On Nov 1, 4:08 pm, "Jim Ludwig" <jim...@sbcglobal.net> wrote:
> I'll give this a try. If it doesn't work, do you think I should get new ram
> chips? Also, there are only 2 ram slots. Does it matter which slot the 512
> mb chip sits in?

None of those replies will solve anything. For example, if
resetting memory causes any improvement, then the defect still
remains.

Your problem is why better computer manufacturers provide
comprehensive hardware diagnostics; to find problems before fixing
anything. Anything useful reported by the BSOD (in your case) would
be stored in system (event) logs. The OS detects problems, records
that problem, and works around the problem; so that you might fix it
later at your convenience. What does system logs report?

Reloading device drivers that were working just fine is another
technique by those who grasp at the only things they understand. Why
would good software suddenly go bad - and only during one program?

What causes a system to fail when more peripherals are consuming
more power? DC voltages. Even a defective power supply many months
ago would have booted a computer, but only sometimes create failures;
more often as the failure gets worse. If the computer boots, still
the power supply system could have always been defective. A defect
only made obvious a certain times such as during a game.

That's the background. Start solving by first identifying the
suspects. What do system (event) logs and maybe Device Manager
report?

Get a 3.5 digit multimeter. Exercise as many peripherals as
possible (multitask to sound card, complex video, floppy, internet, CD-
Rom, ... simultaneously). Then measure critical voltages on any one
of red, orange, purple, and yellow wires. Those numbers must exceed
3.23, 4.87, and 11.7 VDC. Those numbers are also posted here so that
the better informed can reply with something useful (no wild
speculation).

If the computer is not from a responsible source (did not provide
diagnostics), then download comprehensive hardware diagnostics from
each individual component manufacturers or third parties. Get and
execute diagnostics (that execute without complications from Windows)
for peripherals such as video controller, sound card, and memory.
Then heat that hardware as hot as a hairdryer can make them - and test
again.

Nothing says to fix anything yet. That would be shotgunning. First
the failure is identified so that the repair cures a problem - does
not cure symptoms. Ie. reseating memory would only be curing
symptoms. Those who actually know hardware would appreciate why.

If memory has failed, well, depending on the failed location, then
sometimes the system will crash OR sometimes an application only
fails. However the manufacturers memory test (or Memtst86) would find
a memory defect either at room temperature or when heated with a
hairdryer. If memory passes under both temperatures, then eliminate
memory from the list of suspects, move on, and "never look back".
Only useful actions result only in a 'definitively good' or
'definitively bad' answer - no speculation.

Anything can appear defective if those DC voltages are not first
taken. Numbers are useful only when all peripherals draw a maximum
load. No way around what the meter reports. Even motherboard voltage
monitor is insufficient. Get that meter. Get definitive answers from
numbers. Get further information by posting those numbers. Any
answer that is not definitive means nothing was accomplished. Above
is how to get only definitive answers - no speculation.