Well, it took some time and a lot of headaches, but I finally found the cause of problems. Looks like it was the Video card the whole time. After some time, the screen would start spazzing out. I don't know why I always thought it was something other than video, but I did. Probably because the PC would usually be dead by the next AM. I went out and bought another cheapie PCI-e card and it's been fine. Now, I just have to go through the whole RMA process with MSI now. It's been almost 2 weeks withought a reply from them since I submitted an RMA request. I also have to RMA the Giga-byte Mobo too. The Asus board seems to be working fine, altough it won't run my memory at 4-4-4-12 like Crucial said it will. Have to keep it on auto and manually set the voltage to 2.2v to get the VDIMM correct. It's not an issue to me as I don't really notice any difference. I'm just happy it's been working for longer than 2 weeks. It's definitely soured me on MSI, though. I had problems with them in the past and was willing to give them another try since it had been 6 years. My feelings haven't changed. Thanks again for everyone's advice. Good thing the new Comp USA let's you return product, no questions asked. They're actually reasonable for a retail storefont. No wait, either!

Thanks,

Frank

Quote: Originally Posted by Durti Sahnchez Hey guys. 1st off, thanks to everyone that helped me back in Feb. with my hardware advice for my new box. Now I need help in figuring out why it has been a nightmare! Here's the deal. MB-GIGABYTE GA-EP35C-DS3R, Mem-Crucial Ballistix Tracer 2GB (2 x 1GB) 240-Pin DDR2 SDRAM DDR2 800 (PC2 6400), CPU-Intel Pentium E2160 Allendale 1.8GHz, VGA-MSI NX8800GT 512M OC GeForce 8800 GT 512MB, PSU-PC Power & Cooling S61EPS 610W, Cooling-Tuniq Tower 120. I ran it 1333fsb @ 3.0GHz (1.525v), Mem 4-4-4-12 @ 2.2v.

Tricky think about Overclocking, Frank ... you can only push a chip so far, for so long, and it will die. You shorten the life of any chip when you OC. So, given that you were hitting a ... let's see ... nearly 66% overclock, and overvolting the cpu to get it there, which means generating a heck of a lot more heat in the core ... probably, you killed the cpu.

Not all chips perform the same, even if they're from the same run. One person's e2160 will perform worse than the next e2160, and so on.

$SOLID$ probably has better info on this than I, however.

-godling

I would get a multimeter and test the PSU. Also, how are your temps when you are running the computer?

E

Quote: Originally Posted by Fred_G I would get a multimeter and test the PSU. Also, how are your temps when you are running the computer? E

With either CPU, I was running a steady 30c. I can understand killing the e2160 due to overclocking, but I ran the e8200 to spec and had almost the same results. Maybe I just got a bad one?

Also, what do I want to look for on the psu? voltages appeared to be ok in BIOS and on any of the windows apps that monitor voltages.

It's just frustrating troubleshooting a new system when I have no access to parts to swap out.

Quote: Originally Posted by godling Tricky think about Overclocking, Frank ... you can only push a chip so far, for so long, and it will die. You shorten the life of any chip when you OC. So, given that you were hitting a ... let's see ... nearly 66% overclock, and overvolting the cpu to get it there, which means generating a heck of a lot more heat in the core ... probably, you killed the cpu. Not all chips perform the same, even if they're from the same run. One person's e2160 will perform worse than the next e2160, and so on. $SOLID$ probably has better info on this than I, however. -godling

Killing CPUs is pretty hard to do these days. I've seen E21xxs stable at up to 3.6ghz with only moderate vCore...the 65nm can take quite a beating before dying....i stuck 1.75v through my Q6600 a couple of times and it was fine. Something very odd is going on though.

I find it hard to believe that your cpu's are dead. I would check the power supply, PC Power and Cooling make Great Stuff but every company produces a few dudes.

Quote: Originally Posted by Slaymate I find it hard to believe that your cpu's are dead. I would check the power supply, PC Power and Cooling make Great Stuff but every company produces a few dudes.

I thought the same thing. When I RMA'd the MB and memory, I put the e2160 back in and nothing. I bought the e8200 and it only ran 2-3 weeks. The video was squirrely too. Something's not right. I'll buy another psu if that's the problem. I can't warranty it because I clipped a bunch of extra connectors off when I put it together.

Quote: Originally Posted by Durti Sahnchez I can't warranty it because I clipped a bunch of extra connectors off when I put it together.

I can't say how much this is your problem. I can't stress how much you don't ever want to do that. For upgrade reasons alone it's the worst thing you can do.

Edit: I apologize to PC Power and Cooling for questioning the reliability of there stuff.

Quote: I'll buy another psu if that's the problem. I can't warranty it because I clipped a bunch of extra connectors off when I put it together.

Something screwed up is happening.....That's for sure

Pc power and cooling make some pretty stable power supplies. Im wondering if your psu mod did more than trick out your pc. It's the only item left since you already replaced everything with exception of the power supply.

Unless the board could be grounding out somewhere....My cookies are on the PSU.

Quote: Originally Posted by Rockin'male Something screwed up is happening.....That's for sure Pc power and cooling make some pretty stable power supplies. Im wondering if your psu mod did more than trick out your pc. It's the only item left since you already replaced everything with exception of the power supply. Unless the board could be grounding out somewhere....My cookies are on the PSU.

Basically, if anyone's ever seen the standard connectors on that psu, they are daisy chained off of the 1st connector. I de-pinned the connector using my molex tool, then clipped tapped wire off of the molex connector. I'm only using those to run the fans in the Antec 900 case. The SATA power connectors remain intact, along with the VGA and ATX connector. I pulled the motherboard both times and inspected. Everything looks good.

Now, that Giga-byte CPU lock seems to press down pretty hard on the cpu and would appear to apply more force to one side of the chip than the other. Maybe it's to blame? I would think there'd be other people with my problem if it were.

How about the Tuniq tower? The instructions were vague and is it possible to over-tighten it?

Can the video card cause these problems?

I'm tempted to put the system back together tomorrow with all the cables ran loosely for a couple weeks and see if something was pinching.

Thanks for everyone's input. It's good to get input from people with more knowledge than myself.

Quote: Originally Posted by Durti Sahnchez I'll buy another psu if that's the problem. I can't warranty it because I clipped a bunch of extra connectors off when I put it together.

ah, wait! I didn't know this before I posted my cpu analysis! Dan has to give me a break for a mis-evaluation ...

Quote: Originally Posted by Durti Sahnchez Basically, if anyone's ever seen the standard connectors on that psu, they are daisy chained off of the 1st connector. I de-pinned the connector using my molex tool, then clipped tapped wire off of the molex connector. I'm only using those to run the fans in the Antec 900 case. The SATA power connectors remain intact, along with the VGA and ATX connector. I pulled the motherboard both times and inspected. Everything looks good. Now, that Giga-byte CPU lock seems to press down pretty hard on the cpu and would appear to apply more force to one side of the chip than the other. Maybe it's to blame? I would think there'd be other people with my problem if it were. How about the Tuniq tower? The instructions were vague and is it possible to over-tighten it? Can the video card cause these problems? I'm tempted to put the system back together tomorrow with all the cables ran loosely for a couple weeks and see if something was pinching. Thanks for everyone's input. It's good to get input from people with more knowledge than myself.

I'd be tempted to assemble the main parts of the system outside the case, say on the motherboard box for example (or even better, a test-bench). Insert only the CPU, one stick of RAM and the video card. Attach a monitor and keyboard and see if it runs OK. This would hopefully eliminate possible case issues and problems that may be caused by less important hardware.

As for possibly crushing the CPU, I find that unlikely. The heatspreader is there for a reason; to protect the core. If you were running an exposed die CPU such as the old Athlons then maybe. I had a Tuniq Tower and while it's rather difficult to mount, the springs allow you to turn the thumbscrews a certain way before it just becomes too hard. My screws were protruding maybe a few MM through the holes under the motherboard which seemed to be OK.

CPU arrived yesterday. Hope to assemble back as a test bed tomorrow and see what happens. I'll keep you guys informed.

Quote: Originally Posted by Durti Sahnchez CPU arrived yesterday. Hope to assemble back as a test bed tomorrow and see what happens. I'll keep you guys informed.

Well, my nightmare continues. Spent all morning on this w/ no luck. It appears that my 2nd EP35C-DS3R has died. I couldn't get the new CPU to POST. I tried another psu as well w/ no luck. I am so tempted to just buy another mb and sell this one when it gets back from Giga-byte. This just sucks because it has everything I was looking for in a mb. It sucks even more not having any parts to swap out for troubleshooting purposes. I have 3-4 weeks with this new box since I built it back in FEB. The rest of the time it been a frickin paperweight. If there's any one of you that reads this and is in my vicinity, I would love some help. Beer and food provided of course!

Have you tried what Reflux suggested and tried the system out of the case? If you have a bad ground, it can give you hell.

Remember, it is almost always the physical layer...

E

Quote: Originally Posted by Fred_G Have you tried what Reflux suggested and tried the system out of the case? If you have a bad ground, it can give you hell. Remember, it is almost always the physical layer... E

Yes. 2 different psu's. Only 1 stick of RAM. Tried no video card 1st

This is one heck of an issue you got there......

Here's my recommendations, i hope this helps you resolve this nightmare of an issue.

1. Go completely barebone......MB, CPU and VIDEO ONLY and test it outside the case.
2. Make sure the 20/24 pin psu connector and the 4 pin 12v connector are plugged in.
3. Clear the CMOS and make sure its set to not jumped. Meaning no jumper on the pins .
4. Do not use that modified psu you got. It could have been the issue from the start.
5. Fire it up and see if it gives any beep codes.
6. Make sure the speaker is plugged in to hear beep codes.


Continuous short beeps: Power error

Im hoping for a 1 long, 1 short beep code. This is at least telling us that the mb and cpu are working together and detecting there's no memory installed. It's not full proof but its a good sign.

If it comes up with NADA then yank the video card and try again. Just the mb and cpu....Any beep codes?

Now if its still completely dead with no beeps at all. It's a crap shoot as to which one is bad. Since there's no way to test the cpu or motherboard in another system to validate its good or bad.

Why dont you give this advice a shot and let me know what happens.

Good luck!!!!!!!!!!

Quote: Originally Posted by Rockin'male This is one heck of an issue you got there...... Here's my recommendations, i hope this helps you resolve this nightmare of an issue. 1. Go completely barebone......MB, CPU and VIDEO ONLY and test it outside the case. 2. Make sure the 20/24 pin psu connector and the 4 pin 12v connector are plugged in. 3. Clear the CMOS and make sure its set to not jumped. Meaning no jumper on the pins . 4. Do not use that modified psu you got. It could have been the issue from the start. 5. Fire it up and see if it gives any beep codes. 6. Make sure the speaker is plugged in to hear beep codes.

Thanks for the reply. I did try this exactly as described, except for the speaker part. I think there's a speaker in an old eMachine out in the garage. I will yank it and try again.

I started the RMA process with Giga-byte, but it looks like, if you include shipping, this process could take up to 5 weeks. Ouch! I don't think I can wait that long.

Well, here we go:

The Giga-byte board is going through the RMA process.

I received my Asus P5K Pro and spent many hours working through setting it up on the bench and multiple OS installs.

Everything appears to working fine now. Here's the deal. When I tried running the old OS (Vista32 Ultimate) with Nvidia's drivers, the MSI NX8800GT 512 OC will wig the screen out. So, I installed XP64. Same thing. MSI doesn't offer a driver for their card in XP64. The Nvidia one gives the screen the shakes too.

So, I tried Vista64 Business. MSI has drivers for this OS and so far the shakes are gone. Everything seems to be running okay, except for the fact that my Ballistix won't run at 4-4-4-12 @ 2.2v on this board like advertised. I have to run 5-5-5-18. No real biggie.

One question though. Is enabling AHCI worth it? If so, how do I do it now that the OS is installed aleady in IDE mode?

Thanks again for the help. The Asus board seems like a nice piece, but I think if I had my PSU mounted up top in the case, routing the ATX 2x12 would have been a bugger.