>From: "Vintage Computer Festival" <vcf(a)siconic.com>
---snip---
>
>None, I just felt like pointing out that whoever decided to use the plus
>sign for OR was most likely insane.
Hi
Actually from a mathematical sense, it does make sense. The
OR operation can be mapped to the plus operation and the
AND operation can be mapped to the multiply operation. One
finds that when doing this, many mathematical rules correlate
nicely. That is why it was done. Rules like associative and
communicative make sense. It wasn't just wild insane mad men.
When one understands the reasoning fully, one can except
the choices made. It is just that in your schooling, you
were never exposed to much Boolean algebra. That is too bad
because so much of what we do with computers requires a
level of understanding of how it works.
Dwight
>From: "Steve Thatcher" <melamy(a)earthlink.net>
>
>you create your OR functionality with diodes - use two diodes (or as many
inputs as you need) such as the 1N4001 for your inputs. Tie the cathodes
together and attach that to the relay coil. Apply "logic high" (whatever voltage
your relays operate at) to either (or any) diode anode and the relay will turn
on. The relay can be the next stage and does not have to be there to implement
the OR.
>
>best regards, Steve Thatcher
>
Hi
You don't need the steering diodes if the lines that provide the
terms don't cross talk to other coils. This is pure relay logic
and works with AC relays as well as DC. The diodes are only
needed when you share complex terms that use the same stack of
contact terms but would otherwise cross talk. The diodes allow
one to optimize the number of contacts used by sharing common
terms.
Each relay contact can be used as a term in an equation. Normally
open contact can provide inverting function and normally closed
contact can provide isolation buffering ( needed for AC but can
be optimized with diodes in DC ).
Designing with relays takes a slightly different mindset than
with normal logic, like TTL. You have to think in terms of
inverters and transmission gates. There are no NAND or NOR
gates. The function of NAND and NOR is constructed by sharing
the contacts of several relays and then using that shared
stack to control the output ( that may be another coil of
a relay ).
Stacking contacts can provide AND like function and paralleling
contacts can provide OR like function. Choice of normally
open contacts provides OR and AND with inverted inputs,
the Boolean equivalent of NAND and NOR.
This takes a little more thinking of design because with double
throw relays, one has to think of the source as providing the
inverting and non-inverting signals. The contacts only provide
OR and AND, depending on how they are arranged.
Dwight
>
>Question: Wouldn't it be easier to implement an OR gate with no relays at
>all (just two wires joining together)? Or would you need to use the relays
>to keep the output voltage/amperage regulated?
>
>
>Cheers,
>Ade.
>
>
I was just recently given a good sized collection of early Byte
magazines. I've inventoried them, and it is a complete series from
Issue #2 (October 1975) all the way through to Volume 8, #8 (August
1983) - EXCEPT for Vol 6, #7 (July 1981).
.. So .. if anyone should happen to have a copy of Byte Vol 6, #7
(July 1981) that they would part with - please let me know.
I also wouldn't mind completing the 1983 year, so Sep, Oct,
Nov, Dec 1983 would be of interest as well.
[I'd *REALLY* like to get the one missing on the other end as well,
(Issue #1) however I expect it would be much harder to find!]
Regards,
Dave
--
dave04a (at) Dave Dunfield
dunfield (dot) Firmware development services & tools: www.dunfield.com
com Vintage computing equipment collector.
http://www.parse.com/~ddunfield/museum/index.html
>Part of the issue, for modern machines, is that the hardware gets in
>the way. If you change code, the I-cache doesn't change (in machines
>I'm familiar with) so the old instruction may still execute. This
>makes instruction modification a tricky business. Debuggers do it,
>operating systems do it -- and both have to pay attention to all the
>cache magic you have to perform for it to be right.
This reminds me of a point I wanted to make earlier and forgot - another
big issue us that SOFTWARE gets in the way! - todays computing world is
much more oriented toward sharing code, and many systems support multiple
execution threads of a single copy in core - it causes bad things to
happen when one of those threads changes the code :-)
Regards,
Dave
--
dave04a (at) Dave Dunfield
dunfield (dot) Firmware development services & tools: www.dunfield.com
com Vintage computing equipment collector.
http://www.parse.com/~ddunfield/museum/index.html
I used to use a lot more self-modifying code in my early days then I do now,
you could save precious bytes of memory by storing data values into the
immediate operand of instuctions which used them --- I recall a "turning point"
when one day I spent a fair bit of time debugging a problem which was related
to two level deep self-modifying code - ie: Modified an instruction so that
it would subsequently modify another instruction differently - I started to
believe some of the nay-sayers about the evils of SM code after that :-)
Case where I've *HAD* to use SM code: 8080 general purpose monitor - no way
to read/write arbitrary I/O ports - have implemented it both as directly SM
code in RAM based monitor, and as a "built" subroutine in RAM for ROM based
monitor. Same for 8051 (except you have to overlay CODE and DATA address
spaces obviously - this is fairly normal for a debugging environment).
Cheers,
Dave
--
dave04a (at) Dave Dunfield
dunfield (dot) Firmware development services & tools: www.dunfield.com
com Vintage computing equipment collector.
http://www.parse.com/~ddunfield/museum/index.html
if you have telephone relays, then buzzing is not an issue. The ones I used years ago would "always" make contact. I used the N.O. contact to apply power to the relay once it received a pulse. I can't tell you now how long the pulse had to be (too long ago). Even if you wired the relay as a buzzer, it still made complete contact before returning to the open position. It would "buzz" at around 10-20hz.
>I can do a latch... where (in) pulls in the relay and a NO contact
>connects
>v+ to keep the relay in.. but everything I tried just vibrated... :^\
>
>In a relay computer wouldn't one need flipflops for memory?
>
>
>
You just have to think about what happens while the relay is
actuating and all the contacts may be open ( break-before-make ).
Otherwise, you'll be making a lot of nice buzzers. More than
one phase of clock makes things easier.
Dwight
Hullo,
As a result of me taking over another (Dutch) collector's DEC systems,
I now have filled a major gap in my own collection: the Pro 3xx series
have been added. That is to say: I have a 350 in the US (a VAX console),
and this lot came with both a 325 and a 380. The 325 seems to be a
field-upgraded one, as it comes in the "usual" Pro box, not the smaller
one, and it was stuffed with RAM, options and a hard disk. So, I guess
I am still looking for the "small" 325 ;-)
Anyway. The lot also came with a truckload of boxes (binder boxes,
ya know) and floppies, so I am now Teledisking all those floppies for
later use. I'll add these to my Archive, under the Software>POS
section.
All manuals will be auto-feeder scanned later, and the results will
also be in the Archive, under Hardware>PRO300.
Is anything PRO-related online already, other than the update.se
archive with Venix and some P/OS images?
I seem to have:
- P/OS V1.5
- P/OS V2.0
- P/OS V3.0
- P/OS V3.2
- PRO/TK V3.0
- PRO/TK V3.2
- Cobol-81, Pascal, F77, Forth for TK
- DECnet/PRO V2.1 plus Toolkit
- PRO/Sight V1.0
- PRO/Synergy V1.0
- PRO/COMM V3.0
.. and more to unpack.
If anyone wants copies or a trade, lemme know off-list !
PRO-related hardware is always welcome, too... *hint*
Cheers,
Fred
--
Fred N. van Kempen, DEC (Digital Equipment Corporation) Collector/Archivist
Visit the VAXlab Project at http://VAXlab.pdp11.nl/
Visit the Archives at http://www.pdp11.nl/
Email: waltje(a)pdp11.nl BUSSUM, THE NETHERLANDS / Mountain View, CA, USA
>From: "Vintage Computer Festival" <vcf(a)siconic.com>
>
>On Tue, 28 Sep 2004, Steve Thatcher wrote:
>
>> when someone can pick up a logic description and understand it WITHOUT
>> having to figure out what each sytmbol means in the context of the
>> usage, then it is simply more understandable. If I picked up something
>> that had two binary numbers wiht a PLUS sign inbetween, I would not
>> assume that it meant OR, only if you would dig further does the PLUS
>> sign make sense. If you simply said OR, then there is no confusion and
>> it is self documenting. I think Sellam was trying to make this point
>> with regards to symbology.
>
>Right. What I was getting at is that there were plenty of other symbols
>on the typewriter or keypunch to choose from.
>
>Having gotten the explanation as to why those operations were chosen, it
>does make sense and is somewhat elegant (I'll reluctantly concede), but
>it's still stupid as far as I'm concerned. Why didn't they use * or X
>for AND then?
Hi Sellam
Some do use * or X. Would that have helped you? Although, others
seem to to describe things like set theory, I can show a simple
example of the practical use.
A.(B+C) = A.B + A.C ( just like ordinary math )
A+(B+C) = (A+B)+C ( again, just like ordinary math )
Other than that, I don't think there is a good reason.
Dwight
>From: "Ron Hudson" <ron.hudson(a)sbcglobal.net>
>
>I had, at one point, sat down and tried to devise a JK flipflop out of
>relays..
>
>I can do a latch... where (in) pulls in the relay and a NO contact
>connects
>v+ to keep the relay in.. but everything I tried just vibrated... :^\
>
>In a relay computer wouldn't one need flipflops for memory?
>
>
>
Hi
The problem you have is controlling delays. In designing things,
you need to consider things like setup and hold timing as well
as race conditions. In normal circuits, a flip-flop is composed
of two latches with opposite clocks. In normal design, one
needs to either make these two clocks non-overlapping or
if one clock slightly overlaps the other, the data path between
a previous flop and the next have sufficient delay that they
won't race.
Memories are normally composed of simple latches and not flipflops.
Most uP use mostly latch based design and not flop base. Most
ASIC's are still using flop based design but they don't have the
same speed requirements that uP's have.
Another problem you may have been having is that you didn't
consider the contacts as being break-before-make or make-before-
break types. As was mentioned earlier, some of the relays
have things like shunting coils ( loop of copper ) added
to allow the relay to hold longer. One can also do things
like add capacitors in parallel to the coils to hold a few
milliseconds.
It is easiest to just use some overlapping clocking system.
You just have to think about what happens while the relay is
actuating and all the contacts may be open ( break-before-make ).
Otherwise, you'll be making a lot of nice buzzers. More than
one phase of clock makes things easier.
Dwight
