Hi guys,
Does anyone have any recommendations for a good book on the 68000/68010
CPUs (covering assembly language, the programming architecture and the
interrupt system as a minimum)?
I'm trying to disassemble the AT&T UNIX PC Boot PROM and Loader, and
figure out why the Loader doesn't seem to think my WD2797 emulation is
providing valid data... As I said before, it's probing the FDC, reading either two or four sectors, then giving up and trying the HDD instead....
--
Phil.
classiccmp at philpem.me.uk
http://www.philpem.me.uk/
Hi guys,
I've got my 3B1 emulator "mostly working" in that it runs the Boot PROM,
sees the floppy disc in the drive, and proceeds to boot from it, getting
as far as the Loader.
When it gets to the Loader, I get the following display:
AT&T UNIX(tm) pc
Loader version 3.51
Copyright (c) 1985, 1986
AT&T
All Rights Reserved
Searching floppy disk...
####
Searching hard disk...
... and it stops there. I can tell from the emulator log that it's
trying to get the hard drive controller to read CHS 0:0:0 and DMA the
data into RAM at 0x77830, but because the HDC isn't implemented, it
locks.
What I expected was for the Loader to pick up the boot files on the
Diags disk, boot from that, and ignore the HDD. Does anyone know what
"typical boot behaviour" is for a 3B1, 7300 or UNIX PC, when booted from
the Diagnostics floppy (Foundation Set, disk 1) ?
This is a bit of a head-scratcher -- I'm trying to figure out if there's
a problem with my FDC driver (wouldn't be the first one) or the
DMA/interrupt controller, or if the Loader really needs a hard drive
controller (or a really good fake) to boot the system.
I'd also really like to know why the DMA controller has two separate
direction control bits -- DMAR/W- and IDMAR/W-... this seems downright
silly, though in keeping with the rest of the TechRef. My "annotated
edition" corrects about a dozen minor and major errors in the register
set descriptions, and adds a bunch of informational sticky-notes and
scribbly comments to reinforce certain points. Ewwww...
If anyone's interested in playing with my emulator -- go to
<http://www.philpem.me.uk/code/3b1emu/>. Hit the link under "Mercurial
repository", then ".tar.bz2" to get a Tarball of the sources. Untar it.
Grab the boot PROMs, and put them in a directory called 'roms' as
'14c.rom' and '15c.rom'. Use IMDU (Imagedisk utility) on a DOS PC (or
inside Dosbox) to convert the Foundation Set disks from IMDs to binary
files, then copy the first Foundation disk (Diagnostics) as 'discim'.
Compile (you'll need libsdl, aka the Simple DirectMedia Layer) and run.
I know the code is a mess, patches to rectify this (or any of the other
millions of bugs) would be almost certainly be accepted :)
There's also no keyboard or mouse emulation yet, just the CPU, video,
RAM, ROM and a basic memory mapper and DMA emulation. As for Ethernet
emulation... that's on the "maybe later" list, right after "learn how to
send and receive Raw Ethernet frames on Linux".
Thanks,
--
Phil.
classiccmp at philpem.me.uk
http://www.philpem.me.uk/
Hello Everyone!
I would like to identify some mysterious old SIMMs:
- module P/N appears to be "PS 91/344J"
- 30 pin, single sided, 9 chips "SIEMENS HYB514100BJ-60"
- "MADE IN U.S.A." probably manufactured in 1994
I have failed to find explicit information about these chips,
but I'd guess they are 1Mx4 so 9 of them would make a nice
4MB module with ECC/parity.
What kind of machine could these have been made for?
Thanks,
Sab
Hi, All,
I have had no luck googling any pictures of the short rack for the
PDP-11V03 system (CPU + RX01 in a formica-topped office-friendly rack)
that I *think* is called an H9610. For that matter, I was largely
unsuccessful in finding pics of most DEC racks by rack part number
(i.e., finding an 11/70 picture is easy - finding an H960 picture is
not).
Textual material for various rack configurations abound, but not
pictures accessible by Google image search. If anyone knows of a
gallery of rack pictures, I'd be grateful if you'd share the URL.
Thanks,
-ethan
> Subject: Re: 1970s TTL specs and prices
> To: cctalk at classiccmp.org
> Message-ID: <4D006B2A.5090407 at bitsavers.org>
> Content-Type: text/plain; charset=ISO-8859-1; format=flowed
>
> On 8/20/10 11:23 AM, Tom Gardner wrote:
> > For some Computer History Museum work I need information on 7400 series
> Flip
> > Flops (S and Normal, DIPs) circa 1973 (anything 1970-75). Anyone have
> any
> > maximum clock speed and OEM volume pricing information on parts such as
> 7473
> > thru 79 or 74106-116?
> >
>
> At introduction (May 1966) The 7474 (25Mhz toggle freq) was $11.40, $9.10,
> $7.70
> quantity 1/25/100, respectively
>
> The most expensive TI TTL part at that time was the 7491 9 bit shift
> register at
> $31.25, $24.85, $21.25
>
> Their most expensive IC was the linear SN354A Demodulator/Chopper at
> $145, $116, $99
> ------------------------------
There is a fairly complete set of IC Masters at the UC's Northern Reference
Library Facility, Richmond CA, http://www.lib.berkeley.edu/NRLF/. They have
a public reading room and will pre-pull material per an email request,
nrlfreq at library.berkeley.edu.
As it turns out I will be there next week doing some research on DRAM using
all the IC Masters and can see what is in the 1973 version about various
TTL.
FWIW, I frequently use a learning (pricing) curve off a per 100 price to get
a reasonable estimate of the high volume OEM price. You will have to
justify the learning percentage (2% to 10%) and the OEM volume (multiple of
annual volume of a high volume system?) to come up with a price. For
example using the 1966 7474 at $7.70 with a 5% learning curve and a 100,000
OEM volume has an estimated OEM price of $4.60. You will have to do some
research and thinking to come up with the values u use.
Tom
Date: Mon, 06 Dec 2010 17:33:50 -0500
From: "Curt @ Atari Museum" <curt at atarimuseum.com>
Subject: Re: Drive Type International Memories, Inc. (IMI) 5012H
> Weren't IMI's used in much of the earlier Corvus drives too?
Yes, in fact Corvus later acquired IMI AFAIK.
Cromemco also exclusively used IMI drives with their WDI and WDI-II IMI
controllers before they finally went to an ST-506 controller, both the large
7000 series and then the smaller 5000 series; note that the 5000s were
available with either the 40-pin IMI interface or a standard ST-506
interface and the model numbers are not necessarily unique.
mike
>Message: 26
>Date: Fri, 10 Dec 2010 00:27:09 -0600
>From: Doc Shipley < doc at vaxen.net >
>Subject: Re: Wow! Lego Antikythera device -- and it works
>Evan Koblentz wrote:
>> http://therawfeed.com/apple-engineer-re-creates-2000-year-old-greek
>
>?? I thought the Antikythera Research Project decided a few years ago
>that the original did NOT use differential gearing.
>
>?? To be perfectly honest, the fact that the Antikythera incorporates
>the first known User Guide impresses me as much as the technology....
>
>????????Doc
>
The report in The Register ( http://www.theregister.co.uk/2010/12/10/lego_computer/ ) includes some additional links, including one to an interesting article in Nature News from Nov. 24th at http://www.nature.com/news/2010/101124/full/468496a.html .
Bob