Re: [avr-gcc-list] storing a float value to a 8-bit registers

[Abhijeet D Mhatre]

> Delatils of float/double representation should not be interesting for C
> users. In fact we are using C language to "forget" such details. But it is
> very important to know percision of the given float/double representation

>clipping and
> scaling like that:
> double decibels;
> int16_t idecibels;
> if ( decibles < 0.0 )
>       idecibels = (decibels-0.005)*100.0;
> else
>       idecibels = (decibels+0.005)*100.0;

Thanks Artur
I didnt imagine the solution to be so easy.  I was thinking I would have to 
dig into the way the nos are stored.

One more thing. 
Wouldn't it be safer to cast the float to an int.
idecibels =(int) (decibels-0.005)*100.0;

-- 
Regards
Abhijeet D Mhatre

Email: address@hidden