That's it -- I'm building my own PC!.
J. Padron
I'm tired of dealing with Gateway2000, Micron, Dell, even IBM.
I'm tired of spending big bucks buying technology that will be obsolete in
just a few month and then discarding my PC instead of doing the upgrades
myself.
MY QUESTION TO THIS NEWSGROUP IS: is it worth the time and effort to learn
how to assemble one's own PC, buying one's own [good] parts, and doing all
the maintenance and upgrades?. Or it it better to buy a basic PC and
upgrade it myself?.
If so, what book do you recommend to teach myself the basics on how to do
that. I'm not totally clueless -- I'm a software developer and the I.S.
director for a company. However, my knowledge of PC hardware is
limited. (I like the idea of learning and doing this myself assuming it
makes financial sense).
I appreciate your input,
--
Jorge Padron
pad...@gate.net
The Beacon Council
--
bi...@mcs.com
In article <Pine.A32.3.91.950302...@hopi.gate.net>,
<pad...@gate.net> writes:
> Path:
ddsw1!godot.cc.duq.edu!newsfeed.pitt.edu!uunet!cs.utexas.edu!news.sprintlink.ne
t!tequesta.gate.net!hopi.gate.net!padron
> From: "J. Padron" <pad...@gate.net>
> Newsgroups: alt.comp.hardware.homebuilt
> Subject: That's it -- I'm building my own PC!.
> Date: Thu, 2 Mar 1995 15:48:05 -0500
> Lines: 25
> Message-ID: <Pine.A32.3.91.950302...@hopi.gate.net>
> NNTP-Posting-Host: hopi.gate.net
> Mime-Version: 1.0
> Content-Type: TEXT/PLAIN; charset=US-ASCII
individual parts using a great deal of research. Then after
assembly of of parts a burn out period follows.
Dan Finton
writes:
>
>I'm tired of dealing with Gateway2000, Micron, Dell, even IBM.
>how to assemble one's own PC, buying one's own [good] parts, and doing all
>the maintenance and upgrades?. Or it it better to buy a basic PC and
>upgrade it myself?.
sense. You can buy a complete, major brand 4/66 for around 1500 to 1700.
By the time you spend $300 for a motherboard & cpu, another 3 on ram, and 3
to 4 on a hard drive, you're at $1000, and still need a case, monitor,
video card, keyboard, mouse, operating system, etc. If you have an older
machine to start with, like an old 286 or 386, you can save some over the
cost of a new machine. How much depends on the configuration of the old
machine. If you're building a box to handle single line remote access, your
old 40 meg hd and less than hitec monitor may serve just fine.
For Photoshop work, this might not serve.
On the plus side though, I think its fun. It impresses the hell out of
parents and elderly relatives, and you run into more hardware conflicts
than the average 4 year CompSci grad. You also get to pick exactly which
features to beef up, and which to slim down, you can play with 3 flavors of
drive controllers if you like, and generally build your dream box. If you
( like me ) simply have to take home every old, dead or thrown out piece of
hardware you find, you always have a cheap base to start your next project.
Walter Wallenborn
: > I'm tired of dealing with Gateway2000, Micron, Dell, even IBM.
: > I'm tired of spending big bucks buying technology that will be obsolete in
: > just a few month and then discarding my PC instead of doing the upgrades
: > myself.
: >
: > MY QUESTION TO THIS NEWSGROUP IS: is it worth the time and effort to learn
: > how to assemble one's own PC, buying one's own [good] parts, and doing all
: > the maintenance and upgrades?. Or it it better to buy a basic PC and
: > upgrade it myself?.
: >
: > If so, what book do you recommend to teach myself the basics on how to do
: > that. I'm not totally clueless -- I'm a software developer and the I.S.
: > director for a company. However, my knowledge of PC hardware is
: > limited. (I like the idea of learning and doing this myself assuming it
: > makes financial sense).
: >
: >
: > I appreciate your input,
: >
: > --
: > Jorge Padron
: > pad...@gate.net
: > The Beacon Council
: > --
: >
: I do not try to invent the wheel. However, I do carefully select
: individual parts using a great deal of research. Then after
: assembly of of parts a burn out period follows.
:
The bottom line is that if you want the best price there is no way you are
going to build it from scratch. Try going to your local auto supply shop and
buy the bits to build a car! :))
I always shop around for the cheapest package deal and them see what they
want for it with the bells and wistles I need to make it useful. The real
uses for this news group are for the people who had something dropped in
their lap that won't quite fit directly and they want to make the interface
or they have something external that they are just using a PC as a very
specialised keyboard to get data or control onto their SPECIAL stuff.
cya
Walter
C. Deforrest Smith
For the nth time, this is NOT the appropriate place to discuss plugging
together PC clones! Take it to one of the many comp.sys.ibm.pc.hardware.*
groups..
Sheesh.
cds
Mike Inman
> MY QUESTION TO THIS NEWSGROUP IS: is it worth the time and effort to learn
> how to assemble one's own PC, buying one's own [good] parts, and doing all
> the maintenance and upgrades?. Or it it better to buy a basic PC and
> upgrade it myself?.
If you are starting from scratch, it is far more economical to buy
a complete system than the component parts. If, however, you happen
to have a monitor, keyboard, mouse, case, and/or other assorted parts
that you can use in the new system, then it can pay to assemble your
system from bits & pieces.
I did just that (assembled my system from parts), and it still cost
me about $1000, even with a "free" monitor, video card, floppy drive,
keyboard and mouse. But then again, I did purchase a 486DX2-66,
4 Meg of RAM, a 420MB Maxtor drive, and a Full Tower case with a
killer 300W power supply (which carried me through a 2 second power
glitch, sans UPS, thank you very much.)
I like having the Full Tower case (on the floor) with the room and
the power supply to throw in whatever may strike my fancy in the
future. You really can't get that from an "economical" package deal.
There really aren't any secrets to assembling your own system, you
just need all the parts. I purchased my Motherboard, RAM, Hard Drive
and controller card all from a single source. I didn't have any
problems, but if I did, I wanted to be able to call one place for
help instead of getting the run-around. Also, the place I purchased
from was kind enough to set all the appropriate jumpers on the
motherboard. I still read the manuals to make sure for each of them
before putting the power to it, but it was nice to know that both
my and the vendor's interpretation of the typically cryptic manual
agreed.
Caveats:
Check what kind of SIMMs your motherboard takes (30 or 72 pin).
If you have some 30 pin SIMMs laying around, you want to be able
to use them. If you have to buy new RAM, you want to buy the
72 pin variety. Chances are that you will replace your motherboard
sooner or later, and you'll want to carry over any RAM that you buy
to the new motherboard, which will most likely carry only 72 pin
SIMM sockets in the future.
PCI is the bus of the future, but most affordable motherboards today
are VLB. Try not to sink too much into VLB cards that could just as
well be replaced with ISA type cards. ISA cards will be usable for
many motherboards to come.
Attach your motherboard securely to the case. There should be at
at least three nylon stand-offs, and a metal screw down or two
(typically near the expansion card slots). I had a difficult ten
minutes or so while I tried to put it in backwards, but after that
things snapped into place quite logically.
Motherboard power: Black goes beside Black. My system had absolutely
no usable doccumentation for the motherboard power connectors, and
the connectors did not seem to be "idiot proof." Both connectors
contain a couple of black ground wires. These black wires all go
side by side when plugged into the motherboard (on every motherboard
I have ever seen, which is maybe twenty or so different types.)
Still, read your manuals carefully, especially the part about
"MOTHERBOARD MAY BE PERMENANTLY DAMAGED BY IMPROPER CONNECTION OF
POWER CABLES."
Don't build your own system if you can't handle some basic
documentation reading. As I expand my system, the most tedious
part is learning what all the jumpers and driver switch settings
do, and how this affects system performance. Especially in the
software, different switch settings usually entail trade-offs instead
of outright performace increases. If you can't read and understand
the documentation, you'd be much better off going to a computer
super-store and buying a ready-configured system.
To conclude this impromptu article, if you've got some spare parts
lying around, and the savvy to find and read and understand the
documentation, you can economically buy the parts to build a custom
PC that incorporates the features that you value.
Mike Inman
NI...@CRIS.COM
p.s. if this is not what alt.comp.hardware.homebuilt is about, would
somebody please tell me where this type of discussion should take
place?
John Walton
somehow it seems to jinx it.
Juha Kuusama
>From a strict dollars and time perspective, building your own doesn't make
>sense.
A local PC dealer let me choose the features (like type of processor,
brand of video card, interfaces etc.) that made my still somewhat new
home PC. I forgot to tick the box "assembled and tested" and saved a
whopping $40! I spent one afternoon and better part of the night, though...
I quess I could do a second or third PC in an hour or so. The assembly
mechanic clearly makes something like $30 /hour.
--
Juha Kuusama
Mike Inman
You might be more sucessful in squelching the noise if there were
a comp.sys.ibm.pc.homebuilt.systems for the "noise" to move to.
As it is, most of the comp.sys.ibm.pc.hardware. groups are filled
with messages regarding upgrades of a specific component, and
c.s.i.p.h.system is primarialy concerned with Dell, Gateway, Acer,
ad nauseum.
Also, being in alt. a.c.h.h is going to be subject to extra noise
just by being near the top of the list of thousands of newsgroups.
There will be lots of "newbies" over the next couple of years who
will invariably come here before the comp.... groups. Perhaps your
best solution would be a new alt.comp.hardware.homebuilt.pc.systems,
there seems to be enough "noise" to justify it.
I'm not a enough of a Usenetite to know how to make these things
happen, and I'm not interested in investing any signifigant time
in it, but perhaps some of the "old time" a.c.h.h'ers could get
it done.
Mike Inman
Mads Haahr
] csm...@callamer.com (C. Deforrest Smith) wrote:
] >
] > For the nth time, this is NOT the appropriate place to discuss plugging
] > together PC clones! Take it to one of the many comp.sys.ibm.pc.hardware.*
] > groups..
]
] You might be more sucessful in squelching the noise if there were
] a comp.sys.ibm.pc.homebuilt.systems for the "noise" to move to.
Maybe. In any event, it seems to have become bad enough to justify
actually doing something. Obviously, lots of net.newbies believe that
homebuilt is the same as home-assembled. Let's face it, there's
nothing we can do about that, like we couldn't keep people from using
the word hacker to mean criminal.
To keep flaming newbies is not a solution that's going to work in the
long run. Most of them aren't even impolite but simply mistake the
group's subject because it isn't written in neon. Also, people whose
PCs are bellowing forth smoke are likely to be desparate enough not to
test the flavor of a newsgroup like this before posting.
I see two things we can do. One is rename the group to something that
suggests "homebuilt" on the circuit-level, like a.c.h.h.from-chips or
a.c.h.h.hardcore.
Another is to piece together a set of guidelines, maybe a FAQ, and
autopost it frequently enough. And hope that the newbies will read
it, of course; but if it says "How To Fix Your PC [tm]" in neon on the
cover, they actually might.
Thoughts?
--
Mads Haahr [<ha...@cs.ubc.ca>,<ma...@diku.dk>] | Humour
Department of Computer Science | may be
University of [British Columbia,Copenhagen] | unmarked
PE...@news.delphi.com
>er...@wolf359.exile.org (Eric Edwards) writes:
>>In article <D59sI...@indirect.com>, Stan Eker writes:
>>> Getting `homebuilt' completely out of the name would help, although anyone
>>> hurting for aid will *still* continue to trash any remotely related group.
>>I think the problem word is not "homebuilt" but "comp".
>>Perhaps alt.electronics.digital.homebuilt
>>I don't think real hardware hackers will have any trouble finding the
>>group.
> alt.electronics.digital.homebuilt is so...uninspiring. Howabout
>we cordon ourselves from the lost souls of PC hardware angst without
>alienating ourselves from the flash and dazzle of "computer". Howabout we
>reword out .comp and replaceit with .sys, to buffer ourselves from
>PC-platform specific questions. The off topic posts might start to
>include some oddball code jockeys with interesting OS, a definite
>on-topic subthread.
> So, here are my suggestions for The Big Change...
> alt.sys.homebuilt (or) alt.sys.homegrown [naaa...we'd get too
>many gardening questions] (or) alt.sys.unique [a danger from all
>of the restles PDP-11 and Osbourne users out there...] (or)
>alt.sys.homebuilt.technology [a danger from all the yahoos who have made
>a better mousetrap...but at least their off-topic rants would be interesting]
>>----
>>Eric Edwards
>
How about nonstandard some where? I mean most people are trying to build
a standard system, etc... and I think nonstandard might scare them off
and also attract people with idea's for nonstandard things.
-robert
Tony Duell
sou...@enterprise.america.com "Matthew Gabriel" writes:
> alt.electronics.digital.homebuilt is so...uninspiring. Howabout
Quite apart from the fact that computing does not _have_ to be digital.
Although analogue elctronics is not really my subject, I'd rather read posts
about analogue computers (lots of op-amps :-)) than 'fix_my_pc' posts.
> So, here are my suggestions for The Big Change...
>
> alt.sys.homebuilt (or) alt.sys.homegrown [naaa...we'd get too
most pc-goons would thing their home-assembled PC was a homebuilt system -
after all, they thing it's homebuilt, and it's a system...
> many gardening questions] (or) alt.sys.unique [a danger from all
> of the restles PDP-11 and Osbourne users out there...] (or)
Hey... I'm a PDP11 owner (amongst other things...), but I'm not about to discuss
it here. It's hardly a unique system. Anyway, the owners of such systems, are
by necessity, serious hardware hackers, and would quite happily chat about
digital electronics all day. I'd rather have that sort of noise than PC
assembling.
> alt.sys.homebuilt.technology [a danger from all the yahoos who have made
> a better mousetrap...but at least their off-topic rants would be interesting]
Indeed...
-tony
Eric Edwards
> How about nonstandard some where? I mean most people are trying to build
> a standard system, etc... and I think nonstandard might scare them off
> and also attract people with idea's for nonstandard things.
Hmmm. alt.sys.incompatible?
alt.sys.incompatible.design?
the same but whith non-standard.
I dunno. For every name under sys or comp, there are people who will
misinterpret the meaning.
non-standard and incompatible each would attract Macintosh and Amiga
people.
If alt.design.digital.homebuilt is uninspiring, how about
alt.homebrew.digital? Create alt.homebrew.analog for symetry and to
further confuse the pc people.
----
Eric Edwards: Bang= cello.qnet.com!wolf359!eric Domain= er...@exile.org
Remember the home hobbyist computer: Born 1975, died April 29, 1994
O. Hotz de Baar
want. Lets face it very few people have enough facilities to
'homebuild' computers. Even in the days of RadioShack, they just
assembled a kit. The only difference is that then you needed a
soldering iron and now you need a screwdriver. Most 'homebuilt' is
confined to periferal devices, and most of these have already been
designed. I am not too sure that designing (c/w OS) a Pentium (or
whatever) mainboard running @ 100MHz is really the job for 'the home'.
An RS232 keyboard encoder for long distance keyboards might be (Oh any
ideas on this by the way?).
You can however design without any hardware at all (shrieks of rage).
Oz
--
------------------------------
Oz@upthorpe via Demon Internet
------------------------------
O. Hotz de Baar
As a newbie myself (despite many many years on CServe) I have to say
that it is often quite hard to determine what is regarded as 'OK' on IN
newsgroups. It would *definitely* be helpful if a useful pointer to the
ethos of a newsgroup winged it's way down to me as soon as I downloaded
my first block of Subjects. It would (I guess) need someone to post the
message weekly so that newcomers would see it quite quickly, and it
would have to be short to avoid unnecessary downloading time for other
users. An obvious name like "FAQ: new members read me" would be good
enough, and it need only be a short message like
"Before posting to this group please download and read file XXX in
<address>". Given the high number of newbies it might be worthwhile
(perhaps) to spell out how to download in a little more detail (but not
much). If this was the case then group members *would* be entitled to
get *seriously* annoyed with users who infringe the 'rules'. As it is
it's often very difficult to find out what the rules actually are!
As it is it's those poor souls who DO innocently post something that
gets the ire of one of the long term members that gives an idea of what
the rules actually are, and are thus beneficial to all. Sad world,
isn't it?
Oz
--
-------------------------------
Oz@upthorpe via Demon Internet
-------------------------------
Jeff Hunsinger
> It seems to me that hardwaredesign is closer to what 'older' members
> want.
I'm 24 and I know plenty other people active in this group are probably even younger.
> Lets face it very few people have enough facilities to
> 'homebuild' computers.
All it takes is determination. I've built them using wireless breadboards, wirewrap,
and point to point soldering (yuk).
> Even in the days of RadioShack, they just assembled a kit. The only difference is
> that then you needed a soldering iron and now you need a screwdriver.
Radio Shack is still around, in case you haven't noticed. Although they did
offer kits at certain stores for a very limited time, their main line of computers
have always been prebuilt items. Heathkit used to sell kits, but they were often
just screwdriver jobs. I'm not sure who you're referring to as "they". Many
people have built their own machines from their own designs. People just don't
usually depend on that home built machine for all their computer work anymore.
Even with kit builders, you'll notice a large difference in understanding
between the soldering and screwdriver crowds.
> Most 'homebuilt' is confined to periferal devices, and most of these have already
> been designed.
That's sort of like saying everything that's patentable has already been discovered.
> I am not too sure that designing (c/w OS) a Pentium (or whatever) mainboard
> running @ 100MHz is really the job for 'the home'.
Depends on how determined you are. I'd rather use something other than Intel anyway.
> An RS232 keyboard encoder for long distance keyboards might be (Oh any
> ideas on this by the way?).
Sure. Use a microcontroller at both ends. Decode the keyboard at one end, send
the keycode through the built in UART to the the uC at the other end which then
re-encodes the signal to the computer. Easier yet is to just run a long cable.
I used to have a remote keyboard and monitor on my kitchen counter which plugged
into a jack on the wall. I used excess phone wires to run the signals to a
bedroom where the PC was located. It worked fine for me. I was even able to run
CGA monochrome over twisted pair. I was pretty shocked by this.
> You can however design without any hardware at all (shrieks of rage).
But that obviously wouldn't be a topic for alt.comp._hardware_.homebuilt, now,
would it?
I agree with an earlier response that the name should have really been
alt.comp.homebrew
Jeff
David Brooks
>It seems to me that hardwaredesign is closer to what 'older' members
>want. Lets face it very few people have enough facilities to
>'homebuild' computers. Even in the days of RadioShack, they just
[snip]
>Oz
>--
>------------------------------
>Oz@upthorpe via Demon Internet
>------------------------------
I'm not too sure about homebrewers no longer having the resources to brew
up our own CPU architectures, etc.
Many years ago now (late 1970's) I designed & built my own CPU "from the
ground up", using TTL/MSI parts. It ran to about 120 chips, plus the
memory array.
These days, I work professionally with FPGA's mostly. I have more than
once been asked to bring my old design up to date, by re-targetting it
into a FPGA. This would technically be quite feasible - at present all I
lack is the time!!!
FPGA's could be the homebrewer's instant answer (given access to low-cost
software - now *there's* a task for the Free Software Foundation!!)
Such designs might end up on this group, or on ...fpgas, but I would
suggest that complete CPU designs, etc., would go here, while FPGA tools
& tips would go in ...fpgas.
--
David R. Brooks <da...@perth.DIALix.oz.au>
Tel/fax. +61 9 434 4280
sokos mark
O. Hotz de Baar <O...@upthorpe.demon.co.uk> wrote:
>As a newbie myself (despite many many years on CServe) I have to say
>that it is often quite hard to determine what is regarded as 'OK' on IN
>newsgroups. It would *definitely* be helpful if a useful pointer to the
>ethos of a newsgroup winged it's way down to me as soon as I downloaded
>my first block of Subjects. It would (I guess) need someone to post the
>message weekly so that newcomers would see it quite quickly, and it
>would have to be short to avoid unnecessary downloading time for other
>users. An obvious name like "FAQ: new members read me" would be good
>enough, and it need only be a short message like
Well, we are supposed to have a FAQ, but effort on it seems sporadic
at best. If I we don't get regular postings soon, I am going to start
another one. Weekly is a bit often for FAQ's, unless they are under
construction (which ours is/was), but every 2 weeks or monthly is
common, depending on the newsgroup. All newbies should take a look
at news.answers for the newbie intro to usenet type stuff (and our
FAQ should go there, when completed).
>As it is it's those poor souls who DO innocently post something that
>gets the ire of one of the long term members that gives an idea of what
>the rules actually are, and are thus beneficial to all. Sad world,
>isn't it?
It's been a while since I've read them, but the last time I looked,
the newbie files basically said 1) get the FAQ for the group and
2) read for a while before posting. Anyone who does that to this
(or almost any other usenet group) should not have any difficulty.
The only things that tend to get the regular readers here really
ticked off are the "fix my pc" posts, where the reader obviously
did not do 1 & 2 above, nor did they bother to read the IBM PC
hardware questions pointer, which is frequently posted on news.answers
(and does not mention this group, I am happy to report).
*****************************************************************
* Mark Sokos (mso...@umbc.edu) The "M" in A&M Software *
* Control Systems Engineer, Computer Programmer, & *
* Perpetual Student *
*****************************************************************
Tony Duell
je...@oakhill-csic.sps.mot.com "Jeff Hunsinger" writes:
> In article 776039...@upthorpe.demon.co.uk, O...@upthorpe.demon.co.uk ("O.
> Hotz de Baar") writes:
>
> > Lets face it very few people have enough facilities to
> > 'homebuild' computers.
>
> All it takes is determination. I've built them using wireless breadboards,
> wirewrap,
> and point to point soldering (yuk).
Ditto, along with verowire (a self-fluxing wiring system that's quick for
prototypes, reliable, but not that robust for field use).
You don't need that many facilities. Look at the end of Hayes&Horowitz
(Lab manual for the Art of Electronics) to see a 68008-based computer that
_anyone_ could build and understand at home. No, I'm not suggesting that's
the ultimate design, just that homebuilding is still very possible.
>
> have always been prebuilt items. Heathkit used to sell kits, but they were
> often just screwdriver jobs. I'm not sure who you're referring to as "they".
The later heathkit computers were, I beleive screwdriver jobs only. Earlier
ones (H8, H11, etc) had some soldering in them, although the CPU board often
came pre-built.
> Even with kit builders, you'll notice a large difference in understanding
> between the soldering and screwdriver crowds.
Agreed. The 'soldering' crowd often go on to try altering things, and make
modifications, something that's impossible to do with just a 'screwdriver'
THere's nothing wrong in owning a ready-made computer. There's nothing
wrong in discussing it on usenet. But there are other groups to do so
in. This group is for homebuilt hardware. I'm sure the PC crowd wouldn't
like it if we took over their groups and started discussing new CPU
architectures (that don't run Windows!) just because they could be classed
as 'personal computers' (i.e. they are intended to be owned and used by 1
person).
> Jeff
-tony
Tony Duell
sou...@enterprise.america.com "Matthew Gabriel" writes:
> Errr...I am relatively new to .homebuilt(3 mos.), but its what I want.
> I
> want to lean how computers go together, how to design your own from
> scratch. I do not have the facilities to manufacture my own hardware yet,
> but please note the "yet". I want to learn the nitty gritty, how certain
You don't need that much. I do 90% of my debugging with only a cheap logic
probe and an old VOM. That's it. It's nice to have the 'scope and logic
analyser, but I could manage without them for most of my projects.
> memory controllers work, how ANY memory controllers work, what kinds of
I would suggest that you start by using static RAM only, and leave
dynmaic RAM with its wierd timing requiremnts and refresh cycles until
you've got a simple machine working. Similarly forget paging and memory
management for the moment. You can learn about these topics later on.
> microprocessors are suitable for certain jobs, etc.
To start with, I'd pick an 8 bit chip (either a microprocessor or a
microcontroller). Choose one that's got a pd monitor program available
for it, and build a single-board machine usign that. It may be slower
than a 1980's home ocmputer, but it will compute, and you did build it.
You'll never forget the day when your first machine runs its first program
correctly.
You can then go on to build bigger and more complex designs. The idea the
radio constructors used to have of starting with a crystal set and
movng on through the 1 valve regenerative design, via multi-valve TRF's
to superhets and beyond applies to computers as well. Your first machine
will not be a world-beater in terms of facilities or speed, but it is
_yours_.
> Why not? Steve Wozniak designed the first practical home computer
> in his garage.
I'm not convinced the Apple ][ was a good design, but I'll leave it at that
Certainly a machine like that could still be built at home.
> Besides, we will eventually run into the same old dogma wars.
> Most posts have been about PC hardware. What about the Mac? Or the Amiga?
> Or the Acorn? Or Sparc clones? You begin to see how unwieldy such a forum
> will become.
Agreed. I happen to like certain commercial computers (PDP11's, PDP8's, etc),
but I discuss them in the appropriate groups.
> SoupIsGood Food
-tony
Tony Duell
> Tony Duell wrote:
> >Deadbugging (Sticking the chips onto a copper-clad board as a ground plane,
> >and
> >designing the interconnections as transmission lines) will take you to at
> >least
> >100Mhz...
>
> Seems you have a different use of the word "deadbugging" than I do.
I've heard the term used for sticking a chip (pins up) to a normal PCB and
then using thin wire to link it into the circuit (an easy way to kludge in
a chip at the last minute). I've never heard it applied to _removing_
a chip. The term was extended to include _designing_ that way.
> Regarding transmission lines: funny how high speed digital is all
> of a sudden becoming an analog field, aint it? I have had more analog
> problems with digital boards than I have digital problems.
Vonada's law (I forget which one) :
Digital circuits are built from Analogue parts.
I think M.V.Wilkes (Designer of EDSAC) had some similar comment.
> Heh heh... I've gotten "technical specs" for a modern PC motherboard.
> English was obviously NOT the first language of the person who translated
> the material. A lot of it made no sense even when I knew exactly what
> they were trying to say.
And let me guess : The technical content was minimal. I've seen dimensions
and operating temp ranges count as 'technical specs' and those were the
only specs given.
Real technical info (schematics, source code, timing diagrams) are the same
in any language (or at least easy to translate).
> Exactly. I have to say I have no real purpose in building my homebrew
> machine (which is probably why it is taking me so long to do it). It
> is purely for my own enjoyment and education.
I can relate to that. I think the best way to learn is by doing.
> A piece of hardware works when you plug it in at its intended location
> and use it for real for the first time. I've had things work pefectly
> on my desk, and then go psycho when I actually went to use them.
Fortunately, the intended location for a lot of my hardware is in my workshop.
But, I guess you agree that a simulator is not _proof_ of a design.
> * Mark Sokos (mso...@umbc.edu) The "M" in A&M Software *
-tony
O. Hotz de Baar
I enjoyed reading the returns, and I DO actually agree with the spirit
of it all completely. The real question of what you want your forum to
be breaks, I think, into three parts:
1) A support group for old/worthy computers like PDP8, CBM4032 et al.
2) A support group for those wishing to design/build computers from the
chip level to final machine.
3) A support group for designing/building add-ons to existing designs
to perform novel, expensive or interesting facilities.
Perhaps I could take them one by one as to my view of the relevent
points attached to each.
1) With the greatest of respect PDP11 etc users are the guru types for
the computer industry. The machines are NOT home built, although I
would be astonished if the majority of the periferals were (and
probably by now much of the mainboard). I would enjoy reading their
threads, but would have nothing to offer. These guys are the equivalent
of those who lovingly refurbish ancient cars/steamengines. Great
respect, but I don't want to actually have to look after one myself
(well it would be fun I suppose .....). Oh and by the way, has anyone
out there got a repair schedule for a CBM4032 (1979), my old one zapped
out a couple of years ago and it would be REALLY nice to get it going
again.
2) Umm .. err .. Well my son would say something like 'Get real, man'.
The man hours and cost of designing, testing and building a modern (say
1990!) computer complete with periferals would take any individual 10
years+ even with help from the group. By then the parts would be
obsolete and hard to get. The cost would be astronomic, the performance
poor .... Nobody (OK only a very wealthy body) designs and build a car
starting with stock steel sheet & bar, and a few $10,000 of machine
tools. Yup, we could all do a Model T type of design, but that's
already been done. I do NOT include microcontrollers/PLA's etc here.
3) A support group for those meddling in computer hardware. Well here
the opportunities are huge. This is the area for all those really
useful things that either cost a bundle or simply do not exist. We are
talking design and hardware here. You CAN design and build a
microcontroller based subsystem (or PLA) quite easily. What costs and
is 'difficult' is the PCB and development system. I am talking amateur
here, those people who do it at lowest cost in hardware terms for the
fun/thrill of a neat cheap design. I do not particularly like
re-inventing the wheel (how to draw a circle: build computer, OS, HL
language, plotter :draw circle!), here is a group that has done the
bits. I want to design a PIC (chip) based system so I ask the group.
Someone offers software for an emulator, another has a programming
subsytem design, another has the basic SS board design so now I'm
flying. I develop on the emulator and modify the board design to suit
what I want to do and I am in the position to knock out the job in a
few weeks. Then off to the next little project. Maybe this time I'll go
for a 68.... Doubtless there would be discussion about the relative
merits. Here I include build your own process controllers, sound cards,
laser light show, fuel/ignition processor, sattelite subsystem .....
I have to say that there is more to it than that. How do you produce
your boards. For small runs (like one) a useful tip might be that you
can get away (for single sided) by plotting the trace directly onto the
copper using an acetate pen, then a simple etch and you are done.
Except for the holes that is. Lots of useful tips could flow down the
wires to a REAL hardware construction forum.
Or perhaps I am talking total gibberish?
--
--------------------------------
Oz@upthorpe via Demon Internet
--------------------------------
Tony Duell
O...@upthorpe.demon.co.uk "O. Hotz de Baar" writes:
> be breaks, I think, into three parts:
>
> 1) A support group for old/worthy computers like PDP8, CBM4032 et al.
No. There are plenty of other groups for that, for example alt.sys.pdp8,
vmsnet.pdp-11, comp.os.cpm, alt.sys.perq, alt.sys.pdp10, etc. Old machines
happen to be one of my interests, but not one that should IMHO be discussed
here.
> 2) A support group for those wishing to design/build computers from the
> chip level to final machine.
> 3) A support group for designing/building add-ons to existing designs
> to perform novel, expensive or interesting facilities.
I think the group could cover both of those subjects. I've never seen a
flame-war over home built add-ons, and I hope I never do. Not everyone wants to
build the machine from scratch.
There's a lot in common between building a CPU from gates (I'll include
TTL, and FPGAs etc here) and building (say) an interface between a PC and a
paper tape reader. There's a lot less in common between either of those and
assembling a PC from boards
>
> Perhaps I could take them one by one as to my view of the relevent
> points attached to each.
>
> 1) With the greatest of respect PDP11 etc users are the guru types for
> the computer industry. The machines are NOT home built, although I
> would be astonished if the majority of the periferals were (and
> probably by now much of the mainboard). I would enjoy reading their
I'm not sure I follow that. The PDP11 (and like machines) are not homebuilt,
and neither are a lot of the peripherals. Most of us enthusiasts, however do
all our own repairs, and have replaced components in almost all areas of the
machines. Also, home-built add-ons are quite common.
> threads, but would have nothing to offer. These guys are the equivalent
There are plenty of groups which cover this interest. Please feel free to read
them :-)
> of those who lovingly refurbish ancient cars/steamengines. Great
Exactly.
> respect, but I don't want to actually have to look after one myself
> (well it would be fun I suppose .....). Oh and by the way, has anyone
It's great fun. I speak from expereince, having got about 10 TTL-based CPUs to
support :-).
> 2) Umm .. err .. Well my son would say something like 'Get real, man'.
> The man hours and cost of designing, testing and building a modern (say
> 1990!) computer complete with periferals would take any individual 10
> years+ even with help from the group. By then the parts would be
> obsolete and hard to get. The cost would be astronomic, the performance
> poor .... Nobody (OK only a very wealthy body) designs and build a car
> starting with stock steel sheet & bar, and a few $10,000 of machine
No, but plenty of people build model steam engines starting from metal bar.
> tools. Yup, we could all do a Model T type of design, but that's
> already been done. I do NOT include microcontrollers/PLA's etc here.
Well, I'll have to refer to my mad hobby of old machines here. There are plenty
of somewhat obvious extnesions and changes you could make to the old
architectures, and the only way to see what would really happen is to build them
Also, there are machines that for one reson or another are not available, and
thus functional replicas (using modern components) are a way to see what
programming such a machine would be like
Then there's the educational value. The extra experience and knowledge you'd
gain by designing a CPU from scratch would be enormous.
One day I must build my current dream-machine, the 'PERQ 6'. The PERQ was a
graphics workstation (the first), and I want to build a superscallar 32 bit
(at least) machine with other improvements. BAsed on my knowledge of the PERQ
2 architecture, I reckon I could desing and build the CPU in 6 months.
I am quite happy for CPU design to be only a small part of a.c.h.h. , but I
think it should stay here.
>
> 3) A support group for those meddling in computer hardware. Well here
Provided meddling means designing and building (i.e. getting out the soldering
iron), I would support this
> is 'difficult' is the PCB and development system. I am talking amateur
You don't need a PCB for a one-off. YOu could wire-wrap, verowire, or even use
links on veroboard.
> for a 68.... Doubtless there would be discussion about the relative
> merits. Here I include build your own process controllers, sound cards,
> laser light show, fuel/ignition processor, sattelite subsystem .....
Fine. I think all those are on-topic. I'm certainly not going to object to
microcontroller-based posts (or ROM/latch state machines, or ....).
> Except for the holes that is. Lots of useful tips could flow down the
> wires to a REAL hardware construction forum.
Discussions of constructional methods, pros/cons seem OK also. I for one would
be interested in PCB construction at home.
> Oz@upthorpe via Demon Internet
-tony
Matthew Gabriel
>As a general reply to the response about "Real homebuilt" computers.
>I enjoyed reading the returns, and I DO actually agree with the spirit
>of it all completely. The real question of what you want your forum to
>be breaks, I think, into three parts:
>1) A support group for old/worthy computers like PDP8, CBM4032 et al.
I disagree. This is not the "Classic Computers" forum, and most
of the "old timers" only mention and discuss it in passing. Designing
peripherals for these machines does count, however, under your own
"section 3".
>2) A support group for those wishing to design/build computers from the
>chip level to final machine.
>3) A support group for designing/building add-ons to existing designs
>to perform novel, expensive or interesting facilities.
>Perhaps I could take them one by one as to my view of the relevent
>points attached to each.
>2) Umm .. err .. Well my son would say something like 'Get real, man'.
>The man hours and cost of designing, testing and building a modern (say
>1990!) computer complete with periferals would take any individual 10
>years+ even with help from the group. By then the parts would be
>obsolete and hard to get. The cost would be astronomic, the performance
>poor .... Nobody (OK only a very wealthy body) designs and build a car
>starting with stock steel sheet & bar, and a few $10,000 of machine
>tools. Yup, we could all do a Model T type of design, but that's
>already been done. I do NOT include microcontrollers/PLA's etc here.
Well, there is a gentleman in New Zealand who designed the worlds
most advanced V-Twin motorcycle, using cutting edge materials fabrication
and design elements more advanced than those even aircraft manufacturers
use. He did it with a little help from his son in a shed
behind his house in his spare time. Now Mr. Britten has a real production
facility and engineering staff. But he started out with just a few ideas
and a lot of research, and created something totally unlike any other
machine of its type ever made, and advanced the art and science of the
motorcycle further than any before him, corporation or individual. I think
this is the model for most of us would-be hardware engineers. A lot of
research and determination can supplant multi-gillion dollar R&D
programs. Hell, Jobs and Wozniak scooped the entire computer industry
with their little home-built project, and started a revolution that is
still going strong today.
>3) A support group for those meddling in computer hardware. Well here
>the opportunities are huge. This is the area for all those really
>useful things that either cost a bundle or simply do not exist. We are
>talking design and hardware here. You CAN design and build a
>microcontroller based subsystem (or PLA) quite easily. What costs and
Agreed. This will probably take up 90% of the bandwidth here.
Pretty interesting stuff.
>Or perhaps I am talking total gibberish?
No. This is gibberish. GoooeeeeeynbdeyoknoooYEEP! Bingzow! Bop.
Eyeyeyeyeyeyeyeyrieereeeeeii! Zirt. Windows'95 will make all other OSs
obsolete. (Duck!) B)
>--------------------------------
>Oz@upthorpe via Demon Internet
>--------------------------------
SoupIsGood Food
O. Hotz de Baar
Message-ID: <312814...@upthorpe.demon.co.uk>
Date: Tuesday, Mar 21, 1995 07.55.21
Organization: None
Reply-To: O...@upthorpe.demon.co.uk
X-Newsreader: Newswin Alpha 0.7
Lines: 35
Build-Your-Own CPU+:
OK, there are some of you who will design and build your own CPU's out
of TTL for fun. Pretty amazing, but there you go. Obviously it's
soldering iron stuff so it should be in the forum. Me I think this
would have been quite appropriate 20 yrs ago, but who am I to comment.
And to build it in wire-wrap, 250+ packages WOW!!!, I would NOT like to
get that working 100%. OOF.
Obsolete Computer Stuff: OK so you want that to go to it's own forum. I
am not too sure about this. Firstly nothing scares off the
assemble-your-own computer people more than lots of references to PDP8,
Altair et al. People maintaining/using these machines are likely to be
of similar ilk and would make the forum more interesting. If you make
it so restrictive that it's down to 1 message per week, it will die.
Homebuilt periferals etc. Well, that just has to be in, doesn't it?
Homebuilt Process controller Stuff: Should be in too, I would guess.
Of course I have only been subscribing to this group for a couple of
weeks so what the heck would I know? You 'oldtimers' should decide for
yourself. I personally would like a forum to follow, and to get
tools/info as required (and on occasion to express my $2 worth).
Providing the threads are not in the hundreds/week, and comprise
technical chat from skilled people about real problems/solutions I
would favour it to be as wide as possible. That way there are enough
users to keep it 'vital', active, and continuing.
Tony Duell
sou...@enterprise.america.com "Matthew Gabriel" writes:
> O...@upthorpe.demon.co.uk ("O. Hotz de Baar") writes:
>
> >As a general reply to the response about "Real homebuilt" computers.
> >I enjoyed reading the returns, and I DO actually agree with the spirit
> >of it all completely. The real question of what you want your forum to
> >be breaks, I think, into three parts:
>
> >1) A support group for old/worthy computers like PDP8, CBM4032 et al.
>
> I disagree. This is not the "Classic Computers" forum, and most
> of the "old timers" only mention and discuss it in passing. Designing
> peripherals for these machines does count, however, under your own
> "section 3".
One thing some of these older machines (most of them?) had was a very clean and
elegant architecture. Many fanatics would love to see their favourite machine
either made more compact and less power consuming by the use of more modern
chips, or, more likely, would like to see an enhanced version with similar
features. So, as such machine would most likely be home built, then I think
they could reasonably be discussed here (anyone feel like making a 32 bit
PDP8-a-like?).
> Well, there is a gentleman in New Zealand who designed the worlds
> most advanced V-Twin motorcycle, using cutting edge materials fabrication
> and design elements more advanced than those even aircraft manufacturers
> use. He did it with a little help from his son in a shed
> behind his house in his spare time. Now Mr. Britten has a real production
Never underestimate hackers (in the truest sense). There is little that's
impossible
> >microcontroller based subsystem (or PLA) quite easily. What costs and
>
> Agreed. This will probably take up 90% of the bandwidth here.
> Pretty interesting stuff.
Indeed. Not my main interest, but as any machine I do build will certainly
use microcontrollers in the peripherals, I like to read such things. That's
quite apart from all the little embedded controllers I ought to build...
> >Oz@upthorpe via Demon Internet
> SoupIsGood Food
-tony
Tony Duell
O...@upthorpe.demon.co.uk "O. Hotz de Baar" writes:
> 18...@upthorpe.demon.co.uk>
> Message-ID: <312814...@upthorpe.demon.co.uk>
> Date: Tuesday, Mar 21, 1995 07.55.21
> Organization: None
> Reply-To: O...@upthorpe.demon.co.uk
> X-Newsreader: Newswin Alpha 0.7
> Lines: 35
>
> Build-Your-Own CPU+:
> OK, there are some of you who will design and build your own CPU's out
> of TTL for fun. Pretty amazing, but there you go. Obviously it's
Yes, I'll admit to being eccentric :-). But I regard it as an interlectual
puzzle, like crossword puzzles, or whatever, and probably abotu as useful :-)
Still, I enjoy coming up with wierd designs, and it's harmless...
> And to build it in wire-wrap, 250+ packages WOW!!!, I would NOT like to
> get that working 100%. OOF.
It's not that bad. At least if you've designed it you know how it should work.
Debugging commercial stuff (often with more chips - my 11/45 has about 1000
packages in the CPU, and another 400 in the disk controller !) is a lot
worse - you first have to fathom out what the designer intended.
>
> Obsolete Computer Stuff: OK so you want that to go to it's own forum. I
> am not too sure about this. Firstly nothing scares off the
> assemble-your-own computer people more than lots of references to PDP8,
> Altair et al. People maintaining/using these machines are likely to be
> of similar ilk and would make the forum more interesting. If you make
> it so restrictive that it's down to 1 message per week, it will die.
True in a sense. In the first place, old machine owners tend to be hardware
fanatics, as you _have_ to repair these machines to component level youself.
So at least posts concerning them would probably involve a soldering iron.
But the emphasis here should be on homebuilt, rather than home repaired
That said, I'm not going to object to occasional posts about these machines.
100 posts on (say) the DG Nova (to mention a machine I know little about)
are infinitely preferable to the PC crowd...
> ------------------------------
> Oz@upthorpe via Demon Internet
> ------------------------------
-tony