In what version of BASIC was OPT 8-15 introduced?
In what version of BBC BASIC was bit 3 of OPT introduced, to enable checking of the code pointer against a limit in L%?
0
Comments
-
I believe it was introduced in RISC OS 2 with ARM BBC BASIC version 1.04 from 1988, as I found a reference to it in an old scan of the RISC OS 2 reference manual here https://www.4corn.co.uk/archive/docs/RISC OS Programmer's Reference Manual - Volume 4-opt.pdf
... but not in the BBC Master reference manual https://www.stardot.org.uk/forums/viewtopic.php?t=20466 or in the Arthur manual https://www.4corn.co.uk/archive/docs/Archimedes Programmer's Reference Manual - Volume 2-opt.pdf
Edit: Updated with BASIC version in RISC OS 2.1 -
Thank you.I found a reference to it in an old scan of the RISC OS 2 reference manual
That old manual particularly interests me because it includes a description of the Floating point emulator, including support for the IEEE 80-bit extended precision format - although padded to 96 bits by inserting 16 zero bits between the sign and the exponent (it does say "the above layout will vary on future floating point systems").
If only that format had been retained when ARM CPUs eventually gained integrated FPUs, it would have made ARM directly competitive with x86 CPUs for high-precision mathematical applications (and would have provided much better compatibility between the x86 and ARM versions of my BBC BASICs!).
But for some reason I've never understood, when the FPU was integrated on the CPU chip rather than as a separate co-processor, hardware support for IEEE extended precision numbers was dropped. Madness.
1 -
For reference, all the old ARM-era manuals are here https://www.4corn.co.uk/articles/docs/0