From my notes on the string garbage collector bug in OSI ROM BASIC :

> There were two distinct bugs in the garbage collector.
>
> 1. Bad traversing of the arrays area due to wrong string
> step size. OSI Disk and other 6502 Basics use a step
> size of 3. Rom Basic also used 3 but actually needed 4.
>
> 2. Temporary strings were not collected.
>
> Bug 1 was quite serious and often led to crashes if string
> arrays were used. It only occurs in OSI/UK101 Rom Basic.
> (I’m unaware of any 4-byte 6502 MS Basic other than OSI).
>
> Bug 2 would result in corrupted strings but not crash the
> computer. The incidence of the bug hitting was fairly low.
> It occurs in early 6502 MS Basics (OSI/UK101 Rom Basic,
> OSI Disk Basic, PET 2001 etc).

The “step size of 3″ occurs in 5-byte 6502 MS-BASICs. That it
also appears in 4-byte OSI ROM BASIC and was the source of a bug
that occured exclusively in that machine is suggestive that
OSI ROM BASIC was derived from a 5-byte 6502 MS-BASIC.

1 ‘***********************
2 ‘ MICROSOFT! Easter Egg
3 ‘***********************
4 B=#E435 : E=B+8
5 FOR I=E TO B STEP -1
6 J=PEEK(I) : J=J-INT(J/64)*64
7 IF I>B THEN A$=A$+CHR$(J+64) ELSE A$=A$+CHR$(J)
8 NEXT I
9 PRINT A$

It was manufactured by GDR’s company “robotron” and featured a Z80A clone (U880D), a “GW BASIC compatible” BASIC interpreter in ROM and a CP/M 2.2 clone also in ROM (”SCPX5105″). Interestingly, the OS booted depended on the existance of an empty file on the first floppy drive named “SCPXA5105.SYS” – if it was there, CP/M loaded, otherwise RBASIC.

Robotron was known for its “creativity” when it came to “adapting” existing western hard- and software for their own socialist product line (CP/MSCP, dBASEIIREDABAS, SuperCalcKP, MS-DOSDCP etc pp).

Robotron as a hard- and software manufacturer didn’t survive the fall of the Iron curtain and what little is left of it is still very paranoid about discussing this aspect of their creative process ;-)

Well I guess I’ll see if I can find a ROM image somewhere or fire up one of the old machines even :-)

10 CLS
20 FOR I=&HBFEF TO &HBFE6 STEP -1
30 J=PEEK(I)
40 J=J-INT(J/64)*64
50 IF I>&HBFE6 THEN A$=A$+CHR$(J+64) ELSE A$=A$+CHR$(J)
60 NEXT I
70 PRINT A$

Looks like those “UNUSED GARBAGE BYTES” are indeed the

MICROSOFT!

Easter Egg

Rick Weiland and I (Bill Gates) wrote the 6502 BASIC.

I put the WAIT command in.

Mark Chamberlin and I wrote the 6800 BASIC.

When it says Paul Allen wrote the non-runtime stuff that means the development environment which was an amazing piece of work he did on the PDP10 that made development work very productive including simulation and symbolic debugging.

There were a lot of interesting versions of BASIC done for Japanese machines this article misses and then of course we did the IBM PC ROM BASIC (Neil Konzen and a few other people did most of the conversion work).

http://www.tenox.tc/tmp/ww/msxenix/pics/xnx2.png

I remember one really interesting “feature” I discovered on the TRS-80 MC-10 when poking/peeking somewhat randomly (at the time I had no documentation for machine besides the BASIC manual.)

Once the code was executed, the keyboard would lockup (requiring reset) but you could peek $2 which returned a value that seemed to vary with the distance and position my hand was relative to the chiclet keyboard. Years later, I’d often wondered if it was radio interference or something to do with the keyboard switches.

It was accurate enough to have been used as some kind of input device, if only I wasn’t just a young kid living on a farm way out in the boonies at the time :P

10 POKE 0, 0
20 PRINT PEEK(2)
30 GOTO 20

He might never have said “640k is enough for anybody”. But…

Great job, thanks!

FWIW, the Tandy Color Computer line actually had 3 versions of the BASIC in ROM: 8K Color BASIC, sold mostly with the 4K RAM minimum system which was integer only with no graphics extensions. Another 8K ROM which then gave you 16K Extended Color BASIC with all floating point, trigonometry, and graphics extensions. This just plugged into a socket on the motherboard. Last was Disk Extended Color BASIC which came in a ROM on the card which slid into the game slot on the machine on the same board with the disk controller to include the functions for floppy I/O. I believe that all 3 versions were obtained from Microsoft (according to the documentation which came with the computers.)

Another friend actually wrote GEM OS for the Atari ST and later moved to EA Electronics to do there Assembly Programming. I still have an Atari ST, and 800 in the garage. Memory lanes..

As I recall, you could switch between Integer and Applesoft BASIC on an Apple ][+ if you had a language card. The DOS 3.3 commands INT and FP would switch between them.

The Integer BASIC prompt was a greater-than sign as I recall.

TWO BIRTHS
“COLEEN” AND”CANDY”
The development of Atari Basic and Atari DOS

Microsoft BASIC for the Atari 6502 Computer was an optional piece of software that was rather expensive, so most user used the supplied Atari Basic. Later, an optimized Version of Atari Basic was produced by Frank Ostrowski (http://en.wikipedia.org/wiki/Frank_Ostrowski) who later did the GFA-BASIC for the Atari ST series.

This is the keyword list of Microsoft BASIC 1.0:
END, FOR, NEXT, DATA, INPUT, DIM, READ, LET, GOTO, RUN, IF, RESTORE, GOSUB, RETURN, REM, STOP, ON, NULL, WAIT, LOAD, SAVE, DEF, POKE, PRINT, CONT, LIST, CLEAR, NEW, TAB(, TO, FN, SPC(, THEN, NOT, STEP, AND, OR, SGN, INT, ABS, USR, FRE, POS, SQR, RND, LOG, EXP, COS, SIN, TAN, ATN, PEEK, LEN, STR$, VAL, ASC, CHR$, LEFT$, RIGHT$, MID$
BASIC 1.1 only added a single instruction keyword: “GET”

According to your list, Intellivision BASIC doesn’t have GET. It removed NULL, WAIT, POKE and PEEK and added the instructions PLOD, PSAV, VLOD, VSAV, SLOD, PRT and the function GETC. It looks like it’s 1.0-based, although it was released in 1982, and the copyright says 1980.

I put the decoded list here: http://spatula-city.org/~im14u2c/intv/intellivision_basic_keyword_list.txt

The 2-letter error code table appears with the keyword list.

And of course, I would be happy to look at the ROM you have!

@tim lindner: Thanks a lot for the hint, I should have read the full foreword! Fixed.

I didn’t find any of the 40-bit constants mentioned above, so I presume that the Intellivision version that Mattel picked up was the version with 32-bit floating point.

The Intellivision BASIC reports itself as:

INTELLIVISION BASIC
Copyright Microsoft, Mattel 1980

I wonder where it fits in the time line you have above? FWIW, it doesn’t seem to have a WAIT keyword, and it uses 2-letter error codes. (It does have VSAV/VLOD, which seem to be for audio playback on the tape… This *is* a unique machine.) The errors it reports are: GO, NF, SN, RG, OD, FC, OV, OM, US, BS, DD, /0, ID, TM, LS, ST, CN, UF, TR, TP

As for easter eggs, I don’t know if Intellivision BASIC has any. I know the Keyboard’s System ROM does have at least one. There are a number of compressed strings that can be expanded with special escapes. (These were character numbers > 255 if arriving via 10-bit wide RAM shared with the Master Component, or a three byte sequence if generated directly on the Keyboard Component from the 6502. A very odd machine.)

These strings compress down many common phrases and phrase fragments. Two of the strings in this table are “K. Smith” and “L. Zwick”, two of the main programmers for this ROM. So, with an appropriate “PRINT” statement you could get the programmer’s names to appear. In fact, I believe it was possible to get the names to appear in the “typewriter” mode as well.

On a different note, the Mattel Aquarius (a Z80 based machine) also has Microsoft BASIC built in. I don’t have a dump of its ROM though.

@Faried Nawaz: “AppleSoft” is the BASIC based on Microsoft BASIC that was offered for the Apple II. “Apple I BASIC” is written by Woz and unrelated.