[vorbis] bark_noise

[Richard Polton]

I tried to reduce the size of the quasi-infinite value to
10^34 and while this works temporarily, the code still collapses
later on.

Michael Smith wrote:
> 
> >> >At this point, noise[1]==3.4e38. We are trying to add another 3.4e38
> >> >to it and this is generating a SIGFPE on my machine.
> >
> >I'll look at this tomorrow.  There are a number of underflows I need
> >to catch for beta 3 too.
> >
> >(Hmm... values that negative shouldn't be popping up in bark_noise...)
> 
> Values that negative come out (again...) as a result of todB() on an argument of 0. In this particular case, it does todB(f[i]*f[i])+400.
> 
> The +400 has essentially no effect because of the magnitude of the todB() return. Apart from that, it's being called basically directly on the transformed (MDCT) data - there's no reason why it would be bypassed here, so zeros are to be expected.
> 
> In the bughunting earlier today I found another case where -9e38 is too negative and can cause FPEs - so it's probably best fixed just by increasing that by a few orders of magnitude.
> 
> Michael
> 
> --- >8 ----
> List archives:  http://www.xiph.org/archives/
> Ogg project homepage: http://www.xiph.org/ogg/
> To unsubscribe from this list, send a message to 'vorbis-request at xiph.org'
> containing only the word 'unsubscribe' in the body.  No subject is needed.
> Unsubscribe messages sent to the list will be ignored/filtered.

--- >8 ----
List archives:  http://www.xiph.org/archives/
Ogg project homepage: http://www.xiph.org/ogg/
To unsubscribe from this list, send a message to 'vorbis-request at xiph.org'
containing only the word 'unsubscribe' in the body.  No subject is needed.
Unsubscribe messages sent to the list will be ignored/filtered.