Re: [Chicken-users] Why are 32 bit inexact?

[Christian Kellermann]

* Christian Kellermann <address@hidden> [141228 12:47]:
> * Sascha Ziemann <address@hidden> [141228 12:33]:
> > Hi,
> > 
> > I tried to play a bit with ARM assembler, but reading 4 byte from a port
> > seems to be quite hard with Chicken.
> > 
> > Why are 32 bits inexact?
> 
> Because in CHICKEN we use 3 bits of a machine word for type tagging.
> So you cannot use all bits as a number and the number fill overflow
> to a float representation.

I hit send too soon, so if you need the number read in a 32bit
srfi-4 vector or use the bignum tower, i.e. (use numbers).

HTH,

Christian

-- 
May you be peaceful, may you live in safety, may you be free from
suffering, and may you live with ease.