Editorial & Spectrum News

SCENE+ is published by:
L. C. D.
Prager Str. 92/11/12
A-1210 Vienna, Austria
Tel: (01) 272 34 63 (within Austria), or +43 1 272 34 63 (outside Austria)
All classifieds and advertisements are free of charge, as long as they relate to the Spectrum or SAM, and can take any format: text, screen, animation, etc.

Hello SCENE+ readers. Welcome to issue 23! I spent the past few months trying to improve my mouse driver for Atari ST/Amiga mice connected to the Kempston joystick port (does anyone know a good abbreviation for that?). However, it proved impossible—in my opinion, I had reached the absolute maximum achievable in BASIC. Recently, though, as I dove deeper into machine code, I did manage a small improvement. From now on, my mouse driver no longer has to use the screen bits, because the bits are tested and converted using a small MC routine. The routine didn't get any faster, but it is certainly much more elegant, and I learned something new about MC in the process.

Because I am learning more and more about MC, starting from issue 25 I will be listing all Z80 mnemonic codes along with their decimal equivalents in alphabetical order, formatted so that they can be easily printed out (three screen pages per paper page) to keep them always within reach. I might also write a custom assembler for MC beginners and myself.

Both of my MB02 interfaces are finally working properly now, which should please Thomas. On the first one—the caseless model with 128K—I carelessly managed to burn out the 74LS244 chip. I have now replaced it (salvaged from a hard disk interface) and also fitted a massive heatsink onto the DMA controller so that it doesn't get so scorching hot during operation.

With the second MB02 (equipped with 512 KB SRAM, which I bought less than a year ago), I had severe booting issues; it only worked in the rarest of cases. Most of the time the system crashed right after inserting a boot disk. Occasionally I managed to load the system, but once running, crashes became so frequent that I began to think the MB02 was inherently unstable. However, since it worked flawlessly at Thomas's place during the meeting in Sindelfingen, I knew the problem had to be something in my setup at home.

First, I reinforced the Spectrum's edge connector contacts with solder—still nothing! Only when I connected the MB02 to a PC power supply did it finally start working without a hitch. As it turned out, the bulky power supply I had been using to feed the disk drives and the MB02 simply wasn't powerful enough for the MB02. Furthermore, a short circuit had crept into my wiring: I had used a shielded cable, and the heat from the soldering iron had melted the inner lead right through the insulation. On top of that, there was a detail I hadn't noticed earlier: the extension cord to the MB02 and the disk drives was made of wire that was far too thin. In the current setup, I simply extended a PC power cable using a thick 3 mm heavy-duty wire (rated for up to 220V / 800W) on the 5V rail (the MB02 needs 5V, as do the drives, so why bother extending the 12V rail?), soldered two plain wires instead of the shielded cable, and immediately everything worked flawlessly—even the +D to MB02 converter, which previously had been so erratic! The only things that still refuse to run properly are the HiSoft BASIC Compiler and Indian Patience, which triggers an NMI (in the meantime, I have managed to solve the problem with Indian Patience: the MB02 forces the 128 into USR mode).

This means I currently have the following disk systems up and running at home: 2× +D, 2× DISCiPLE, 2× MB02, and a Beta Disk built into the Pentagon.

And while we are on the subject: I also repaired one of the +2 motherboards that Thomas gave away at the meeting, and installed it into an old PC mini-tower case along with a Multiface 128, a DISCiPLE, and a keyboard interface. This time it looks even more professional than my previous attempt—one always learns along the way. As it turned out, the original power circuitry was failing to supply +12V and -5V. I disconnected those rails from the onboard voltage generator and fed them directly from the external PC power supply, and the board sprang back to life. Power to the rest of the Spectrum board was also tapped from the +12V line, decoupled beforehand with diodes. The video output is monochrome, but I mostly use a black-and-white TV anyway for programming and typing text—supposedly it causes less eye strain than color monitors, which is why I don't need glasses. Since there was no sound coming from the TV speaker and I didn't feel like prying open the TV modulator, I installed a headphone jack. Using an adapter cable, I can route the audio straight into a stereo system, giving me rich, full sound at all times.

Usage tip for combining DISCiPLE + Multiface 128:
Before pressing the Multiface button, the DISCiPLE should be made invisible using its Inhibit button. Furthermore, unless you are using a special DISCiPLE-compatible version of the MF128, you should also enter OUT 31,0. This completely masks the DISCiPLE. Releasing the Inhibit button reenables the system; without OUT 31,0, you can still access disks even while the Inhibit button is held down.

Once I get some spare cash again, I will probably buy two more tower cases to house a Spectrum 128 with an MB02, and the Pentagon.

That is all for today, see you soon!
Yours, L. C. D.

NEWS FROM THE WORLD OF SPECTRUM

• Prism PD has published a multi-page Public Domain catalogue featuring thousands of programs from Prism, Hammer, and FPD.
• Classix magazine will now be published three times a year. SCENE+, however, will not be joining this trend.
• Apparently, Peter Rennefeld isn't the only one trying to clone new +D interfaces: Prism PD is also attempting to reproduce this outstanding interface, albeit without the snapshot button. Curiously, Datel doesn't seem to have any objection. Did Peter simply offer them too little bribe money?