Date: Tue, 3 Jul 2012 11:07:27 -0500
From: Mark Tapley <mtapley at swri.edu>
To: <cctalk at classiccmp.org>
Subject: Modern computers with docs (was: Re: PM 6100/60, was:
Powermac...)
Message-ID: <p0624080bcc18c0b8657a(a)[129.162.151.118]>
Content-Type: text/plain; charset="us-ascii"; format=flowed
At 19:48 -0500 7/1/12, ARD wrote (more or less):
> >More seriously, can you name a present-day computer where the
> >manufactuers do supply schematics, data on ASICs, and the like?
>
Ummm, you want the transistor-level prints on a Core 2 Duo CPU?
290 million transistors! Really, if put on paper, it ought to
fill a large room.
The last machines I ever saw where schematics were available to the
end users were the VAX-11/780 and the uVAX-II. The 780 schematics
were as thick as a phone book and 11 x 17" pages. The uVAX-II
was a lot smaller, it was a hierarchical description, but gave a lot
of insight on how the processor and interface chips were organized.
Jon
Does anyone recall anything about the availability of (cross-)
assemblers for 1802 development back in their day?
Not that I need to find an original, I'm disassembling and 'reverse-
sourcing' the firmware (~700 bytes) for some equipment from 1977 with
an embedded 1802, and just idly assessing what the likelihood is that
it was produced with a machine assembler vs. hand-assembled.
Date: Tue, 3 Jul 2012 23:53:51 -0400
From: Ethan Dicks <ethan.dicks at gmail.com>
To: "General Discussion: On-Topic and Off-Topic Posts"
<cctalk at classiccmp.org>
Subject: Re: Modern computers with docs
Message-ID:
<CAALmimn-hHdHA8xCf4CLfdkkvwCjzptq-tnh4V4AbwK=TvPT-Q at mail.gmail.com>
Content-Type: text/plain; charset=ISO-8859-1
On Tue, Jul 3, 2012 at 10:33 PM, Jon Elson <elson at pico-systems.com> wrote:
> > The last machines I ever saw where schematics were available to the
> > end users were the VAX-11/780 and the uVAX-II. The 780 schematics
> > were as thick as a phone book and 11 x 17" pages. The uVAX-II
> > was a lot smaller, it was a hierarchical description, but gave a lot
> > of insight on how the processor and interface chips were organized.
>
The 11/750 and 11/730 also had full schematics and assembly drawings
on thick slabs of 11"x17" paper, accompanied by several 8.5"x11" tomes
of technical descriptions. I think even DEC stopped providing that
level of documentation in most cases after about 1983 or so.
I remember plenty of detailed Qbus docs but I don't remember running
across the uVAX-II printset. That would be interesting to review.
I have a KA-630 print set, if anyone in interested. Probably it should
be scanned
and archived with bitsavers. It is 11x17, and about 3/8" thick.
Jon
Hi,
This is a generic fixit for PowerPC Macs but always worth a try:
- Unplug the unit from the mains.
- Remove the NVRAM battery
- Leave to stand for *at least* 2-3 hours
- Power up
The issue is the NVRAM (referred to as PRAM on a Mac just to be
contrary) becomes corrupt as the battery runs down. If the machine has
been stood unused for a long time it may have this aforementioned
amnesia.
I have, however, heard a lot of sorry tales of PowerMac G5s suffering
all manner of weird problems with PSUs, leaking liquid coolers and
more. It's worth thoroughly checking the whole system over before
proceeding.
--
Mark Benson
http://markbenson.org/bloghttp://twitter.com/MDBenson
On 4 Jul 2012, at 09:37, Rob <robert at irrelevant.com> wrote:
>> There is a chance, I reckon, that the RasPi may end up like the IBM
>> PC. Quite respectably successful in its own right, but more important
>> for the whole market sector it spawned of enhanced-but-compatible
>
> I think that's actually the point!
Exactly. They aren't doing this for the profit, they are doing it for
the drive into the community and to get people programming as a hobby.
While yes I appreciate people have battered on about the documentation
being part under NDA and that sucks for driver development, really
that's not the part they are most interested in leveraging, they want
a small device that people, kids especially, can doodle about with
languages like C, Python etc on to get them enthused about creating
software in a modern context in order to inspire a new generation to
grow up with computer programming on the brain.
The project is at an early stage, really largely still late
development, but it's got the important thing - large amounts of
traction with the community and a friendly introductory price tag.
--
Mark Benson
http://markbenson.org/bloghttp://twitter.com/MDBenson
Date: Tue, 3 Jul 2012 14:19:58 -0700 (PDT)
From: David Griffith <dgriffi at cs.csubak.edu>
To: cctalk at classiccmp.org
Subject: IMSAI switch paddle replacements
Message-ID:
<alpine.DEB.2.00.1207031419440.9191 at sleipnir.cs.csubak.edu>
Content-Type: TEXT/PLAIN; format=flowed; charset=US-ASCII
Can someone point me to a source for replacement switch paddles for the
IMSAI 8080? Todd Fischer of imsai.net tells me that he has red ones, but
no blue ones left.
In this day of 3-d printers, maybe you can find somebody to print some.
You ought to
be able to get the ABS filament in blue.
Jon
Date: Mon, 2 Jul 2012 20:11:01 -0400
From: Michael Thompson <michael.99.thompson at gmail.com>
To: cctech <cctech at classiccmp.org>
Subject: Strange Core Memory Behavior in a PDP-8/L
Message-ID:
<CAH1BU=9z5sOj5rw0dzb1hPe4PTrMvYpjiOv2dPRnrXJ-5CuqwA at mail.gmail.com>
Content-Type: text/plain; charset=ISO-8859-1
The RICM is still wrestling with the core in the PDP-8.
After replacing some diodes on the core stack we have all addresses
working.
We observed an interesting core memory behavior during our debugging
last Saturday.
We started the memory alignment procedure by looking at the
STROBE FIELD 0 signal and the amplifier output on pin E1 of the sense
amplifier. The STROBE signal was very late compared to Figure 5-6 in
the 8/L Maintenance Manual. We ran a short JMP loop and adjusted the
relationship with the trimpot on the M360 delay module. When we
halted the
processor and tried a examine core we only got just zeros.
We adjusted the M360 delay back where it was and single step worked
again. We found that the strobe-to-one-bit relationship was almost
100ns earlier when in single-step than it was with the processor
running. We checked the whole timing path from MEM START at pin N2 of
the M113 in slot C03, through all of the gates, delays, and
flip-flops, and found no timing difference between single-step and
running. Right now it looks like there is a 100ns delay difference
between the READ(1) signal that turns on the current in the core and
the bit signal showing up on the E1 pin of the sense amplifier when in
the single-step and running.
Is this normal behavior?
The first thing that comes to mind is one-shot recovery time. I might
guess that some
component has degraded, and the one-shot has a different delay when it
has been
recently triggered vs. when it has sat for a while. (I'm assuming a
classic 8, with
discrete transistors, not a later machine built with ICs.) Most
specifically, there probably
is a transistor that resets the capacitor at the end of the one-shot
cycle, and if that
transistor is weak, it may not get the cap fully discharged before the
next memory
cycle starts. Most core memories had circuits to adjust drive current based
on core plane temperature, maybe some adjusted timings as well. Also,
check for
dead decoupling capacitors, these could allow noise when running to alter
timings.
Jon
[I hope this is on-topic; I believe the machine is at least 20 years old]
I have a NEC Spinwriter 5525 printer that is available for the cost of
shipping (free if you pick it up). The printer is a wide carriage and
appears to have a RS232 serial interface. I do not know if the
printer works or not.
The machine is located in Langdon Alberta Canada (postal code is T0J
1X1) which is approximately 10 minutes east of Calgary.
The machine is rather heavy. I estimate 50 pounds or more. If there
is no interest, the machine is headed to the e-waste recycling.
I can send pictures upon request.
Contact by e mail:
i a m v i rt ihatespam u al @ @ @ g ma il . c om <-- remove
spaces and ihatespam
Thanks!
--barry
At 21:58 -0500 7/2/12, Cameron wrote:
>Fortunately they only made this assumption a few places, but figuring out
>what broke was maddeningly fiddly.
...and for the record, the resulting (I assume) TenFourFox rocks!
http://www.floodgap.com/software/tenfourfox/
There are multiple websites I use daily that load faster and better
on TenFourFox than on Safari on my PowerBookG4; the (on-topic) iMac
G3 is a no-contest win for it.
Cameron, thank you!
Anyone else surfing the web with a PowerPC Mac, I commend the above
site to your attention.
(no relation, just satisfied user).
--
- Mark 210-379-4635
-----------------------------------------------------------------------
Large Asteroids headed toward planets
inhabited by beings that don't have
technology adequate to stop them:
Think of it as Evolution in Fast-Forward.
> So, if there is someone out there that could pick these items up
> (Woodridge, IL - a southwest suburb), I would appreciate it. Time may
> be short. Please let me know offlist.
I just received a note saying this must be done today or tomorrow, as
the family departs on Wednesday!
--
Will
You never do know what you'll find on Craigslist. This time my bad
terminal habit gained another 200lbs of trouble with the acquisition
of two very interesting Teletype Model 40 CRT terminals, attractively
badged as the Bell System Dataspeed 40. The accompanying printer came
with them, as well as two very thick volumes of the Shop Manual for
these machines (which may be the same as what is already on Bitsavers
- I will check.) A very clean KSR33 Teletype was thrown in to round
out the deal. Here are the initial pics (shop manuals are stuck
somewhere in the car, will add them soon:)
https://picasaweb.google.com/102190732096693814506/TeletypePickup
Not much info out there on these terms. In fact, given the number of
labeled cards inside them, I'm not even sure they're strictly
terminals. I have read about local storage options, cluster
controllers and other peripherals which suggest that TT or Bell may
have tried to sell them as an office computing solution as well as
terminals. All I know is they are extremely heavy, probably thanks to
the massive PSUs in the bases. The CRTs slide off the two stalks in
the base, which contain the power and signal connectors. The
keyboards detach from the front of the base as well. Very modular,
novel design - and essential if one was to have any hope of lifting
these things.
The "printer" is only a shell. I thought at first it may have been
harvested for parts (I was told they came from the estate of a
deceased IL Bell repair instructor) but given the connections inside
the enclosure, I am wondering if it was designed to accept a standard
Teletype mech.
Of course, any knowledge out there that can be added here is appreciated.
(No I haven't powered them on yet...)
-j
--
silent700.blogspot.com
Retrocomputing and collecting in the Chicago area:
http://chiclassiccomp.org
Hi,
I'm doing a circuit diagram of a SA1100 Disc Controller built
with a Z80 CPU. On the controller, I have two blue devices
lables "Z80 CLK DRV" on the board, and HPI-1014-782 on the
device itself. the "-782" is the date of production most
likely. I have no idea about the pinout of those devices nor
do I know where to finde datasheets to make a component for
my CAD software.
Heare you'll find a picture of such devices I mean (different
board, but same devices)
http://www.picfront.org/d/8Ik2
Any hints of what this is, how the pinout is, and where I can
find a functional description of this device would be welcomed.
Greetings, Oliver
We have an excellent opportunity for an apprentice applicant to join a rapidly expanding company.
An at home Key Account Manager Position (Ref: 92552-205/6HR) is a great opportunity for stay at home parents or anyone who wants to work in the comfort of their own home.
This is a genuine offer and not to be confused with scams!
The successful candidate must have the ability to handle calls efficiently whilst maintaining the highest levels of customer service and being courteous.
Applicants must have an excellent telephone manner, have a friendly approach, excellent communication skills and be computer literate.
You must have the ability to type and talk at the same time to customers,
as you will be taking customer details over the phone and inputting data onto company database.
Requirements: computer with Internet access, valid email address, good typing skills.
If you fit the above description and meet the requirements, please apply to this ad stating your location.
You will be processing orders from your computer. How much you earn is up to you.
The average is in the region of US$600- US$750.00 per week, depending on whether you work full or part time.
If you would like more information, please contact us stating where you are located and our job reference number - 92552-205/6HR.
Please only SERIOUS applicants.
Our contacts: Josue at careerin-finance.com
Thank You!
Folks,
I have happened upon a fairly nice-sized Data General Power Supply
board. It's about 8" square with 2 x 120-pin plug connector the length
of the board. It says "650W AC autoranging power supply, (C) 1996 Data
General 107003838_02/04" and has a very tiny printed label with a part
number and serial number. The part number I've made out as 665643739,
although the printing is so tiny it could be?885643739. My wife says
005043739. Google turns up nothing.
I actually have two. These are essentially NOS parts and seem unused/clean.
Here's a photo:?http://i.imgur.com/ZyYYq.jpg
Is there any interest in these? What are they?
--
-Jon
Jonathan Katz, Indianapolis, IN.
I am looking for some help in the Chicago area. Very recently, an IBM
mainframe old-timer passed away. He has some old manuals and so forth
that the family would like to save, but they understandably are too
busy. Of note is a 16 mm film concerning the release of OS/360, which
was to go to CHM.
So, if there is someone out there that could pick these items up
(Woodridge, IL - a southwest suburb), I would appreciate it. Time may
be short. Please let me know offlist.
--
Will
Does anyone know what the spec and polarity is of the 12V PSU port for the AlphaServer 4100 Remote Console. I'd like to be able to use it when the machine is offline.
--
Mark Benson
http://markbenson.org/bloghttp://twitter.com/MDBenson
Date: Sun, 1 Jul 2012 21:42:31 +0100 (BST)
From: ard at p850ug1.demon.co.uk (Tony Duell)
To: cctalk at classiccmp.org
Subject: Re: Z80 CLK DRV
Message-ID: <m1SlQyb-000J4jC at p850ug1>
Content-Type: text/plain
I cna't help wit hthe pinout, but I can guess the function.
IIRC the clock input of the S80 is not actually TTL compatible, it needs
to be pulled up to the 5V line.My guess is that thes are clock driver
cirucits for this signal, they take a TTL level singal in and give out
the genuine 5V signal for the Z80 clock input.
I used to do a lot of Z80 designs, most with classic Zilog Z80s, but
also some later
ones with Harris CMOS Z80 clones for battery-powered applications. I don't
recall a lot of difficulty with the clock generation, I used classic
74xx chips.
Generally I started with a faster xtal oscillator and divided down with
something
like a 74HC161, so maybe that was all that was needed to get the right
clock swing.
Jon
Hi all,
does anyone have a hint how to configure a Philips P2701E serial text
terminal. I found zero Information in the net.... Are these some kind of
rebadged equipment?
Regards,
Wolfgang
Got a bit of a scare this week. I have most of my equipment stored and I received a call indicating
that the lock on one of my units was missing. The management put a "temporary" lock on the unit
until I could get down and survey what was going on (and put a new lock on it).
I went down this morning and there in front of the door to my unit was my lock that had been cut off
(I suspect with bolt cutters). Fortunately, there doesn't appear to be any mischief (other than having
to replace the lock). I suspect that someone was confused (and probably lost the key to their lock)
and opened my unit. When they saw what was in it they moved on. The reason that I suspect this
was that nothing was disturbed and there are several obvious items near the front that would've probably
gone missing if theft was the motivation.
I put a new high security lock in place of the old padlock and will probably replace my other padlocks
with high security locks to prevent this sort of thing in the future.
TTFN - Guy
Continuing to make progress on the 11/04. The H7441 supplies are still to
be tested. The 24000uf input capacitor is faulty on one and while waiting
on the replacement I also noticed one of the 1200uf capacitors in the
output CLC filter was leaking. Now waiting for parts to replace both CLC
filter capacitors (560uf and 1200uf) in each supply before I power them up.
Moved on to the H745. The two output capacitors test file but the big
31000uf electrolytic after the rectifier is faulty and needs replacement.
Any suggested on a replacement other than same spec... it seems a bit
highly spec'ed for what this module is doing.
Regards
Andrew
Hi all,
Do you guys know of a failsafe way to check if a Teletype is for 110
or 220V? I've bought one which has a power connector which seems like
US mains. The motor is 50Hz, but I guess that won't make a difference.
Any quick way to check without completely dismantling the machine?
re,
Sander
--
~ UNIX is basically a simple operating system,
? ? ? ? ? ?but you have to be a genius to understand its simplicity. ~ dmr
cctalk-request at classiccmp.org wrote:
Message: 6
Date: Sat, 30 Jun 2012 01:47:24 -0300
From: "Alexandre Souza - Listas" <pu1bzz.listas at gmail.com>
To: "General Discussion: On-Topic and Off-Topic Posts"
<cctalk at classiccmp.org>
Subject: Adapting linux for other ARM devices. Was:Raspberry Pi and
America,
Message-ID: <08b401cd567b$8e3cd550$6600a8c0 at tababook>
Content-Type: text/plain; format=flowed; charset="UTF-8";
reply-type=original
An interesting question that may fit on this list
There are lots of old ARM devices around. One that is very common in
Brazil is the "easybox" internet computer which is an ARM device with an ISA
slot (for originally a modem and optionally a NE2000 network board), some
peripherals, video output and remote keyboard. These can be had for (very)
cheap. How hard is to port e.g.: Angstrom or something like that for a
system like this?
Maybe I hadn't expressed myself well. I want to UNDERSTAND how the port
is done
Well, depending on the exact flavor of ARM CPU, it could be easy or
VERY, VERY hard.
The problem is that the ARM CPUs may all have essentially the same
instruction set,
but the on-chip peripheral configuration can be quite different. Also,
some chips had
significant bugs that needed workarounds. The ARM architecture spans a
wide range
of performance and capability.
The Beagle Board is an amazing system on a 75 mm square board, with USB,
HDMI
256 M RAM, SD memory card and runs on less than 2 W power. You can run a
full Linux Ubuntu system on it. I use them for net-attached devices
that export
TCP services or Glade GUIs to control little boards that stack with the
Beagle.
If you have an existing port for a related chip, or a board that uses
the same ARM chip,
it may not be that difficult. There will be a source code base and a
development
tools environment. You mostly customize the list of drivers to be
included, possibly
providing some parameters to select what peripherals are enabled and how
to route
them to the board.
If you are starting with a generic source tree and a chip that has not
has Linux ported
to it (or the port has not been updated in years) then it can be QUITE
messy.
I built an updated kernel for the Beagle board ON the beagle board, it
took about
a day to compile the kernel. That's why they use toolkits to do it on a PC.
Jon
