Can anyone explain what's going on with this dual 2.0 G5???

Question

I've been trouble shooting my PowerMac G5 dual 2.0 recently. I've gotten to the point where I think it's the processors. I was getting random freezing and Kernel Panics so I bought a set of used 2.0 CPUs (U1 and U2)in unknown working condition. Well, I took out the original CPU (O1) that I though was causing the trouble and replaced it with one of the untested ones (U1).Everything worked great except that the computer only recognized one CPU (don't know which one). Swapped out U1 for U2 and again it powered right up but froze after a few minutes. So I replaced U2 with U1 and went about using it as a single G5.
I later aquired a 100% tested and working 2.0 CPU (T1) and popped it in with O1. I got nothing. No chime or start up. I switched their positions and got as far as trying to calibrate the fans only to repeatedly get an error saying the processor was bad.
After this I tried something I hadn't thought of before. I put U1 and U2 together and to my surprise, the computer started. It was working with both CPUs once again. The only thing I noticed was that CPU A's temp was about 103 degrees F and CPU B's was 118 F. There was always about a 15 degree difference. The computer ran fine for a while so I opened 6 full length movies in Quicktime to stress the CPUs. After a few minutes of running and expose work, the computer froze. I didn't see what the temps were when it froze, I just did a hard shutdown and left it off.
My point is, I'm starting to wonder if it's the logic board and not the CPUs. Both O1 and O2 work together and are recognized as dual, but the computer will freeze. O1 and U1 work together but are recognized as a single 2.0 G5. U1 and U2 work together and are recognized as dual but, again, the computer froze. I couldn't get T1 to even register, but I've been told Rev A and Rev B CPUs cannot be mixed so maybe that's why.
Does anyone have any ideas? Two (or more) heads are better than one and I'd love to figure this out. I hope this doesn't sound too confusing. I'll give any additional info if needed. Thanks in advance.
Message was edited by: fiveinchwrench

Answer 1

Hey DaddyPaycheck and Japamac,
Thanks both for your replies.
+You are a tenacious one, that's for sure! :-)+
I don't know if that's a good thing in this case.....;)
I do agree with the statement that it cannot handle dual processors. I swapped the the "original" working CPU with the "original" one I thought was giving me problems. It is now running (although showing as single) with the CPU I thought was bad. Both sets, original and untested, will work together and register as dual, but will be unreliable and randomly freeze.
+The defining bit of data would be if you could determine that one socket or the other is continually working, or if one is sporadic (thus explaining the occurrence of the pair working).+
I'm having trouble narrowing down that part. I can swap them back and forth and get the same result-so both sockets seem to be working. But like I said when both pairs are couple together, it shows as dual, and it works for a certain amount of time. So I know they are working to a certain degree. But when they're mixed, only one (don't know which one) is being seen.
Yes, I've run thermal calibration and everything goes fine. Even with the diagnostic disc, when the CPUs are mixed, hardware specs show one CPU. But when the original pairs are coupled, it will show both.
This is like a splinter that I can't get out. It runs great as one CPU, and I could let it go like that, but knowing that it could (and should) be dual is driving me nuts.
Is there anything either of you can think of that I can try?
Message was edited by: fiveinchwrench