| |
-----Original Message-----
From: John Cooper [mailto:robots@jpsc.co.uk]
Sent: Tuesday, May 11, 1999 11:07 AM
To: Ralph Hempel
Subject: Re: New pbFORTH
> My FORTH is rusty too. I looked at the source for 15 minutes and did some fiddling
> before I saw the 2DROP at the beginning of #>....
>
> If you look at the definition of U. you will see that they put ZERO on the stack
> and then do the conversion...
>
> In other words the <# blah blah blah #> sequence expects a double length number
> on the stack!!! Referring to my trusted copy of Starting FORTH, this has always
> been the case.
How could I have forgotten... I've looked at that 'till I'm blue in the face.
> This brings up an interesting point, double numbers will have to be stored in
> a small-endian word fashion. In other words: LSW,MSW
>
> On a regular (Motorola) implementation, this would be (word big-endial) MSW,LSW
> and in the Intel world it would be LLSB,LMSB,LMSB,MMSB (byte small endian)
>
> At this point, I'm not too concerned about doubles, but with all the talk on the
> robotics group, fixed vs floating will be an issue...
Of course, then the copmplaints about precision start. You could always transmit
complex calculations to the PC I suppose.
> BTW, do you mind if I forward all our correspondence so far to the lugnet pbforth
> group so that others can see some activity?
Not at all, once it's up I'll continue the conversation there.
--
John Cooper, Wallington, UK
|
|
1 Message in This Thread:
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
