9

archive.org's DG collection has Single User BASIC and Extended BASIC manuals, but no BB manual. Does anyone have a PDF?

The underlying question... the earliest mention of BB I can find is from a 1977 ComputerWorld article announcing its release. This includes some rather confusing language about a "double-precision integer arithmetic facility". "double precision" and "integer" are not normally seen in the same statement.

It also states that it was based on the same "research" as Extended BASIC. Extended BASIC includes a complete FP system, and FP libraries and/or hardware are a standard part of DG systems of the era.

I strongly suspect they are actually referring to a BCD or fixed-point system. I find it difficult to believe that any business-oriented language would not have some way to handle pennies, especially considering they include output formatting for dollar formats.

Can anyone shed some light on this?

5
  • It is very confusing. If "double precision integer" meant what is more conventionally called "long int" (and "precision" meant what we call "range"!), you'd expect 32 bits giving a range of +/- 2 billion, not a lousy 21 million. If the BASIC does all its arithmetic in floating-point, you'd expect DG double-precision to be able to handle 56-bit integers exactly, unless there's some subtlety about base-16 floating point I don't understand. Dec 27 '21 at 22:33
  • 3
    @another-dave But with a .3 decimal fixed point format the range of a signed 32-bit integer would indeed be around 21 million.
    – WimC
    Dec 28 '21 at 9:01
  • DG BB does not have any floating point. Everything is in integers: single and double precision. You basically scale up all the calcs. Eg for finance, do everything in cents/pence etc and then convert to dollars/pounds before display.
    – cup
    Dec 28 '21 at 17:12
  • @cup - then what does "double precision" refer to? 32-bit integers? Then why is the range so small? Dec 29 '21 at 21:55
  • DG Novas and Eclipses were 16-bit machines in the 70s. Double precision refers to 32-bit integers. In those days, 32 bit integers were considered big.
    – cup
    Dec 29 '21 at 22:34

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.