Can you use MIDI notes and convert them to frequency. This way you can
stay in a linear scale, but have the appropriate exponential output.
The midicps udo here should work ok I think?

http://www.csounds.com/udo/displayOpcode.php?opcode_id=64


On 8 February 2013 21:23, Alex Weiss  wrote:
> Hi list
>
> For an instrument of mine, I need to pick random frequencies between 20Hz
> and 20kHz. Because we perceive pitch logarithmically, a uniform distribution
> won't work as it will be biased towards higher pitches. My limited
> mathematical knowledge leads me to believe that an exponential distribution
> should fit the bill (correct me if I'm wrong). But this is where my
> mathematical knowledge ends; should i adjust the klambda parameter in
> exprand empirically, or is there a mathematically sound value that would
> make sense in this context?
>
> Thanks,
> Alex

Another possibility might be to use the uniform distribution and then
map it to a logarithmic curve (could try using the logcurve opcode).

On Fri, Feb 8, 2013 at 9:40 PM, Rory Walsh  wrote:
> Can you use MIDI notes and convert them to frequency. This way you can
> stay in a linear scale, but have the appropriate exponential output.
> The midicps udo here should work ok I think?
>
> http://www.csounds.com/udo/displayOpcode.php?opcode_id=64
>
>
> On 8 February 2013 21:23, Alex Weiss  wrote:
>> Hi list
>>
>> For an instrument of mine, I need to pick random frequencies between 20Hz
>> and 20kHz. Because we perceive pitch logarithmically, a uniform distribution
>> won't work as it will be biased towards higher pitches. My limited
>> mathematical knowledge leads me to believe that an exponential distribution
>> should fit the bill (correct me if I'm wrong). But this is where my
>> mathematical knowledge ends; should i adjust the klambda parameter in
>> exprand empirically, or is there a mathematically sound value that would
>> make sense in this context?
>>
>> Thanks,
>> Alex
>
>
> Send bugs reports to the Sourceforge bug tracker
>             https://sourceforge.net/tracker/?group_id=81968&atid=564599
> Discussions of bugs and features can be posted here
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"
>

thanks - interesting alternative for some situations!
	j


Am 12.02.2013 07:31, schrieb Alex Weiss:
> Hi list
>
> As an aside for anybody that's interested: Another method that seems to
> work fairly well is to generate the numbers with a uniform distribution
> in the Mel scale, then convert from Mel to frequency. Never used Mels
> before and was pleasantly surprised: http://en.wikipedia.org/wiki/Mel_scale
>
> Alex
>
> On Sun, Feb 10, 2013 at 1:56 PM, Alex Weiss  > wrote:
>
>     Thanks everyone. I ended up using a combination of unirand and
>     expcurve, which works fairly well.
>
>     Alex
>
>
>     On Fri, Feb 8, 2013 at 1:53 PM, Steven Yi      > wrote:
>
>         Another possibility might be to use the uniform distribution and
>         then
>         map it to a logarithmic curve (could try using the logcurve opcode).
>
>         On Fri, Feb 8, 2013 at 9:40 PM, Rory Walsh          > wrote:
>          > Can you use MIDI notes and convert them to frequency. This
>         way you can
>          > stay in a linear scale, but have the appropriate exponential
>         output.
>          > The midicps udo here should work ok I think?
>          >
>          > http://www.csounds.com/udo/displayOpcode.php?opcode_id=64
>          >
>          >
>          > On 8 February 2013 21:23, Alex Weiss          > wrote:
>          >> Hi list
>          >>
>          >> For an instrument of mine, I need to pick random frequencies
>         between 20Hz
>          >> and 20kHz. Because we perceive pitch logarithmically, a
>         uniform distribution
>          >> won't work as it will be biased towards higher pitches. My
>         limited
>          >> mathematical knowledge leads me to believe that an
>         exponential distribution
>          >> should fit the bill (correct me if I'm wrong). But this is
>         where my
>          >> mathematical knowledge ends; should i adjust the klambda
>         parameter in
>          >> exprand empirically, or is there a mathematically sound
>         value that would
>          >> make sense in this context?
>          >>
>          >> Thanks,
>          >> Alex
>          >
>          >
>          > Send bugs reports to the Sourceforge bug tracker
>          > https://sourceforge.net/tracker/?group_id=81968&atid=564599
>          > Discussions of bugs and features can be posted here
>          > To unsubscribe, send email sympa@lists.bath.ac.uk
>          with body "unsubscribe csound"
>          >
>
>
>         Send bugs reports to the Sourceforge bug tracker
>         https://sourceforge.net/tracker/?group_id=81968&atid=564599
>         Discussions of bugs and features can be posted here
>         To unsubscribe, send email sympa@lists.bath.ac.uk
>          with body "unsubscribe csound"
>
>
>