On Thu, Jul 12, 2001 at 01:08:14AM +0300, n++k wrote:
> [Greg Berchin <[EMAIL PROTECTED]>]
> | Steve Harris wrote:
> | 
> | >I don't understand the hardware issues, but as the filter coefficent 
> | >appraoches 0.0f the number of cycles taken to multiply it goes up 
> | 
> | This confuses me.  In reading the documentation for the FPU on the
> | Athlon,
> | they state that it can perform two pipelined double precision floating
> 
> http://www.smartelectronix.com/musicdsp/text/other001.txt

Thanks, but the example code I just gave allready has denormal protection
in it.

It is for the large numbers case.

- Steve

Reply via email to