Abit AI7....

[kc8adu]

Re: Abit AI7....

Oddly enough, we've got a Averatec 3150 laptop at work that has been exhibiting the same types of errors ( stop 50 and 8E). The thing with the laptop though, is that sometimes only half the added SODIMM is recognized. So instead of POSTing with 640M of RAM, it POST's with 384. I did an advanced RMA on the ram through Kingston, and received it today, so we'll see if the new DIMM fixes the issue or not.

The funny thing about the laptop was that Memtest86 worked perfectly overnight, as well as prime95.. I was just looking at it one time and then i remembered that it has 128M soldered down, as well as one SODIMM slot, which has a 512M module. That's how i figured the SODIMM has a quirky fault that makes only half of it work sometimes.


If you have a couple of other sticks of RAM to try, or just try one stick at a time. Maybe try bumping Vmem +.1V or so..

had an averatec that acted like that.had bad soldering on the sodimm socket.
resoldering fixed it.
see this on a bunch of sony laptops.

[Rainbow]

Re: Abit AI7....

I've seen the same on Compaq.

[gg1978]

Re: Abit AI7....

had an averatec that acted like that.had bad soldering on the sodimm socket.
resoldering fixed it.
see this on a bunch of sony laptops.

Oh joy.... So basically that means that the socket is only making connection to one bank of the DIMM instead of both... As i recall, it appeared that to resolder the connector would mean the motherboard would have to come out, as the socket is to one side of the opening. Well, when the GM's laptop starts BSOD'ing again, at least i'll know what's actually wrong with it..

[skiddz]

Re: Abit AI7....

Send me the minidump file from the crash saved in %windir%\Minidump

Sometimes WinDBG can be quite useful in diagnosing this stuff...

No such animal on my machine. Most BSODs totally hang the system and require a power-off restart - can't even mash the reset button to get it to reboot.

[Per Hansson]

Re: Abit AI7....

No such animal on my machine. Most BSODs totally hang the system and require a power-off restart - can't even mash the reset button to get it to reboot.

If you get a BSOD the file should be there...

Check under Control Panel > System > Advanced > Startup And Recovery;

Make sure that under "write debugging information" this is selected "Small memory dump (64KB)"

---

The reset button should always work, it is a 100% hardware feature, the OS is not asked for any confirmation at all, the reset button is actually interfaced directly into the systems CPU...

The only way I can imagine that it would not work (I mean if it normally does when the system is running or booting i.e. have not crashed) would be if the CPU is so overheated that it does not receive the signal, or rather "see" the signal...

This has commonly happened to me when I have overclocked systems too far; but it still tries to reboot, I just do not get any POST because the CPU can not work at the selected frequency...

[skiddz]

Re: Abit AI7....

Not a heat issue unless Motherboard Monitor and the BIOS's System Health info are lying to me. I've never seen any sensor temp top 42C, even when rendering or playing games. Machine was overclocked until the problems started (p4 2.8GHz @ 3.15Ghz - no voltage increases required) when I set it back to 'stock' and it's been there since.

I'll look for the minidump file.. One of my other machines puts it in C:\Documents and Settings\[username]\Local Settings\Temp\[some_odd_named_folder]\

so I'll look there...

[skiddz]

Re: Abit AI7....

Not a single minidump file on my entire machine. Settings say it should be in %systemroot%/minidump.

When the machine BSODs, it locks up hard and requires a power-off reset.. Expecting my caps from Mouser today so we'll see what recapping does...

[RoadKill]

Re: Abit AI7....

I have an Abit AI7, infact, this is my second.

My first had united chemi-con and rubycon, this second, all rubycon.

The first only failed because of my stupidity, I accidentally scratched a track that lead to the RAM.. with a screw.

I'm happy with the board, but ocz ram running in dual channel gives me instabilities, but my old, random assortment of RAM ran fully stable even at much overclocked FSBs.