Brett, I have two little PC programs to disable and enable the Pentium internal
cache. With the
internal cache disable, the Pentium 350 machine runs like a 80386 SLC40 :)...  i
f
you are interested I can send them to you.  Some old hardware like eprom program
mers that runs
timming loops got totaly confused and don't work with new and fast mahcines, so
I
need to disable the internal cache.

Wagner Lipnharski
http://ustr.net

> Brett Paulin wrote:
>
> Greetings Piclisters,
>
> a question for those of you with new fast PC's. (or knowledeable in such thing
s)
>
> Recently, I dusted off my Pic Programmer board.. (home made, a variety of Tait
, DonMcKenzie/
and others hybrid) that was once working very happily on my old Pentium 100 mach
ine,
> and attempted to program some 16F84's..  no go
>
> EVERY program I can find that supports the parallel port programmers produces
the same
result.  Verify Failure at address 0, Config Fuses verify failure.
>
> Manually testing the hardware with Nigels Windows PicProg2 in debug mode shows
 all the lines
to be toggling and doing their thing correctly.  VDD, VPP, clock data etc all be
have
> as expected.   No apparent damage to programmer, Voltages seem correct.   But
when a program
or verify is attempted, it fails
>
> after a failed program cycle, read shows a single word being changed to a psue
do random value
>
> Erase appears to work, in that after erasing, the single incorrect word revert
s to 3FFF
>
>
> The only thing I can come up with is that with my recent upgrade to a Celeron
running at
450Mhz (and yes it works fine for everything else), something is too fast for th
e timing
> requirements of the PIC.  but seeing as how a friend has borrowed my 'scope at
 the moment, I
cant verify this right now.  I tried slowing the CPU back down to 300Mhz (its no
rmal
> speed), with no improvement in results.
>
> One side issue is that with old P100 machine, The programmer didnt appear to w
ant to work on
a Card-based Parallel port but was quite happy on the onboard port..  similair
> symptoms.. static line tests showed all signals running OK, programming failed
.   I've tried
it on both the Motherboard and card based Ports on the new machine with no chang
e in
> result..
>
>
> Thats about all I can tell you right now.   Can anyone let me know if there ha
ve been
similair problems with high speed machines and pic programmers  ??
>
> And a solution would be really nice too.. (Of course). B-)
>
> Thanks muchly.
>
> ---
> Brett Paulin : Trance DJ Spock, Electronic Engineer & Gyrocopter Pilot