#: 19219 S12/OS9/68000 (OSK)
29-Oct-93 20:57:09
Sb: #19211-Screen program for OSK
Fm: Bob van der Poel 76510,2203
To: robert brose 72067,3021
Robert, funny about screen not working for you. I just set up a couple of
screens and got basic running in both. Loaded some programs and did a dir and
all worked just fine.
#: 19221 S12/OS9/68000 (OSK)
30-Oct-93 14:03:11
Sb: #MM1-68340
Fm: John R. Wainwright 72517,676
To: all
I got one of the MM/1 68340 accelerator boards at the
Atlanta 'Fest. Zack Sessions has already posted an
excellent technical description and review of the upgrade,
but I thought I would add a couple notes on my own
experience with the faster system - and a couple more
"gotcha"s.
I did not have the "slow" PAL chips on my I/O board, but
just when I thought I was safe, I discovered some on the
8-meg memory board. Replacing them was not as hard as I
thought it would be, but not a job for someone without
experience in repairing electronics. Like a famous coach
said about passing is a football game -- "Three things can
happen - two of them are BAD".
It is, of course, FASTER. Thats why we bought it. The
TeX and PostScript (and Ghostscript) programs I have been
playing with lately take a lot of time. Building a new font
for TeX ran about 15 minutes before, now its about 7. I
don't have a "dhrystone" program (or even a whetstone), but
I can easily see the increase in speed. I did run the
little basic "benchmark" program that Frank Hogg posted some
time ago, and the 68340 came in right about where you would
expect it.
Results of Frank's basic benchmark
(the program)
SHELL "date -j"
FOR j=1 to 12
FOR i=1 to 1000
a=SQRT(i)
b=INT(a)
IF a=b THEN a=b
ENDIF
NEXT i
NEXT j
SHELL :date -j"
(Frank's posted results)
Year Model Speed CPU FPU Seconds
'92 KiX\30 33 Mhz 68030 Yes 2
'92 KiX\30 25 Mhz 68030 Yes 3
'92 KiX\30 16 Mhz 68030 Yes 5
'87 QT20x 12 Mhz 68020 Yes 12
'92 KiX\30 33 Mhz 68030 No 14
'92 KiX\30 25 Mhz 68030 No 19
'92 KiX\30 16 Mhz 68030 No 29
'90 TC70* 15 Mhz 68070 No 105
'85 QT+ 10 Mhz 68000 No 105
'84 QT 8 Mhz 68008 No 219
* On board memory only.
(my MM/1)
mm1 68340 16.59 68340 N/A 35
It boots from the hard drive. Twice it has stumbled a
little on a cold start when it looked for the drive before
it was up to speed, but it got it on the second try. Both
of my hard drives, a 40 meg conner and a 40 meg quantum work
fine as "/dd".
Zack said he got a new "shell" -- I didn't. I have the
same shell I had before. Also, "dEd" runs fine - but that
is probably because of the different shell. I asked Kevin
Pease about that new "shell" - he said he didn't know where
that came from. Apparently Zack got an accidental "gift"
from somewhere.
Windows have a little quirk - changing colors on one
window does it to all of them. Kevin Pease has a fix,
provided by a well known OS9 Guru - he said he would upload
it very soon.
I have heard of problems with MM/1's talking to
printers, especially with long cables, but I was getting
away with it up to now. I had ribbon cables from the MM/1
and a big ugly "386" running to switch boxes, and parallel
cables from the boxes to the Epson and the Deskjet so I
could switch either printer to either computer. --- No
more. With the faster CPU, the MM/1 wants a short cable
directly to the printer. (I can live with it).
The reason I haven't said a lot in the forum about my
new toy is that it had this little problem -- every once in
a while, especially after a cold start, it would trash the
hard drive. It seems there was this little timing problem
that required some attention from Kevin Pease. The good
news here is the response I got on the problem. I bought
the upgrade at the "Blackhawk" booth in Atlanta, which was
manned by Bill Whitman (another IMS rep). When I decided
something was wrong, I called Bill (put a message on his
machine), and left CIS email for Kevin Pease. BOTH of them
called me at home within a day. Both had suggestions and I
tried them all - but finally had to send the insides of my
machine to Kevin. He tracked down the problem, made a new
PAL to correct it, called to let me know he had it fixed,
sent it back, and called me again today, just to make sure I
had it running OK.
As Delphi and CIS OS9 forum regulars know, response from
IMS has, in the past, been.. well... less than
outstanding. Paul Ward has not appeared on CIS or Delphi
for many moons. ((I think Frank Hogg ran him off. Frank
was mean to him. Look ashamed, Frank.))(GRIN)
David Graham (Blackhawk) (NIMITZ) has obviously been
trying to correct the situation. Although he was not at the
Atlanta 'fest himself, David left his booth in good hands.
I am very pleased with the 68340 and the quick technical
support.
Like Zack, I feel my money was well spent. Get one.
Check with David Graham (NIMITZ) for current price and
delivery time.
John R. Wainwright
<<CIS -- 72517,676>> <<DELPHI -- JOHNREED>>
*********** InfoXpress ************
There is 1 Reply.
#: 19222 S12/OS9/68000 (OSK)
30-Oct-93 15:16:39
Sb: #19221-MM1-68340
Fm: Zack Sessions 71532,1555
To: John R. Wainwright 72517,676
> machine to Kevin. He tracked down the problem, made a new
> PAL to correct it, called to let me know he had it fixed,
> sent it back, and called me again today, just to make sure I
> had it running OK.
Hmm, was that U18 on the IO board by any chance? Reason I ask was that I
received just today, a replacement PAL for U18, from Bill Wittman. He states in
his letter that "on some systems, not many" require the new PAL to help solve
"hard drive problems". Even though I am not having hard drive problems (unless
you consider not being able to boot from the hard drive as a "problem"), "Pease
highly recommends that you put it in".
And I concur, support from Kevin Pease and Carl Kreider has been excellent.
------------------------------------
Zack C Sessions
ColorSystems
"I am Homer of Borg, prepare to be assimi ... OOOOHHH, DOUGHNUTS!"
Press <CR> !>