Looks like the latest Memtest86+ might kill BIOS!

[bluemax]

The story so far: I was working on an in-house machine which up until now was working perfectly. I shut everything down, powered off the PSU, left it plugged in. Changed PC133 RAM to one stick of DDR for testing purposes (yes, one of those boards with both RAM types.) Memtest86+ found about 5000 errors on the one stick of RAM in about 15 seconds so I decided, "yes, that's a dead stick-o-RAM."

Instead of hitting ESC to reboot, I simply powered down the machine.

Pulled RAM, replaced it with another DDR for continued testing.... black screen. Won't beep. Won't post. No beeping error codes for no RAM or no video.

Yes, I was sure to touch all kinds of metal thingys before touching anything computery - static is bad.

I'm only day 3 on the job, but I've been in hardware for 10+ years! I can't let something like this ruin a good job just because of some lousy luck!

Any ideas?

 

[SlitheryDee]

What brand/model MB was in the comp?

 

[Hyperlite]

ya know, this makes me think. My initiall response would definantly be no. BUT. i have seen more and more threads with people asking if Memtest could have killed their mobo or ram....i'm begining to think its possible, somehow or another. i would defiantly be interested in further investigation into this, but i doubt there will be many volunteers. (for obvious reasons.)

 

[bluemax]

Originally posted by: SlitheryDee
What brand/model MB was in the comp?

ECS P6S5AT, P3-750

 

[ViRGE]

Originally posted by: Hyperlite
ya know, this makes me think. My initiall response would definantly be no. BUT. i have seen more and more threads with people asking if Memtest could have killed their mobo or ram....i'm begining to think its possible, somehow or another. i would defiantly be interested in further investigation into this, but i doubt there will be many volunteers. (for obvious reasons.)

The thing is how would MemTest86 kill a machine? It doesn't write to any drives, it doesn't overclock the system or change any voltages, and it doesn't try to flash your BIOS; what possible vector would it have that it could kill a machine with?

 

[Hyperlite]

i know, it doesn;t make any sence....its just coincidence i suppose...

blumax did you try a new PSU?

 

[bluemax]

Originally posted by: Hyperlite
i know, it doesn;t make any sence....its just coincidence i suppose...

blumax did you try a new PSU?

That's my last option. I've pulled this machine to pieces, right down to the board on the desk, PSU hooked up outside - nothing but a video card and a known-working stick-o-RAM. Nothing. Not even a post.

Either the BIOS is dead, or the PSU is. And I'm leaning BIOS, as this has now happened twice.

Maybe it's because of this new version of Memtest86+ less than a month old? It MUST do some BIOS functions as it's able to relay on-screen just what motherboard chipset and northbridge is being used.... but why would it KILL the BIOS?? It makes NO SENSE! And yet the evidence is in front of me inthe form of TWO dead motherboards! Both from the same thing - memory testing with the new Memtest86+!

This is not a good sign, and I'll be switching to another RAM tester IMMEDIATELY! This is going to reflect on my technical abilities and could cost me my new job!

 

[bluemax]

Tried new PSU. Yep. BIOS is dead.

 

[allanon1965]

there must be another common denominator here.....i cant see how its possible for memtes to kill a bios, all it does is test the ram, its a small dos-like based program....bios is merely the instructions for the computer to get thru post and into the operating system....its either a bad coincidence or a problem with two different boards, personally i have never had any good luck with older ecs boards, did you look to see if by chance there were any swollen caps? not saying 100% that it is not memtest, but i have used it for a longtime as well as the new version, with no problems yet.....just trying to toss out other possibilities...good luck... btw, never leave a system plugged in while you change out components, thats the first thing they taught us in pc repair class....always unplug the power cord and make sure you are discharged of any static electricity either by touching the metal case or by a strap on your wrist....

 

[ViRGE]

Originally posted by: bluemax
It MUST do some BIOS functions as it's able to relay on-screen just what motherboard chipset and northbridge is being used....

It recognizes the chipset and mobo because those devices have equipment ID's, which MemTest then knows what the IDs mean. IIRC, MemTest is built upon a compact Linux kernel, which makes this all the more confusing since it would be going through the OS to get to the BIOS.

 

[bluemax]

Originally posted by: allanon1965
there must be another common denominator here.....i cant see how its possible for memtes to kill a bios, all it does is test the ram, its a small dos based program....bios is merely the instructions for the computer to get thru post and into the operating system....its either a bad coincidence or a problem with two different boards, personally i have never had any good luck with older ecs boards, did you look to see if by chance there were any swollen caps? not saying 100% that it is not memtest, but i have used it for a longtime as well as the new version, with no problems yet.....just trying to tos out other possibilities...good luck...

Checking capacitors now.... I wish it were the case, as it would be beyound a doubt NOT my fault! But I don't see how this could have happened.... All I did was power down to change RAM sticks, then *POOF*! Okay... not "POOF"... that would imply a sound other than spinning fans along with ghastly, deadly silence.

Caps look OK... bunch of big greenies, nothing scorched or really abnormally-shaped.

 

[SlitheryDee]

Have you tested all the memory you've been using in other computers?

 

[NokiaDude]

Did you even try a PC133 stick of RAM?

 

[aatf510]

Originally posted by: NokiaDude
Did you even try a PC133 stick of RAM?

 

[Navid]

Doesn't your motherboard allow you to remove the battery and set a jumper to reset the bios?

 

[Spikesoldier]

Originally posted by: toattett

Originally posted by: NokiaDude
Did you even try the original PC133 stick of RAM that was working before?

fixed

and also did you make sure that theres some kind of special jumper that you have to short or dip switches when you change from SDR to DDR?

 

[Bozo Galora]

actually memtest86 does probe the bios for memory maps
and older bios do not support new probing
you are using a PIII 750 so that means old bios
I would reflash the bios
and use vers prev to 2.9 for memtest

Quote:
Memory Sizing
The BIOS in modern PC's will often reserve several sections of memory for it's use and also to communicate information to the operating system (ie. ACPI tables). It is just as important to test these reserved memory blocks as it is for the remainder of memory. For proper operation all of memory needs to function properly regardless of what the eventual use is. For this reason Memtest86 has been designed to test as much memory as is possible.
However, safely and reliably detecting all of the available memory has been problematic. Versions of Memtest86 prior to v2.9 would probe to find where memory is. This works for the vast majority of motherboards but is not 100% reliable. Sometimes the memory size detection is incorrect and worse probing the wrong places can in some cases cause the test to hang or crash.

Starting in version 2.9 alternative methods are available for determining memory size. By default the test attempts to get the memory size from the BIOS using the "e820" method. With "e820" the BIOS provides a table of memory segments and identifies what they will be used for. By default Memtest86 will test all of the ram marked as available and also the area reserved for the ACPI tables. This is safe since the test does not use the ACPI tables and the "e820" specifications state that this memory may be reused after the tables have been copied. Although this is a safe default some memory will not be tested.

Two additional options are available through online configuration options. The first option (BIOS-All) also uses the "e820" method to obtain a memory map. However, when this option is selected all of the reserved memory segments are tested, regardless of what their intended use is. The only exception is memory segments that begin above 3gb. Testing has shown that these segments are typically not safe to test. The BIOS-All option is more thorough but could be unstable with some motherboards.

The third option for memory sizing is the traditional "Probe" method. This is a very thorough but not entirely safe method. In the majority of cases the BIOS-All and Probe methods will return the same memory map. For older BIOS's that do not support the "e820" method there are two additional methods (e801 and e88) for getting the memory size from the BIOS. These methods only provide the amount of extended memory that is available, not a memory table. When the e801 and e88 methods are used the BIOS-All option will not be available. The MemMap field on the display shows what memory size method is in use. Also the RsvdMem field shows how much memory is reserved and is not being tested.

http://www.memtest86.com/#philo

Edit: and 2 diff types of RAM availibility might really throw memtest for a loop

 

[SlitheryDee]

Originally posted by: Bozo Galora
actually memtest86 does probe the bios for memory maps
and older bios do not support new probing
you are using a PIII 750 so that means old bios
I would reflash the bios
and use vers prev to 2.9 for memtest

Out of curiosity, how do you reflash the bios on a computer that won't POST?

Edited: For easier reading.

 

[allanon1965]

very interesting, i was gonna read up on memtest later tonight, looks as if now i wont have to, thanks......but i probably will anyway

 

[bluemax]

Originally posted by: NokiaDude
Did you even try a PC133 stick of RAM?

Yes, and yes to checking RAM in other machines. Trust me, all options have been exhausted. The machine won't even give you the error beeps for "NO RAM".

 

[bluemax]

Originally posted by: SlitheryDee
Out of curiosity, how do you reflash the bios on a computer that won't POST?

That's what I'd like to know!

 

[VirtualLarry]

Originally posted by: bluemax

Originally posted by: SlitheryDee
What brand/model MB was in the comp?

ECS P6S5AT, P3-750

LOL. I put together a system with that board. Finicky as all heck. In one case, the reset button wouldn't work, even though the switch worked fine. In a different case, it did. No extra standoffs shorting anything - I checked. I put it down to "board flex", when mounted into the case. The board is disturbingly thin. I had one appear to "die" on me, thus necessitating a replacement board, and then it mysteriously started to work when tested again some time later.

Memtest86+ didn't kill your board. The problem is that it's an ECS/PCChips board.

Edit: Btw, if one wanted to easily test this theory, then all you would need to do, is take the flash BIOS chip off of a "dead" board, and "hot swap" it onto another board (like in preparation for a "hot-flash"), but instead of flashing it, dump the contents to a .BIN file instead, and then compare that against a known-good BIOS .BIN dump. That would allow proving that theory, if for some reason, some of the chipset-programming code in Memtest86+ actually could be doing what you suggest. But based on personal experience, I would find that to be far less likely. You might try simply removing the CMOS battery for several hours / a day, and then see what happens, it might come back to life...

 

[Bozo Galora]

flashing bios to a PC that wont post is easy - ASSUMING you can boot to floppy!!!! (no hope with DEAD mobo - dead = no floppy controller)
its called blind flash
You set bios to boot floppy first
You put bin or rom and flasher on floppy
insert floppy and boot - see if you get recongition light action
then you carefully type in (blind) the A:> flasher command needed and watch the floppy light carefully
there will be lots of activity, then when that stops, it is asking you if you want the file to be flashed, so hit enter, watch light until it stops, do you want to save old, watch light, enter etc.

if you load correct flasher and file it will work, done it lots of times.
some bios flashers already have autoexec feature

and if that doesnt work, add files onto a Win98 boot disc, if you dont have enough room, remove ebd.cab

Edit: Googled this up:

PROBLEMS ENCOUNTERED AFTER FLASHING BIOS: If no display appears after flashing the BIOS, but the beeping sound from the system can be heard and the floppy disk drive is activated, then all is not lost. Try the following method to recover the system:

Prepare a bootable floppy disk with an AWDFLASH.EXE utility and BIOS binary file in the disk (you can use the prepared diskette mentioned in Q3).
Create an AUTOEXEC.BAT file in the diskette with the following command line:
AWDFLASH xxxxx.BIN/SN /PY
The options /SN /PY will answer the questions automatically that you would normally answer with "no" or "yes". That way the system will boot up from the floppy disk, program the BIOS and restart the computer It will take a while to carry out the procedure, don't turn off the system, wait for the system to restart again.
If you want to see what it does, use an ISA VGA card instead of PCI or AGP. If you don't have any ISA VGA card, the BIOS can still be fix in blind mode (no screen, but the BIOS is flashed by the Auto-run diskette). After the mainboard reboots itself, the floppy disk can be taken out and your system should have recovered. However if that doesn't work, return the mainboard to your Vendor. Then you may resume BIOS update.

http://www.qdigrp.com/qdisite/eng/support/f_BIOS_f.htm

And FWIW - I wouldnt use an ECS board if they gave them out free. BUT - that doesnt help you in your current situation.

 

[bluemax]

Originally posted by: VirtualLarry
LOL. I put together a system with that board. Finicky as all heck. In one case, the reset button wouldn't work, even though the switch worked fine. In a different case, it did. No extra standoffs shorting anything - I checked. I put it down to "board flex", when mounted into the case. The board is disturbingly thin. I had one appear to "die" on me, thus necessitating a replacement board, and then it mysteriously started to work when tested again some time later.

Memtest86+ didn't kill your board. The problem is that it's an ECS/PCChips board.

You may have something there! The board did feel a bit "bendy" and I had to push pretty hard to get that stick of DDR into the slot (no, not backwards!) It was all stiff from never having been used with DDR before, like the ISA slots of old - super tough to use the first time, but became easier with each insert....... hmm..... sounds naughty.

Chalk it up to bad luck and crappy board, but it looks bad on me as a technician in my first week on the job.... Thanks to everyone with your advice!

 

[shoRunner]

i really doubt memtest kills the bios i have literally used it 500+ times with out ill effects, not once.