn the image to a CD. The system you want to test boots directly from the CD.

Odd.. I had no problem running it (the ISO I burnt to a disc) before I removed the HDD... after I removed it, I couldn't run either version (4.0 or 4.2). It says "DISK BOOT FAILURE, INSERT SYSTEM DISK AND PRESS ENTER". The only thing I did since running version 4.0 the first time was remove the hard drive. Weird.

That link is EXACTLY was I downloaded and burnt to disc.

You need to go in your bios and set it to boot from your CD before it trys to boot the hard drive.