[seqfan] Re: Breaking news on partition numbers.

Joerg Arndt arndt at jjj.de
Tue Jan 25 22:01:15 CET 2011


* peter.luschny <peter.luschny at googlemail.com> [Jan 25. 2011 20:04]:
> jj> ..a nice "proof without words".
> Yes.
> 
> jj> I cannot, however, see how much more can be read from this picture
> jj> and certainly not something worth calling "model" of any sort,
> 
> Sure this is a model: an object made with paperboard, Indian ink
> and color nicely visualizing a mathematical idea. Today these
> 'models' are often made out of binary digits, are called
> 'demonstrations' and can be found on the Wolfram site.
> However, I also like these old artistic models.
> 

OK, I think I see what you mean.
However, some sort of formal definition may well
lead to more clarity and more infromation using less sequences.


> jj> Until this happens I'll remain the usual grumpy German,
> jj> also known as Sauer Kraut.
> 
> Germans also have an old recipe against being grumpy: They drink
> beer by the gallon. Hope that will help you also.

I am not actual grumpy  8-)
(cite was from a T-shirt of mine saying "A grumpy German is a Sauer-Kraut.").
Btw. Nuremberg == beer heaven!


> 
> Cheers, Peter
> 
> P.S. By the way, have you already done your C++ implementation
> of the new formula? We need support from the expert also on
> non-soft questions.

Aaargh, real work.
Isn't any good implementation in Pari/gp good for pretty huge n?

For the last bit one would have to use GMP (I'd do through CLN).
However, I am pretty sure there will be a few good implementations
(and reports os timings) in the near future (so I may just skip it).

> 
> _______________________________________________
> 
> Seqfan Mailing list - http://list.seqfan.eu/




More information about the SeqFan mailing list