Microcode's great advantage over hard-wiring is that, as Lampson
notes, "it's much easier to express something complicated in program
instructions, or micro-instructions, than directly in gates." Detect a bug
in a hard-wired machine and you have to rip out the errant circuits and
solder in new ones. With microcode you just rewrite the program. It is as
though in traversing the United States from Boston to San Francisco you
find that the paved highway through Cincinnati leads not to San Francisco but San Diego. Microcode would be the equivalent of getting on an
airplane and changing the flight path rather than ripping up and relocating the highway.
Translating hardware into software in quest of this simplicity is, as
processes go, fiendishly complex—the digital signals take a lot more
twists and turns in microcode than in a hard-wired machine; but to Taylor's group its mysteries were an open book. The BCC 500 was micro-
coded, as was the FLEX machine Kay designed for his Utah doctorate.
"We knew a lot about microcode," recalled the latter, who contributed
some conceptual ideas while observing CSLs actual design and construction of the machine from the sidelines. "One of the world's great micro-coders, Ed Fiala, was with us, so that was a big plus. One of the great
hardware designers, Thacker, was there, that was a big plus. The minus
was that there were only about ten people to do it all. And it had to be
done quickly."
Once they settled on the new course of action, a whole world of possibilities suddenly opened to them. They could not only clone the PDP-10;
they could
improve
on it. The first place to do so was the memory. PDP
machines used ferrite core memory, the standard in the industry since its
invention in 1951 by Jay Forrester at MIT. Core memories were made of
tiny rings (or "cores") woven into a mat of copper filaments that allowed
each ring to be magnetically polarized. The cores held, or "remembered," their charge until deliberately overridden, leaving a pattern to be
read back and interpreted as data bits in storage.
Core's great virtue was its reliability. But it had to be manufactured
by hand and suffered from the defects of great bulkiness and slow
speed. The 300-kilobyte core memory of Wes Clark's TX-2, for
instance, was a handmade block of about a cubic yard, or the size of a
large file cabinet, costing roughly half a million dollars. This volume of
memory was large for its time but trivial by todays standards, when a
desktop computer's semiconductor memory can hold about one hundred times as much information in the physical volume of a couple of
credit cards, at a cost of about forty dollars.
By 1971 core memories were about to be supplanted by a brand-new
technology based on silicon semiconductors. The previous October a
one-kilobit memory chip (that is, 1,024 bits of memory per chip) had
been introduced by Intel, a young engineering company co-founded
by Gordon Moore of "Moore's law" fame. Intel's 1103 chip was struggling for acceptance in the computer industry, largely because its peculiarities gave system designers migraines. The
1103 memories were "volatile," meaning that all the stored data were
wiped out whenever the chip's charge was lost. Because the 1103's charge
had a tendency to gradually leak away, the chip had to be recharged, or
refreshed, by zapping it with an electrical impulse several thousand times
a
second
to keep data from evaporating. (Cores, by contrast, were "non-volatile," meaning they held their stored data indefinitely, charged or
not.) The 1103 required users to supply it with all sorts of "weird voltages," as Lampson later put it, and looked like it might be prone to a host
of data errors arising from the density at which designers crammed it
with microscopic transistors.
Such flaws made the 1103 a spectacularly stubborn and perverse
contrivance. Its patriarch, Gordon Moore, termed it "the most difficult-
to-use semiconductor ever created by man." It was also hard to manufacture. Intel had so much difficulty turning out an economical volume of
working chips that it had to assign entire teams of engineers and technicians to the drudgery of picking good chips out from the river of useless
silicon coming off the fabrication line, a job so fervently detested it was
labeled "turd polishing."
To the CSL team, however, the 1103 s shortcomings were obstacles to
be overcome. "It seemed pretty clear to us that the memory should be
semiconductor, although we didn't really know whether those chips
worked—and it turned out later that they don't," Lampson recalled. "We
certainly would have preferred a more robust chip. But we were very
confident that by putting in error correction we could make up a very satisfactory system that
would
work." That was an understatement. As CSL
well knew, if they could only overcome the 1103's manifold obstinacies,
their machine would boast the speediest and most reliable memory on
Earth. CSL's resolution of the PDP-Sigma imbroglio failed to quell entirely
SDS's discontent with the outcome. From El Segundo was heard continuous carping that PARC had caused the division insupportable
embarrassment by spurning its top-of-the-line product. This provoked
Pake into an outburst that settled the matter once and for all. In a blistering memo he reminded headquarters that his best new engineers
had voluntarily agreed to suspend bona fide research projects for the
year or more it would take them to satisfy SDS's querulous concerns.
"It is unthinkable to me that Xerox sets me the task of hiring creative,
imaginative, top-rank researchers and then expects me to insist that
they handcuff themselves with
inappropriate
equipment," he wrote.
"I
will do my best to provide them with the kind of first-rate technical
support it is reasonable
to expect
in
Xerox
research laboratories.
If
that
is the wrong way to build a
first-rate
corporate research center for
Xerox,
then
I
am the wrong man for the job."
Meanwhile, the
CSL
rank and files lingering resentment at their
Southern California in-laws
was
manifested when the time came to
give their clone
a
name. One afternoon a group of engineers gathered
at someone's house to mark an intermediate milestone on the project.
"A
moderate amount of beer had been consumed and
we
were trying
to figure out what
to
name this thing and there was considerable hilarity,"
Thacker recalled of the day they came up with the formal moniker of
"Multiple Access
Xerox
Computer."
It
sounded conventional, but everyone on the scene got the joke.
In
honor of the man who had sold his lousy
computer company to
Xerox,
the first major project undertaken by
PARC
would be known for all time by the acronym
"MAXC." No
member of the
lab ever forgot to remind outsiders, "The
'C'
is silent."
MAXC's
christening was
more
than an opportunity to tweak
Max
Palevsky.
*
It
crystallized their
awareness
that what was coming together
in the Computer Science
Lab
was no longer a
DEC
machine, but their
own.
"We
did everything, from
soup to
nuts,"
McCreight
said. Liberated
from slavish adherence to the
PDP-10 design,
they were able to get
dozens of functions running faster or cheaper.
McCreight
performed one
such feat
with his disk controller. Conventional
disk controllers were generally equipped
with
their
own
separate processing units, like Stegosauruses with their
second
brains, which added significantly to their cost
and complexity.
Poring
over the
system
schematics in his office one day,
he was struck
by the realization
that
there
would occur certain periods
when, having
executed one
instruction and
not
yet received
the
next,
MAXC's central processor would be
idle
but
available, like a
car
left running unattended in
the
driveway.
*
The point was probably lost on the target. As Pake said later, "I doubt that Max
Palevsky ever cared, or even knew about it."
"I learned enough about the processor to realize I could use some of
those spare cycles," he recounted. "In effect I could kidnap the processor
to do some arithmetic for the disk controller. I wouldn't have to put so
many gates into the disk controller"—saving another few thousand dollars in hardware—"if I could periodically borrow the processor to compute some of the things I needed to compute." McCreight's realization
was their first embrace of the concept of "multitasking"—giving the
processor numerous jobs to juggle at once. Implemented on this modest
scale in MAXC, it was destined to pay enormous dividends later.
Meanwhile, they continued to inject refinements into the PDP-10
design. If along the way they discovered some flaw, an inelegance or
vulgarity committed by the original designers, they had no compunction about fixing it. At least once this resulted in making MAXC
too
good. This was the episode of Fiala's floating-point bug.