Re: [Protux-devel] Peak building/drawing question

[Luciano Giordana]

On Tuesday 01 July 2003 12:11 pm, Remon Sijrier wrote:
> Hi all,
>
> I made (as you know) some optimizations to the peak build process.
> But I have a few questions:
>
> 1. In macro view, only the upper side of the waveform is painted. I
> inverted the negative part of the waveform, so it is painted at the upper
> side of the x-axis. But this is NOT the actual waveform. If it is not
> converted, but set to zero, it looks like if there is no signal, which is
> of course not true. The question is: should the negative values be painted
> at the upper side of the x-axis, or set to zero?

We use the Samplitude standard, we show the representative peaks in module (abs)
so all negatives must be treated as positives and displayed as it


>
> 2. The other question isn't a quesion actually :).
> The peak files are much to large (about 1/8 of the actual audioFileSize). I
> changed the primitive type, so the peak buffers are now made with "char"
> instead of "short", reducing the peakfile and buffersize with a factor 2.
> If there are no objections to this change, I will upload it within a few
> hours. Peakfiles made with an older version of the peak build proces do not
> work then anymore. Simply delete them, so they are regenerated. (Well, that
> should be the case).

short is 16 bits (2 chars) type. It should not make ANY difference !!! 
using shorts is a little faster , though. if I use chars, I have to calculate 
the sample, while using shorts, I dont have to
r u sure you are keeping the consistency of peaks ?

>
> Thanks,
>
> Remon
>
>
> _______________________________________________
> Protux-devel mailing list
> address@hidden
> http://mail.nongnu.org/mailman/listinfo/protux-devel

-- 
Best Regards
--
Luciano Giordana - Musician - Certified Java/GNU C++ Developer - Free Software 
Evangelist
Project Mustux - http://www.freesoftware.fsf.org/mustux
-- GNU/Linux adoption grew 65% only this yer. Smile : we're winning. --