GA-P35-DS3R Boot Problem

[the oven]

I have had this system running for about 3 years now.
I've always had hit and miss luck with overclocking it

Yesterday I tried again, adjusting oc, and also the memory timings and voltage

I have used these settings before but could have put the wrong info

The system started to reboot after saving settings, then powered down and went to reboot but with no signal to monitor


This used to happen before when i would OC and try to reboot, it would cycle trying to boot and never make it


I used to unplug it for a few min and it would start back up

no luck this time

I tried resetting the cmos several times
removed battery
used jumper
even let it sit for a few hours with everything unplugged except the proc

still will not post

All the fans and such spin up for about 3 sec then it powers down, then back up and just sits there, with all it's fans spinning

no signal to monitor

I don't have any other parts or system I can hook it up to

I don't have a internal speaker, would that be worthwhile to get?

would the beeps tell me what is wrong?


If I cleared the cmos why wont it start up?
could the memory have gotten fried?

 

[bilbat]

Have you tried the exact process in the "...to break a 'boot-loop'" section of the 'sticky'?

 

[the oven]

Have you tried the exact process in the "...to break a 'boot-loop'" section of the 'sticky'?

was going to try that next, but it doesn't loop it just sits there

how vital is a internal speaker for correct diagnosis of problems?

can I cannibalize an old dell for a speaker?

 

[bilbat]

Check the pin header - really older boards used a two pin header, where the pins were right next to each other. Your speaker pins have two 'empties' in-between 'em, on a four pin header. If you have the old style plug, and a stock of 'leftover' single pin plugs, a little judicious soldering will fix 'em...

Another thing: might just be time to replace the CMOS battery - they don't last forever (though, CMOS draws so little power that the 'service life' of a battery is pretty much equal to the battery's shelf-life...)!

 

[the oven]

Check the pin header - really older boards used a two pin header, where the pins were right next to each other. Your speaker pins have two 'empties' in-between 'em, on a four pin header. If you have the old style plug, and a stock of 'leftover' single pin plugs, a little judicious soldering will fix 'em...

Another thing: might just be time to replace the CMOS battery - they don't last forever (though, CMOS draws so little power that the 'service life' of a battery is pretty much equal to the battery's shelf-life...)!

gotta look at the speaker in the old system and see what it has


I got a brand new bat and it still won't post

does the double boot and then sits, no signal to monitor, all fans spining


maybe frys has a speaker I can get

 

[bilbat]

Here's what I'm thinking so far: the 'double-boot' is pretty much a standard thing for GBs; if you change some BIOS parameter (and I don't know which ones - some don't seem consistent - maybe it's a issue of "change X by more than Y amount?" - but I know static tRD is one - any change to tRD will make the board 'double boot'), the board seems to 'partially post', 'notice' the changes, shut down for an instant (maybe re-calculating some 'auto' parameters?), and then boot again. If that second boot does not succeed, it will boot a third time, this time resetting itself to 'Load Fail-Safe' parameters... Occasionally, that third boot will also 'go awry' resulting in really strange operation. I'm trying here to think of what the possibilities might be - I'm guessing you're getting that double boot, but for some reason, it's getting 'stuck', and not doing the third, 'restorative' boot! I've said it here before, and I'm sure I'll say it many more times: "I'd kill to get my hands on source code for a modern BIOS!"

I could just cry for the days when I started doing this Had a Z-80 KayPro, and ran ZCPR instead of CPM. You coud get complete schematics forevery component and IC on the board, right down to the terminating resistors; and ZCPR used a custom BIOS extensively - there was a 'patch area' in the OS/BIOS, so if you needed some obscure function in the OS, you just coded it, loaded your patch in, and called the vector to it! Everyone who was 'serious' about it had an ultraviolet PROM eraser, and a handfull of 'favorite' modded BIOS - and used to 'trade' the patches like baseball cards! One of the favored activites was to give a buddy the assembler source for a patch as a challenge - "betcha can't guess what this one does!?!"

Just as a check - you're sure you haven't 'forgotten' some USB gizmo that's remaining plugged in? Great amusement once here, when someone got a case of the dreaded 'reboot loops', and we tried to figure out for a day or two, what could be the cause? He had had an external backup drive plugged into USB for months, so we never suspected it! Turned out it had been plugged in for months without the power switch on the unit turned on!! Unbeknownst to him, his daughter had flipped the switch, powered it up, and it had 'toasted' his CMOS [:isamuelson:8]

 

[the oven]

Here's what I'm thinking so far: the 'double-boot' is pretty much a standard thing for GBs; if you change some BIOS parameter (and I don't know which ones - some don't seem consistent - maybe it's a issue of "change X by more than Y amount?" - but I know static tRD is one - any change to tRD will make the board 'double boot'), the board seems to 'partially post', 'notice' the changes, shut down for an instant (maybe re-calculating some 'auto' parameters?), and then boot again. If that second boot does not succeed, it will boot a third time, this time resetting itself to 'Load Fail-Safe' parameters... Occasionally, that third boot will also 'go awry' resulting in really strange operation. I'm trying here to think of what the possibilities might be - I'm guessing you're getting that double boot, but for some reason, it's getting 'stuck', and not doing the third, 'restorative' boot! I've said it here before, and I'm sure I'll say it many more times: "I'd kill to get my hands on source code for a modern BIOS!"

I could just cry for the days when I started doing this Had a Z-80 KayPro, and ran ZCPR instead of CPM. You coud get complete schematics forevery component and IC on the board, right down to the terminating resistors; and ZCPR used a custom BIOS extensively - there was a 'patch area' in the OS/BIOS, so if you needed some obscure function in the OS, you just coded it, loaded your patch in, and called the vector to it! Everyone who was 'serious' about it had an ultraviolet PROM eraser, and a handfull of 'favorite' modded BIOS - and used to 'trade' the patches like baseball cards! One of the favored activites was to give a buddy the assembler source for a patch as a challenge - "betcha can't guess what this one does!?!"

Just as a check - you're sure you haven't 'forgotten' some USB gizmo that's remaining plugged in? Great amusement once here, when someone got a case of the dreaded 'reboot loops', and we tried to figure out for a day or two, what could be the cause? He had had an external backup drive plugged into USB for months, so we never suspected it! Turned out it had been plugged in for months without the power switch on the unit turned on!! Unbeknownst to him, his daughter had flipped the switch, powered it up, and it had 'toasted' his CMOS [:isamuelson:8]

thanks for all the help

Right now nothing is plugged in, no keyboard or mouse
the hard-drives are unplugged too


tried with only 1 ram stick in each slot and nothing
tried other, same thing

is the boot loop the same thing I have happening?
it only does it twice and the second time just keep running
forever
never tries to boot again
no signal to monitor, or old monitor

<edit>old computer has some ancient looking round thing that does not look compatible with newer system

 

[the oven]

ok

so I went out and bought a motherboard speaker and other then a small click sound when I first try and boot the system, the thing is silent

does not beep even once

what now?

defective speaker?

 

[bilbat]

Now that you have the speaker, try booting it without any memory - that should make it 'complain' (1 long, 1 short beep); that test is done early enough in the POST so we should be avoiding whatever else is 'hanging it'... There is a way to check the speaker - do you have access to any 'bare wires' on it? If it's a piezo (small cylinder 'bout ½" by ½" - ~12mm x 12mm), mine always come with the wires soldered to the back, which is open - and I always wind up putting a bit of electrical tape on the back, as I figure - Murphy's Law, it will eventually 'relax' the wires into the position where not only will the grounded side short something, but it will fall into the position from which it can inflict maximum damage!

 

[the oven]

Now that you have the speaker, try booting it without any memory - that should make it 'complain' (1 long, 1 short beep); that test is done early enough in the POST so we should be avoiding whatever else is 'hanging it'... There is a way to check the speaker - do you have access to any 'bare wires' on it? If it's a piezo (small cylinder 'bout ½" by ½" - ~12mm x 12mm), mine always come with the wires soldered to the back, which is open - and I always wind up putting a bit of electrical tape on the back, as I figure - Murphy's Law, it will eventually 'relax' the wires into the position where not only will the grounded side short something, but it will fall into the position from which it can inflict maximum damage!

tested speaker and it works fine

tried with no ram and it does not make a peep

that mean the mobo is dead?

 

[bilbat]

Well, I think you're down to one of two items: the motherboard, or the CPU...

Almost any kind of problem should result in beeps of some kind; two PSU possibilities - if it's completely dead, it'll be easily observed, as nada - no fans, no LEDs; if it's got any kind of serious, but non-fatal problem, it will almost always result in successive reboots, or beeping...

If the CPU's gone, there's 'no one home' to run the BIOS, so you get - nothing. If the MOBO's got a problem serious enough to not let it run the CPU, at all, you get - nothing. I think that's pretty much where we're at... If you can get your hands on a different 775 to test with, it's probably worthwhile - would confirm it's the CPU. If not, my money's on the motherboard, simply as there's so much more to go wrong there, than in the CPU. If it wasn't excessively overclocked (especially overvolted), and all the fans were working, CPUs should last a long time - I've got some ancient P4's around that have been running forever!