Re: [RFC/PATCH] sbc: new filtering function for 8 band fixed point encoding

[Siarhei Siamashka <siarhei.siamashka@nokia.com>]

On Friday 12 December 2008 21:19:20 ext Brad Midgley wrote:
> Guys
>
> One mistake we made was not keeping track of what functions we were
> applying to the published tables as we worked things to look less and
> less like the pseudocode in the spec. So if you start again from one
> of the tables in the spec, keep a comment like
>
> /* ourtable(x) = (int32)(table1(x) << 16) */

This part can be probably done just by having a macro, something like
#define F_TO_Q15(x) (int16_t)((x>0) ? ((x)*(1<<15)+0.5) : ((x)*(1<<15)-0.5))
And then using plain floating point numbers from the SBC specification in the
table, wrapped into this macro. Though I wonder if it is possible to use such
conditional expression in the static table initializer list with all versions
of gcc/other compilers.

> Which in this case would mean that in ourtable we've shifted the
> original table1 float value left 16 bits and truncated it to an int32.
>
> I also combined tables or split tables to simplify our loop logic or
> eliminate operations; an explanation in a comment would have been
> appropriate.

Yes, any transformations or simplifications should be extensively commented.
So that it will be always possible to reproduce them or verify their
correctness.

Can anybody try to remember/explain what transformations were applied to
the existing fixed point implementation?

-- 
Best regards,
Siarhei Siamashka