Page 1 of 1
Doubles?
Posted: 21 May 2008, 18:55 PM
by GTBecker
A question just posed on the Yahoo Basic-X group asks if 64-bit math can be done; the answer there will be no - but it's a good question to ask here.
Might a complete set of 64-bit math functions be done in native mode?
Posted: 21 May 2008, 19:32 PM
by mikep
This is an item I have asked Don about before; more to round out the language than anything else. I think that the main reason that only single precision arithmetic is supported is because there really aren't that many requirements to use double precision arithmetic (DP) on a 8-bit micro. If you really want DP then you probably also want a much faster machine like a 32-bit ARM microprocessor. After all, you are now doing some serious number crunching.
Certainly with native support you can mix C code that does DP arithmetic with ZBasic code in the same application. You would need to also handle the conversions in and out of DP from other data types as well.
Re: Doubles?
Posted: 21 May 2008, 23:57 PM
by dkinzer
GTBecker wrote:Might a complete set of 64-bit math functions be done in native mode?
It is certainly feasible. There has been some discussion on the AVR Freaks forum about this. I suspect that double precision will eventually be added to the avr-gcc compiler.