• Hello MLAers! We've re-enabled auto-approval for accounts. If you are still waiting on account approval, please check this thread for more information.

Dream peripherals (the return)

Here's what i want:

A way to make 68k mobos use far cheaper, far more available PC RAM!

I have oodles of 72pin simms for PC (8, 16, and 32mb) but I have to pay money for ram to add to my LCIII? what the heck.

 
??The LCIII does, in fact, use 72-pin simms. Feel free to stick in a 32MB simm from the PC world. 'Tis the same thing! It just has to be 100ns or faster (which should not be a problem).

 
PDS/Nubus IDE/ATA/RAIDI wonder: take an existing Nubus SCSI card and replace [some stuff] to make it an ATA card ... that still looks like a SCSI card to the Mac. In my dreams it's a case of replacing one IC ::)
Click to expand...
you could piggyback an acard ATA-SCSI onto the card ;)

 
This is the dream peripherals thread. The dream peripheral is a Nubus card that doesn't need a $100 adapter to run an ATA drive.

 
Does anybody have some sort of NuBus documentation? I've seen specs etc. for the various PDS slots but none for NuBus.

 
Re-reading the very first post in this thread I remember a do-it-yourself hardware project I read about in the year 1994: some computer magazine, the German c't, published complete instructions to set up an ADB interface hardware, with an additional RS232 port, several digital I/Os and even expandability to attach a chip card reader. Instructions how to program this device are given, as well. The author Carsten Meyer is very skilled in various electronics and programming and works for the publisher, still.

Here you find the article (in German language; probably the paper has been translated somewhere else):

http://www.heise.de/ct/94/11/230/

The company eMedia supplies an empty pcb (Euro 19.50) and a ready to use programmed EPROM (Euro 14.50) as well as hardcopies of the paper, still. A complete set of parts should be around a probably given 100 $ limit.

With the I/O capabilies one could set up a closed loop to control some other device, let's say switching off a toaster after detecting the bread is nicely browned with some homegrewn image recognition software for a QuickTake webcam in the kitchen ;-)

Did someone make this device or even did some useful liberation with it ?

 
Someone previously mentioned a dual-output video card for 68k Macs.

I'd just like one with DVI. As mentioned by others, USB would be nice, too. (While DVI wouldn't require any special drivers, since VGA vs. DVI could be handled purely in hardware, USB would require drivers be written.)

As a side note, I just discovered that my über-fancy ViewSonic 19" CRT (supports 2048x1536 at 60 Hz, or 1600x1200 at 75 Hz,) does not support 640x480 at 60 Hz! The frequency is too low for it. My Performa 600 defaults to that, and I can't use my BNC-to-DA-15 cable; I have to use the VGA cable with a dip-switch-equipped adapter to force it into 1024x768 fixed frequency mode, which somehow tricks the computer into sending out 640x480 66 Hz. (No, it won't display 1024x768.) I suppose I should just throw in a NuBus video card, but I haven't cracked the case open yet.

Another funny artifact of this monitor is that it will only go up to 1600x1200 over VGA, even if I override its DDC claims; yet it will go up to the full 2048x1536 over BNC. In spite of the two high resolutions, its phosphor grid can only realistically display 1280x960 without losing data. If I expand the picture to the very edges, 1440x1050 is ALMOST usable. (By "losing data", I mean that if I display a 50% black/50% white grid, like the default background on black and white Macs, I can see the individual black and white dots at 1280x960 or lower. At 1440x1050, I get moiré patterns on the edges, which have lower phosphor density than the middle. At higher than 1440x1050, I get moiré patters all over. At 2048x1536, the pattern looks completely gray.)

 
do-it-yourself / 1994: some computer magazine, the German c't, published complete instructions to set up an ADB interface hardware, with an additional RS232 port, several digital I/Os and even expandability to attach a chip card reader. /Here you find the article (in German /): http://www.heise.de/ct/94/11/230/
Click to expand...
there was also a workshop in the ct back in the days on how to build your own nubus card. though they still want 2€ for each article... :/

 
That's a nice little project there register. I wonder if the source code could be useful as a starting point for other ADB hacks.

 
how to build your own nubus card
Click to expand...
Would that be these articles?

Der NuBus in Theorie und Praxis, Teil 1: GrundlagenKnow-how, NuBus-Grundlagen, -Multi-I/O-Karte, 6821, Wrap, Prototyp, Burst-Transfer, Apple Macintosh II, Next, Doublet, Quadlet

c't 2/93, Seite 164

Der NuBus in Theorie und Praxis, Teil 2: PIO-Karte

Projekt, NuBus-PIO-Karte

c't 3/93, Seite 238

Guido Körber (cm)

Teil 3: Programmierung der NuBus-PIO-Karte

Projekt, NuBus-Karte, c't-PIO-Card, XFCN-Erweiterung

c't 6/93, Seite 198
Click to expand...
 
Does anybody have some sort of NuBus documentation?
Click to expand...
The Nubus developer's docs are still available on the Apple website, somewhere. I've got a copy saved at home.

 
A rambling thought or two:

ISA cards provide a lot of the missing functions we're talking about for older machines. ISA is a fairly simple bus to interface with as I understand it, especially if you're not dealing with multiple cards. 8 bit ISA has been interfaced with PIC microcontrollers to provide NICs, sound cards, IDE, etc

So how about a PDS-ISA adapter?

Or a universal card:

  • FPGA on a card with:
  • Clip-on socket headers for:
  • 030 PDS, '040 PDS, LC PDS, Nubus, ISA, DDR RAM, 68k/G3/x86 CPU etc, etc in either direction
  • Re-programming the FPGA allows:
  • Software pin remapping of all connectors
  • 68k or other softcore CPU runs A/ROSE or:
  • 8088/x86 running DOS * for driver support


* Or FreeDOS, whatever. The expander runs as a headless DOS computer just to support the ISA card and pass data back to the Mac. This being just one implementation. The CPU, software and pin mappings can all be changed in a flash, giving you a whole new card.

 
Thought I'd try and explain my universal card a little better.

It requires an FPGA with lots of GPIO pins. Every pin is brought out to 0.1" edge connections at the four sides of the card.

You pick it up and shove a Nubus card connector onto the bottom. Four microswitches mate with little notches you've previously cut out of the Nubus connector and provide a 4-bit ID. The card recognises this and loads the "I am a Nubus card" module into the FPGA. This contains the Nubus pin map, and the FPGA configures all pins correctly for a Nubus slot. It also contains a Nubus controller emulator.

Similar things happen if you plug an LC PDS/ 68k CPU male/ 040 PDS etc connector into the bottom edge.

You do the same at another edge with an ISA (or whatever) female connector. It loads the "ISA host" module.

Taking the case that you'll be using it today as an IDE drive controller, the following happens:

It pretends to be a known SCSI card to the Mac. Thus existing SCSI drivers can be used on the Mac side.

It configures itself as a DOS machine as far as the ISA card is concerned, and loads existing DOS IDE drivers.

In this way it acts as a middleman between the host Mac and existing, readily available ISA cards. This (in theory) takes a lot of the hardware and software development pain out of the loop. As time goes on, patches could be developed to allow it to natively support IDE drives in pretend-SCSI-PDS-card mode, and other functions as time and imagination permit.

Comments are welcome.

 
Wikipedia says: "Field-programmable gate array".

The FPGA is some kind of microcontroller that can be configured to perform complex logic operations. This can be used to program the behaviour of the device, for example to emulate a circuitry of discrete elements, or to emulate any other controller that might not be available or as easy programmable. A FPGA is a great solution to produce small lots of specialised controllers with least developing costs.

Best of all: developer kits start at about USD 100 (manufacturer Atmel, supplier Digikey, kit including software, breadboard, adaptors etc.).

 
Dev boards start a lot cheaper than that. A dev board probably won't do for the universal board though. Probably have to make up a new board.

There's no reason (short of time, cost and programming) that I can see why an FPGA couldn't handle the tasks described above. That doesn't mean it would be easy or cheap. It's on a par with some of the feasibility rants in the '060 upgrade thread.

The Xylo boards in the shop (see link) look pretty cool; USB and an ARM CPU on board as well as the FPGA

 
feasibility rants

Apple's guidelines for cards for PDS slots are very similar to NuBus cards, so there's actually not that much difference in developing one or the other.
With FPGAs it's not that difficult to build a card which will work regardless of which PDS slot it's plugged into. Although making it autodetect that might be a bit of a trick. It'd be nicer if the user set a jumper.

Of course, there's only one PDS slot in each machine. So a multi-function card such as IDE/USB mentioned above becomes attractive.
Click to expand...
 
You must log in or register to reply here.

Similar threads

bottleworks
Apple Network Server RAID w/NetBSD & in other Macs
Replies
1
Views
2K
dramirez
dramirez
David Cook
Macintosh IIsi SuperMac Spectrum/8 SI Restoration
Replies
0
Views
2K
David Cook
David Cook
David Cook
Stealth Macintosh IIfx No Chime and Bad F-Line Repair
Replies
5
Views
3K
imactheknife
imactheknife
Tek466
Looking to have a working Compact Mac system again!
Replies
5
Views
2K
LaPorta
LaPorta
David Cook
Technical explanation of why the LC, LC II, and Classic II have a 10MB RAM limit
2 3 4
Replies
61
Views
19K
Phipli
Phipli
Back
Top