I have played with this a bit and have no luck yet. A lot of computer time has convinced me it is true but...
]]>I have a feeling that a disproof for p=k could be tweaked to become a proof for p=1/k.
]]>I know that for p>1 it does not work.
There are solutions for p = 2, the pythagorean triples. What restrictions are on a,b,c?
An example where it does not work for this p?
I think we could find a lot of them. I cannot find one in that range yet.
]]>There are some compelling reasons why I do not think you will find any.
]]>Thanks
]]>