Does anyone know how to “initialize” Model I Disk BASIC so that its ROM calls (ie, floating point math) can be used from a TRSDOS machine code application? I believe that at least one program did this “back in the day”—Phelps Gates amazing APL/80.
- VCF South West - June 14 - 16, Davidson-Gundy Alumni Center at University of Texas at Dallas
- VCF West - Aug 2 - 3, Computer History Museum, Mountain View, CA
- VCF Midwest - Sept 7 - 8 2024, Schaumburg, IL
- VCF SoCal - Mid February 2025, Location TBD, Southern CA
- VCF East - April 2025, Infoage Museum, Wall NJ
-
Please review our updated Terms and Rules here
Accessing BASIC ROM Calls From Assembly
- Thread starter VinnieMan
- Start date
It depends on how much of BASIC you want to use.
If all you want is the math functions, and you use the two floating point accumulator areas 411Dh and 4127h properly (so no error that would trigger BASIC's error handling), then only one small area needs to be set up. And even that is only for floating point division or exponential. You just need to copy 14 bytes from ROM starting at 18F7h to RAM starting at 4080h:
LD HL,18F7H
LD DE,4080H
LD BC,000EH
LDIR
But if you want to use more, like BASIC's convenient floating point ASCII conversion routines, or many printing routines, then you'll probably have to plug the Disk BASIC RAM "exits" from 41A6h to 41E2h with C9 "ret" instructions the way Level 2 does. And if you want a functional disk, you have to be very careful with RST 28h, which is shared with both DOS function calls and Break key handling.
I recommend reading The Alternate Source's "ROM Routines Documented" book, which explains how to use much of the ROM properly. Plugging Disk exits is covered on page 4, and the whole Chapter 2 is on using the math routines, with my example LDIR above coming from page 30. It even explains string functions that other books skip. You can find it on archive.org at
(Note the "live" pages of that link are scanned so horribly as to be unreadable. But under Download Options, either PDF version is perfectly readable, so use those to read the book.)
If all you want is the math functions, and you use the two floating point accumulator areas 411Dh and 4127h properly (so no error that would trigger BASIC's error handling), then only one small area needs to be set up. And even that is only for floating point division or exponential. You just need to copy 14 bytes from ROM starting at 18F7h to RAM starting at 4080h:
LD HL,18F7H
LD DE,4080H
LD BC,000EH
LDIR
But if you want to use more, like BASIC's convenient floating point ASCII conversion routines, or many printing routines, then you'll probably have to plug the Disk BASIC RAM "exits" from 41A6h to 41E2h with C9 "ret" instructions the way Level 2 does. And if you want a functional disk, you have to be very careful with RST 28h, which is shared with both DOS function calls and Break key handling.
I recommend reading The Alternate Source's "ROM Routines Documented" book, which explains how to use much of the ROM properly. Plugging Disk exits is covered on page 4, and the whole Chapter 2 is on using the math routines, with my example LDIR above coming from page 30. It even explains string functions that other books skip. You can find it on archive.org at
Tandy/Radio Shack Book: ROM Routines Documented (1983)(The Alternate Source) : Free Download, Borrow, and Streaming : Internet Archive
Tandy/Radio Shack Book: ROM Routines Documented (1983)(The Alternate Source)
archive.org
(Note the "live" pages of that link are scanned so horribly as to be unreadable. But under Download Options, either PDF version is perfectly readable, so use those to read the book.)
Bruce Tomlin
Experienced Member
You might get more insight to it by looking at some later genuine Microsoft source code. There shouldn't have been too many changes to the math code between the TRS-80 version and the CP/M 5.2 version.
Look for F4.MAC, and you can probably find a lot of common code.
Anyhow, as long as you know where the floating point accumulators are, you should be able to use the floating point math functions without any special initialization.
WinWorld: Microsoft Basic-80 5.2 Source
WinWorld is an online museum dedicated to providing free and open access to one of the largest archives of abandonware software and information on the web.
winworldpc.com
Look for F4.MAC, and you can probably find a lot of common code.
Anyhow, as long as you know where the floating point accumulators are, you should be able to use the floating point math functions without any special initialization.