On 29-Sep-07, at 11:02 AM, Daniel Stenning wrote:
>
>> Perhaps the fact that Rb doesn't really like 4-byte floats has
>> something to do with this, but that's just another excuse.
>
> Whats the issue with 4 byte floats ? This is news to me - ( but
> then a lot
> of such things are.. )
I don't know all the technical details, but from what I can gather
floating point literals, constants, and intermediate expression
values are all double precision, so if you're working with Singles in
RB you get a lot of unavoidable double-to-single-to-double conversions.
I meant to reply to this sooner but my outbox went all screwy...
Frank.
<http://developer.chaoticbox.com/>
_______________________________________________
Unsubscribe or switch delivery mode:
<http://www.realsoftware.com/support/listmanager/>
Search the archives:
<http://support.realsoftware.com/listarchives/lists.html>
|