[maxmsp] Re: Random between +/- 128.0??

Quote: Chris Muir wrote on Tue, 14 August 2007 23:44
----------------------------------------------------
> At 6:23 PM -0600 8/14/07, UCDMSRA wrote:
> >Your demo patch is very nice. I never thought that I can use table to build probabilities. I am sorry if this question is really silly, I think I understand what you said and what your patch doing. I also know how to use table.......but I don't understand how to set probabilities of table. In you example, there are several 64 and 127, what does that means? I mean, is it the probabilities 64 and 127 will come out if the result?
> 
> When you bang a table, it treats the data in it roughly as probabilities for that address. For example if the table had data only at address 1, when you bang it, it would only output 1. The, probably simplistic, way I think of this is that the value at an address is the probability that that address will be output (as data). It sort of turns the table on its head, the address is the output data, and the data at an address is the probability of that address being output.
> 
> The significance of the 64 and 127 is that they are at a 2:1 ratio (although I should have used 64 instead of 64). It might as well have been 1 and 2 instead of 64 and 127.
> 
> The manual has more to say on this.
> 
> 
> >I tried to figure out but I can not connect the way of building probability table. For example, if I hope the number 1,3,5,7,13 have 25% chance, what's the table should look like? How to build it?
> 
> A 25% chance compared to what?
> 
> 
> >At the end, which object I can use for reading the random result from result table so that I can send the result numbers to other patches?
> 
> I'm not sure what you mean by this. Could you clarify?
> 
> -C
> 
> -- 
> Chris Muir           | "There are many futures and only one status quo.
> cbm at well.com         |  This is why conservatives mostly agree,
> http://www.xfade.com |  and radicals always argue." - Brian Eno
> 
----------------------------------------------------

Hi,

Yeah, I think I got it. For the building probabilities of the table, I understand what you means and I know how to build the table now. Thank you so much.

Sorry for my English makeing you confused. Although I understand how the probability table working, I still can not make this to other case. I hope I can build a patch run a 0~100 random numbers, but there are 5 numbers' (for example: 1,3,5,7,11) probabilities are 20 times more then others. So, if in the original random case each number has 1% probability, I would like to let those 5 numbers have 20% probabilities compared to other numbers. 

For what I understand right now, I probably need to build a 100x100 table which have y = 1,3,5,7,11 and x=20 in each y position. Please correct me if I was wrong.

At the last part, I still can not figure out how to build the whole patch. If I use the same way you show me (probability table and result table), I hope I can use something which be able to read results from the results table and express properly. I hope I can use something to read the result table and unpack it so that I will get the whole results numbers once at a time. Is it possible?

I hope you can understand what I try to say and I am very appreciate your patience and kindly help.

thank you so much