Re: Repeatable ways to crash the MathKernel
- To: mathgroup at smc.vnet.net
- Subject: [mg116080] Re: Repeatable ways to crash the MathKernel
- From: "Sjoerd C. de Vries" <sjoerd.c.devries at gmail.com>
- Date: Tue, 1 Feb 2011 06:53:27 -0500 (EST)
- References: <ii5rka$fjh$1@smc.vnet.net>
I don't see any problemms with the first 3 sequences on my $Version
"8.0 for Microsoft Windows (64-bit) (November 7, 2010)"
However, I can confirm that the fourth example does crash my kernel.
I'd suggest you contact Wolfram support.
Cheers -- Sjoerd
On Jan 31, 9:24 am, Alexander <alexander_elk... at hotmail.com> wrote:
> The first two sequences are repeatable with
> $System=="Microsoft Windows (32-bit)" (sometimes the crash happens
> when Plot3D is evaluated as In[1] after the intial startup before any
> crash has happened):
>
> 1) The following sequence works on versions 5.2.0.0(has no
> exclusions),
> 6.0.3.0 & 7.0.1.0 but crashes version 8.0.0.0 of the kernel:
>
> In[1]:= Plot3D[Im[Log[((x+I*y)^2-Pi^2)^2]],{x,-2Pi,2Pi},{y,-2Pi,2Pi}]
> Out[1]=...(output not shown here)
>
> In[2]:= ?Plot3D
> ...(output not shown here)
>
> In[3]:= Plot3D[Im[Log[((x+I*y)^2-Pi^2)^2]],{x,-2Pi,2Pi},{y,-2Pi,2Pi}]
>
> After a few seconds the kernel crashes...
>
> 2) The following sequence works on versions 5.2.0.0(has no
> exclusions),
> 6.0.3.0 & 7.0.1.0 but crashes version 8.0.0.0 of the kernel:
>
> In[1]:= Plot3D[Piecewise[{{Im[Log[((x+I*y)^2-Pi^2)^2]],x^2-y^2!
> =Pi^2}}],
> {x,-2Pi,2Pi},{y,-2Pi,2Pi},Mesh->False]
> Out[1]=...(output not shown here)
>
> In[2]:= ?Plot3D
> ...(output not shown here)
>
> In[3]:= Plot3D[Piecewise[{{Im[Log[((x+I*y)^2-Pi^2)^2]],x^2-y^2!
> =Pi^2}}],
> {x,-2Pi,2Pi},{y,-2Pi,2Pi},Mesh->False]
>
> After a few seconds the kernel crashes...
>
> 3) The following line always crashes versions 5.2.0.0, 6.0.3.0,
> 7.0.1.0
> & 8.0.0.0 of the kernel:
>
> FullSimplify[Limit[-(1/2)Log[1/#]-Log[I#]+Log[#]/2&[Complex[y,y]],
> y->0]==-I*Pi/2]
>
> The following works (changed Complex[y,y] to y+I*y):
>
> FullSimplify[Limit[-(1/2)Log[1/#]-Log[I#]+Log[#]/2&[y+I*y],
> y->0]==-I*Pi/2]
>
> Anyone else see this happen?