Re: prerelease checks

[Steve ROBBINS <stever at bic dot mni dot mcgill dot ca>]

On Sun, 14 May 2000, Brian Gough wrote:

> Interesting. This is what I get,
> 
>  PASS: miser(f1), dim=7, err=0.0266 (0.963855 observed vs 1 expected)
>  PASS: miser(f1), dim=9, err=0.0693 (0.9571 observed vs 1 expected)
> 
> The first case is pretty similar. The second case seems to be
> significantly different.  I have no idea why it should be.
> 
> A long shot: if the environment variables GSL_RNG_TYPE or GSL_RNG_SEED
> were set when you ran the tests then that would affect the
> results. (We should definitely modify the monte functions not to
> depend on the environment variables and to take a generator as an
> extra argument).

Sorry, I should have described my environment more clearly.

NO GSL_ variables are set.  
I did "./configure; make; make check"; i.e. no special configure options.
Compiled with gcc 2.95.2 (the debian package is version 2.95.2-10).
The CPU is PII 400MHz; cat /proc/cpuinfo reads:

processor       : 0
vendor_id       : GenuineIntel
cpu family      : 6
model           : 5
model name      : Pentium II (Deschutes)
stepping        : 2
cpu MHz         : 400.920218
cache size      : 512 KB
fdiv_bug        : no
hlt_bug         : no
sep_bug         : no
f00f_bug        : no
coma_bug        : no
fpu             : yes
fpu_exception   : yes
cpuid level     : 2
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 sep mtrr pge mca cmov pat pse36 mmx osfxsr
bogomips        : 399.77


I'm willing to run extra test cases, but I don't really have the expertise
to delve into the code itself at this time.

 
> Steve ROBBINS writes:
>  > However, 'make check' reported these failures:
>  > 
>  > FAIL: miser(f1), dim=7, err=0.0209 (0.951217562909207448 observed vs 1 expected)
>  > FAIL: miser(f1), dim=9, err=0.1950 (1.20951997339804262 observed vs 1 expected)
>  > FAIL: miser_test
>  > 
>