🚀 go-pugleaf

After several years of suckling at the teat of Mother Big Blue, we had 
accumulated an interesting collection of spare parts in our humble little 
laboratory. These parts consisted of cast offs and hand-me-downs from users 
who went on to bigger and better things. One day an interesting idea dawned in 
the mind of one of the workers in this vineyard of science. "Let's take all 
these spare parts," he said excitedly, " add just a little bit of money and 
create a PC Clone."
 
Like Dr. Frankenstein's demented assistant, Igor, we accumulated the boards 
and other paraphenalia that would be stitched together to form a new bred of 
computer in our midst. The following pieces were retrieved from various 
graveyards:
       
             * Amdek 300A amber monitor 
             * IBM  Enhanced Graphics Adapter
             * Keytronic KB 5150 keyboard
             * IBM serial card
             * IBM parallel card
             * Shugart half-height floppy
             * CDC half-height floppy
 
Like many others, we had drolled over the ever lowering prices for clone parts  
that we saw in the back pages of Byte. With grand confidence and our fingers 
crossed we placed an order with JDR Microelectronics for the following parts:
 
             * XT compatible motherboard -- $129.95
             * Floppy disk drive adaptor --  $34.95
             * Pro-BIOS                  --  $19.95
             * 150 WATT Power Supply     --  $79.95
             * Flip-top case             --  $39.95
             * 256K RAM                  --  $26.55
                                    Total = $331.30
 
Unlike Frankenstein's monster which never had a price tag attached to it, our 
little creation would be quite a bargain. The P.O. was cut and it was with 
anxious anticipation that we received a rather large box several weeks later.
 
An eager and disgustingly bright work-study undergraduate was assigned the 
task of assembling the clone. The motherboard was populated with RAM and ROM 
and mounted in the sturdy-looking case. The power supply was bolted in place 
and plugged in. Without any ceremony the switch was thrown. Nothing happened! 
The rather meager instruction we received from JDR (the nth xerox of a mangled 
master) said that the fan on the power supply should spin merrily. A heavy air 
of dread and foreboding descended upon the room. Sure, Tricky Dickie 
recognized the Red Chinese, but why should we suffer such terrible retribution 
from the Taiwanese.  
 
Not to be conquered without a fight, I took my trusty screwdriver in hand and 
proceeded to crack the shell of the offending power supply. Lo and behold the 
quality control inspector must have drunk too much plum wine the night before 
he passed this particular power supply. A connector marked "AC IN" was not 
plugged into its designated receptacle. Thank the Lord that this power supply 
wasn't riveted together like so many of Big Blue's. Putting the energy box 
back together, we were greeted with a hearty hum from the fan and all voltages 
checked out within specs.
 
Next came the mounting of the disk drives. Suddenly the flip-top case did not 
seem like such a terrific convenience when we were told that the drives had to 
be mounted first in their bracket and then mounted in the case using screws 
that were inserted through the bottom. Not too bad the first time around, but 
I'm certain that this "hairy barbarian" will be directing mumbled expletives 
eastward when one of these drives needs to be removed. The flip-top sides also 
make it difficult to mount certain cards. Recently we received a hard card for 
another machine. The manufacturer recommends you put it into the first slot so 
that other slots are not wasted. The flip-top case makes it impossible to 
insert the hard card into the first slot.
 
At this point all the card were in place and the fateful moment of sending our 
monster to the roof to be invigorated with electricity arrived. The switch was 
manipulated and again nothing happened. After some worried scrambling we 
discovered that the switch settings on the motherboard were wrong. Our fault.  
After several "mea culpa's", the switch was thrown once again and we saw a 
wonderous sight. RAM memory was being checked, then a message appeared telling 
us that I/O boards were being tested and then, with bated breath, drives 
whirled into action and the monster opened its eyes and asked us for the date. 
Wild jubilation ensued.
 
After our emotions were contained, we proceeded to check out the various 
boards. The printer worked. That only left the serial card to check out. With 
mounting hope we ran PROCOMM and dialed up our local UNIX node, only to be 
greeted by smiley faces and male symbols. The IBM serial card obviously didn't 
like the shot-gun wedding we had arranged. First checking to make sure the 
card performed flawlessly in a true-blue PC, I called JDR for the first time 
thinking that someone in the universe had run into the same problem. JDR was 
courteous but hadn't a clue as to what the problem might be. They thought that 
everyone had their serial port on a multifunction card, but a wearer of hand-
me-downs can't keep up with the latest fashion trends. Anyhow I borrowed a 
multi-function card from a neighbor's PC and it worked flawlessly in the 
clone. Next I stared at the board and its schematic in the Technical Reference 
Manual. All I could find was a mysterious jumper on the board labeled J13. 
This jumper enables the Data Input Strobe (bar DISTR). For some reason only 
known in Tapei the Taiwanese motherboard didn't like this signal being 
asserted. I removed the shorting block from J13 and was relieved to see the 
familar login message. 
 
All in all, I guess our clone experience hasn't been that bad. The DOA power 
supply may have been a fluke and the hardware incompatibility with the IBM 
serial card was solved without too much trouble. I did, however, neglect to 
mention that in the midst of all this frenzied activity, there was a period of 
time in which the system just refused to response.  After some panic-stricken 
moments of flexing the motherboard and resetting cards, the system resurrected 
itself. I expect this kind of behavior from my wirewrap boards, it makes me 
nervous with a multi-layered printed circuit board. Please don't flame about 
getting what you pay for, etc. The risks seemed acceptable and we do have a 
functioning system. How long it will remain so is anybody's bet. Just thought 
those of you out there in netland might like to learn from our experience.
 
                                                    Larry Taylor (alt@duke) 
                                                    Duke Biophysipedic Lab