14 Jun
2009
14 Jun
'09
1:03 p.m.
The PACE was National Semiconductor's first single-chip 16-bit
microprocessor. Its architecture was similar to National's earlier
IMP-16 multichip processor, which in turn was inspired by the Data
General Nova architecture.
Godbout designed and advertised a PACE system, and built a prototype,
but never put it into production. PACE never caught on with hobbyists,
possibly because it was slower than the more common 8-bit processors.
PACE, and the later INS8900 NMOS version, did get used in some embedded
systems.
In May I wrote an assembler and simulator for the PACE, and typed in the
published source code for PACE FIG-Forth. I found and corrected a bug
in the U/ word.
In case anyone actually has a PACE system, I've put the sources online:
http://www.brouhaha.com/~eric/retrocomputing/national/pace/figforth/
I'm not quite ready to publish my assembler and simulator, though if
anyone feels an urgent need for such things, let me know.
Eric
14 Jun
14 Jun
2:49 p.m.
On 14 Jun 2009 at 13:03, Eric Smith wrote:
Godbout designed and advertised a PACE system, and
built a prototype,
but never put it into production. PACE never caught on with
hobbyists, possibly because it was slower than the more common 8-bit
processors. PACE, and the later INS8900 NMOS version, did get used in
some embedded systems.
At a Wescon (can't recall which one) in the 70's, NSC was giving away
samples of the PACE, along with documentation. Well, I can't resist
anything *free* at trade shows, even if it's a dead i386 die encased
in resin and promoted as a keychain fob, so I grabbed a sample
package.
I eventually wired up an S100 board for it, complete with 8-to-16-bit
bus conversion circuitry. If you thought that a PACE was slow in
native mode, try double-cycling the bus to support it! I also recall
that the PACE requiired a substantial number of support chips. At
some point, I scrapped the board as being unworkable.
For the life of me, I couldn't figure out what market NSC was
shooting for. A 16-bit micro that was glacially slow with a small
register file and a stack that needed interrupt servicing to make it
really usable. A Z80A or 8085 would run rings around it.
It seems to me that NSC has always had a problem with maintaining
focus on its own microprocessor products. The SC/MP, PACE, NSC800,
32xxx series of CPUs all seemed to be handled with a big marketiing
splash and then ignored. Some of these, such as the 32xxx MPUs were
technical tours-de-force (and very late). Something like corporate
attention-deficit disorder. I think National probably did best
second-sourcing other people's MPUs, letting the primary source
handle promotion and marketing.
--Chuck
17 Jun
17 Jun
5:08 p.m.
At a Wescon (can't recall which one) in the
70's, NSC was giving away
samples of the PACE, along with documentation. Well, I can't resist
anything *free* at trade shows, even if it's a dead i386 die encased
in resin and promoted as a keychain fob, so I grabbed a sample
package.
I'd **love** to have one of these, principally FREE :o)
I'll try to do that with encapsulated chips, with the bottom open :oD
5:21 p.m.
Chuck Guzis wrote:
At a Wescon (can't recall which one) in the
70's, NSC was giving away
samples of the PACE, along with documentation. Well, I can't resist
anything *free* at trade shows, even if it's a dead i386 die encased
in resin and promoted as a keychain fob, so I grabbed a sample package.
Alexandre Souza wrote:
I'd **love** to have one of these, principally
FREE :o)
One problem with taking advantage of a PACE chip (free or otherwise) is
that the I/O pads aren't compatible with TTL, or with much of anything
else either. The specs are very strange. National made some buffer and
transceiver chips designed to work with it, but of course they are even
more like unobtanium these days than the PACE itself. I suspect that
the easiest thing to interface to it nowdays would be CD4000B series
CMOS logic, which is extremely slow, but then so is the PACE.
The later INS8900, which was basically an NMOS version of the PACE, had
more conventional I/O pads so it was easier to interface.
Eric
7:12 p.m.
On 17 Jun 2009 at 17:21, Eric Smith wrote:
One problem with taking advantage of a PACE chip (free
or otherwise)
is that the I/O pads aren't compatible with TTL, or with much of
anything else either. The specs are very strange. National made some
buffer and transceiver chips designed to work with it, but of course
they are even more like unobtanium these days than the PACE itself. I
suspect that the easiest thing to interface to it nowdays would be
CD4000B series CMOS logic, which is extremely slow, but then so is the
PACE.
That's exactly what I did. The various PACE interface chips
(STE,MILE, etc) were expensive and not easy to findt, so I made do.
That was also my first exposure to 4000 series CMOS logic.
The whole thing pretty much filled up an S100 prototype card. Lotsa
wires.
--Chuck
9:04 p.m.
Alexandre Souza wrote:
No, I'm not talking about the PACE chip, but the keyfob with the 386 die
encased in plastic :o)
I never got to encase in resin a chip and make it crystal clear. I tried
many many times with no success :(
I'd **love** to have one of these,
principally FREE :o)
One problem with taking advantage of a PACE chip (free or
otherwise) is
that the I/O pads aren't compatible with TTL, or with much of anything
else either. The specs are very strange. National made some buffer and
transceiver chips designed to work with it, but of course they are even
more like unobtanium these days than the PACE itself. I suspect that
the easiest thing to interface to it nowdays would be CD4000B series
CMOS logic, which is extremely slow, but then so is the PACE. 
18 Jun
18 Jun
12:14 a.m.
On Jun 18, 2009, at 12:04 AM, Alexandre Souza wrote:
Did you have problems with air bubbles, or something else? The
way to solve air bubble problems is to place it under vacuum before
it cures.
-Dave
--
Dave McGuire
Port Charlotte, FL
One problem
with taking advantage of a PACE chip (free or
otherwise) is
that the I/O pads aren't compatible with TTL, or with much of
anything
else either. The specs are very strange. National made some
buffer and
transceiver chips designed to work with it, but of course they are
even
more like unobtanium these days than the PACE itself. I suspect that
the easiest thing to interface to it nowdays would be CD4000B series
CMOS logic, which is extremely slow, but then so is the PACE.
No, I'm not talking about the PACE chip, but the keyfob with the
386 die encased in plastic :o)
I never got to encase in resin a chip and make it crystal clear.
I tried many many times with no success :( 
3:57 a.m.
On Thu, 2009-06-18 at 01:04 -0300, Alexandre Souza wrote:
My fiancee has done that, with a couple of resistors, an SSM2044
analogue filter IC (dead) and the knackered ceramic resonator from the
memory board on my old PDP-11.
The answer seems to be to use a clear-coat lacquer with the same
refractive index as the resin to smooth over the bumpy surfaces. Pics
and further instructions from when she gets back from the crafts
store ;-)
Gordon
Alexandre
Souza wrote:
No, I'm not talking about the PACE chip, but the keyfob with the 386 die
encased in plastic :o)
I never got to encase in resin a chip and make it crystal clear. I tried
many many times with no success :( I'd **love** to have one of these,
principally FREE :o)
One problem with taking advantage of a PACE chip (free or
otherwise) is
that the I/O pads aren't compatible with TTL, or with much of anything
else either. The specs are very strange. National made some buffer and
transceiver chips designed to work with it, but of course they are even
more like unobtanium these days than the PACE itself. I suspect that
the easiest thing to interface to it nowdays would be CD4000B series
CMOS logic, which is extremely slow, but then so is the PACE. 
17 Jun
17 Jun
5:35 p.m.
Alexandre Souza wrote:
Hey for fun I was thinking of taking some transistors, octal coil holders
and plastic covers to make a 'solid state tube'. The heat sinks would
look like Seleium recitfiers fins... :)
Ben
At a Wescon
(can't recall which one) in the 70's, NSC was giving away
samples of the PACE, along with documentation. Well, I can't resist
anything *free* at trade shows, even if it's a dead i386 die encased
in resin and promoted as a keychain fob, so I grabbed a sample package.
I'd **love** to have one of these, principally FREE :o)
I'll try to do that with encapsulated chips, with the bottom open :oD
7:23 p.m.
On 17 Jun 2009 at 18:35, bfranchuk at jetnet.ab.ca wrote:
Hey for fun I was thinking of taking some transistors,
octal coil
holders and plastic covers to make a 'solid state tube'. The heat
sinks would look like Seleium recitfiers fins... :) Ben
I don't know if it was ever done for active circuits, but solid-state
"tubes" for the old 0Z4 cold-cathode gas recitifer found in auto
radios (with vibrator supply) were very popular as you didn't get any
RF "hash" in the sound. Same sort of fixture for other common
rectifier types, such as 5Y4.
I recall when the early silicon recifiers came in little ceramic
cartridges, held in fixtures that looked like fuse holders. IR used
to advertise kits to replace various selenium rectifiers. At least
silicon didn't stink when it went bad.
--Chuck
18 Jun
18 Jun
12:10 p.m.
> Hey for fun I was thinking of taking some transistors, octal coil holders
It's probably simpler if you use FETs.
and plastic
covers to make a 'solid state tube'. The heat sinks would
look like Seleium recitfiers fins... :)
I've seen that as replacement for rectifiers and RF Power tubes...
1:43 p.m.
Tony Duell wrote:
I seem to rememebr seeing references to little PCBs
you could fit over
the pis nf defective 01As (between the valve and the holder) that
contained an FET and a few other bits and let you keep 1920's radios in
operation.
Why bother? You can't find any AM music from 20's anyhow.
More and more I find out about old Radios the tubes are the parts that work
the best. It is the old CAPs that you have watch out for, and the mechanical
parts. I tossed two old radios since the dial cords broke as well as other
major problems.
-tony
Ben.
PS. I have a old short wave radio I plan to oil up today in my spare time today.
(No WD-40 here!)
2:02 p.m.
"bfranchuk at jetnet.ab.ca" wrote:
Why bother? You can't find any AM music from
20's anyhow.
More and more I find out about old Radios the tubes are the parts that work
the best. It is the old CAPs that you have watch out for, and the mechanical
parts. I tossed two old radios since the dial cords broke as well as other
major problems.
?? Aside from corrosion and broken dials, there's not much inhibiting repair
(of the electronics) of an old radio, dial cords are easy (and yes, I've done
some complex ones). Not to say that sometimes one just might as well toss it in
terms of effort vs. benefit.
2:02 p.m.
On 18 Jun 2009 at 14:43, bfranchuk at jetnet.ab.ca wrote:
Why bother? You can't find any AM music from
20's anyhow.
More and more I find out about old Radios the tubes are the parts that
work the best. It is the old CAPs that you have watch out for, and
the mechanical parts. I tossed two old radios since the dial cords
broke as well as other major problems.
Actually, it seems as if NEW caps are just as bad if not worse.
Chinese capacitor disease seems to be a pandemic.
--Chuck
3:31 p.m.
Chuck Guzis wrote:
Actually, it seems as if NEW caps are just as bad if not worse.
Chinese capacitor disease seems to be a pandemic.
One of the few remaining local parts retailers just switched resistor brands
after many years. The new ones are visibly crap
(but apparently they 'sound cleaner' to the audioph**les who are some
significant portion of his market).
Just one retailer, or is this is the next step towards death of 'conventional'
home-brew electronics and repairs? With the market for wire-lead components
having been so reduced by SMD, what market remains is filled by low-quality manufacturers.
3:46 p.m.
New subject: new crappy components
Brent Hilpert wrote:
Chuck Guzis wrote:
What are they ... I am building a amp and just using 1/2 watt carbon composition
for everything but power resistors.
Actually, it seems as if NEW caps are just as bad
if not worse.
Chinese capacitor disease seems to be a pandemic.
One of the few remaining local parts retailers just switched resistor brands
after many years. The new ones are visibly crap
(but apparently they 'sound cleaner' to the audioph**les who are some
significant portion of his market). Just one retailer, or is this is the next step towards
death of 'conventional'
home-brew electronics and repairs? With the market for wire-lead components
having been so reduced by SMD, what market remains is filled by low-quality
manufacturers.
I am looking at the new resistors ... all film resistors to blow at .1% greater
wattage than specified
@ 70C. I planned to use ordinary 5 watt power resistors, but they don't handle
momentary overload
at all nowdays, so I had buy 'audio' resistors at 10x the cost. 50 cents vs 5
dollars.
Ben.
4:11 p.m.
New subject: new crappy components
Gordon JC Pearce wrote:
On Thu, 2009-06-18 at 16:46 -0600, bfranchuk at
jetnet.ab.ca wrote:
> I am looking at the new resistors ... all film resistors to blow at .1% greater
> wattage than specified
> @ 70C.
[citation needed] --> ME
Fact #1 power rating you start de-rating @ 70C.
Fact #2 They have a very high thermal resistance. I can't find the data
but something like 150K/watt if I am guessing right.
Fact#1+Fact#2 == *poof*
Gordon
4:03 p.m.
New subject: new crappy components
"bfranchuk at jetnet.ab.ca" wrote:
Brent Hilpert wrote:
For 15 years he was selling a grey-body metal-film resistor which seemed of
reasonable or good quality, at least they were consistent and reliable. I think
they were Philips.
The new ones are a deep-blue-body (not flame-proofs) carbon-film type. Aside
Chuck Guzis wrote:
What are they ... I am building a amp and just using 1/2 watt carbon composition
for everything but power resistors. Actually, it seems as if NEW caps are just as bad
if not worse.
Chinese capacitor disease seems to be a pandemic.
One of the few remaining local parts retailers just switched resistor brands
after many years. The new ones are visibly crap
(but apparently they 'sound cleaner' to the audioph**les who are some
significant portion of his market). from the difficulty in distinguishing some color codes
against the deep-blue of
the body (leading to packaging errors, like 47K's in
the bag of 4.7K's), the
visible problem is that a fair portion of them out-of-the-bag have the final
protective coating formed into drips as they dried and too thin in spots (one
can see the resistive helix). They do not inspire confidence. I don't know the
actual manufacturer.
5983
days inactive
5987
days old
19 comments
8 participants
participants (8)
-
alexandre-listas＠e-secure.com.br -
ard＠p850ug1.demon.co.uk -
bfranchuk＠jetnet.ab.ca -
cclist＠sydex.com -
eric＠brouhaha.com -
gordonjcp＠gjcp.net -
hilpert＠cs.ubc.ca -
mcguire＠neurotica.com