Re: Question about DiracDelta

*To*: mathgroup at smc.vnet.net*Subject*: [mg68645] Re: Question about DiracDelta*From*: Bill Rowe <readnewsciv at earthlink.net>*Date*: Mon, 14 Aug 2006 06:44:24 -0400 (EDT)*Sender*: owner-wri-mathgroup at wolfram.com

On 8/13/06 at 5:52 AM, michev at gmail.com (vasil michev) wrote: >Plot[NIntegrate[DiracDelta[3 - Cos[x] - Cos[y] - Cos[z] - e], {x, >-Pi, Pi}, {y, -Pi, Pi}, {z, -Pi, Pi}, MaxPoints -> 1000000]/8/Pi^3, >{e, 0, 6}, PlotPoints -> 30] >the problem is that the function always returns 0, no matter what >the value of 'e' is. I saw that other people have reported incorect >evaluation/simplification related to DiracDelta, but I was wondering >can anyone tell me if there's some workaround? I don't think the issue is a problem with DiracDelta specifically. You are asking Mathematica to do a numerical integral of a function that will be zero everywhere except a few specific points. So, unless NIntegrate causes the function to be sampled at one of those specific points, it will evaluate to zero. I don't think increasing MaxPoints the way you have done forces NIntegrate to sample the function at more points. Instead, I believe this merely allows NIntegrate to sample at more points. -- To reply via email subtract one hundred and four