where can i find that random object that lets you set the range of the random numbers to chose from? i think i have heard it called random ab. does any one know what im talking about?

i-think-its-called-random-ab

http://www.timara.oberlin.edu/~gnelson/maxdownloads/downloadmax.htm

that is one of gary nelsons objects
http://www.timara.oberlin.edu/~gnelson/maxdownloads/downloadmax.htm


Yes, of course you can abstract this with the stock random, or use Gary's objects. But lp.tata uses a stronger random number algorithm *and* you can instantiate it as [lp.tata 10 20] or whatever you want.

And it's part of the Litter Starter Pack, so it's free.

I think the URI is in the .sig, but when I write directly to the forum I'm never sure, so: www.bek.no/~pcastine/Litter/

UB versions under beta test by Litter Pros.

And it's part of the Litter Starter Pack, so it's free. 

On Oct 11, 2006, at 2:46 PM, Peter Castine wrote:

> But lp.tata uses a stronger random number algorithm *and* you can 

> instantiate it as [lp.tata 10 20] or whatever you want.

any chance of getting an inlet to feed lp.tata a new seed value?

> But lp.tata uses a stronger random number algorithm *and* you can  
> instantiate it as [lp.tata 10 20] or whatever you want.
any chance of getting an inlet to feed lp.tata a new seed value?


There is a design point to be aware of: if you want a particular instance of, say, lp.tata to maintain its own seed, you must instantiate it with an explicit non-zero seed (for instance [lp.tata 1 127 -12345], in which case -12345 is the initial seed).

If you don't include a seed argument--ie, you just write [lp.tata 1 127]--then the object uses a global random number pool and the seed message re-juggles the pool for all instances of (in this case) lp.tata, except any that were initialized with explicit non-zero 3rd arguments.

This may not be entirely obvious, but it's quite a powerful set up.

The seed message is sort of buried in the old .PDF documentation. I hope it's a little more obviously accessible in the "Litter Messages" item in the See Also popup menu in the new Litter .help files.

There is a design point to be aware of: if you want a particular instance of, say, lp.tata to maintain its own seed, you must instantiate it with an explicit non-zero seed (for instance [lp.tata 1 127 -12345], in which case -12345 is the initial seed). 

If you don't include a seed argument--ie, you just write [lp.tata 1 127]--then the object uses a global random number pool and the seed message re-juggles the pool for all instances of (in this case) lp.tata, except any that were initialized with explicit non-zero 3rd arguments. 

This may not be entirely obvious, but it's quite a powerful set up. 

The seed message is sort of buried in the old .PDF documentation. I hope it's a little more obviously accessible in the "Litter Messages" item in the See Also popup menu in the new Litter .help files.


On Oct 11, 2006, at 3:40 PM, Peter Castine wrote:

thanks! :) when I opt-crtl'd on the object I saw the 'seed [int]' 

>
> Won't the seed message do the trick?
>
> All Litter RNGs grok seed.

thanks! :) when I opt-crtl'd on the object I saw the 'seed [int]'  
message...

> UB versions under beta test by Litter Pros.

How's that going ? Any chance us amateurs will see

> UB versions under beta test by Litter Pros.
>
> Best,
> Peter
> --
> --------------   

How's that going ? Any chance us amateurs will see
them soon ?
cheers
Roger


Nice demo patch, Kim. On the loadbang front, note that all Litter RNGs and noise objects seed themselves during class initialization based on time of day, uptime, machine address of the seed pool, and various other exotica. So the auto-seed is already pretty random.

There has been one issue on XP reported, which I suspect was related to a broken QTML install. Currently Litter Power on XP uses QTML for things like time-of-day and uptime. I'm moving all this stuff to native Windows calls as part of general maintenance, but it's only been medium priority (QTML seems to work fine for almost everyone, at least the to the extent that Litter uses it).

Finally, to Roger's question about availability: work on finishing up Litter Power 1.7 has been delayed due to death in the family requiring my presence in the US. I have limited Internet access here but software development is not possible at this time. There are few things more important to me than Litter Power, but this is one of them. Sorry. A more positive announcement will be made ASAP, but no sooner.

i think its called "random ab"