octave-maintainers
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: octave 4.0.3 - win 10


From: Tatsuro MATSUOKA
Subject: Re: octave 4.0.3 - win 10
Date: Mon, 4 Jul 2016 09:48:37 +0900 (JST)

----- Original Message -----

> From: Tatsuro MATSUOKA 
> To: PhilipNienhuis ; "octave-maintainers
> Cc: 
> Date: 2016/7/4, Mon 08:54
> Subject: Re: octave 4.0.3 - win 10
> 
> ----- Original Message -----
> 
>>  From: PhilipNienhuis
>>  To: octave-maintainers
>>  Cc: 
>>  Date: 2016/7/3, Sun 19:38
>>  Subject: Re: octave 4.0.3 - win 10
>> 
>> t macchant wrote
>>>>   From: JohnD 
>>> 
>>>>   To: octave-maintainers
>>>>   Cc: 
>>>>   Date: 2016/7/3, Sun 10:14
>>>>   Subject: octave 4.0.3 - win 10
>>>> 
>>>>   On my 64 bit machine using the official installer:
>>>> 
>>>>     PASS     13022
>>>>     FAIL         0
>>>>     XFAIL       27
>>>>     SKIPPED     54
>>> 
>>> 
>>>   I have gotten slightly different results (win 10 64bit) but 
>>>   I do not think it is serious.
>>> 
>>> 
>>> 
>>>     PASS     13023
>>>     FAIL         0
>>>     XFAIL       26
>>>     SKIPPED     54
>> 
>>  Hmmm, with an --enable-windows-64 cross-build I get 3 FAILs on Windows 10:
>>  - axis.m
>>  - assert.m
>>  (both the "usual" tolerance errors for these functions)
>>  and
>>  - ichol.m:
>>>>>>>   processing
>>>>>>> 
>> 
> C:\Programs\Octave\Octave-4.0.3_w64\share\octave\4.0.3\m\sparse\ichol.m
>>  ***** error <pivot equal to 0> ichol (sparse (-0))
>>  !!!!! error failed.
>>  Expected <pivot equal to 0>, but got no error
>> 
>> 
>>  Philip
> 
> 
> With --enable-windows-64 cross-build I get 2 FAILs on Windows 10
> 
> 
>   PASS     13021
>   FAIL         2
>   XFAIL       26
>   SKIPPED     54
> 
> For axis.m and assert.m, perhaps the same tolerance errors occured as those 
> by 
> Philip.
>  
> 
>>>>>>  processing 
> C:\octave\Octave-4.0.3-x64\share\octave\4.0.3\m\plot\appearance\axis.m
> 
> ***** test
>  hf = figure ("visible", "off");
>  unwind_protect
>    a = logspace (-5, 1, 10);
>    loglog (a, -a);
>    axis tight;
>    assert (axis (), [1e-5, 10, -10, -1e-5]);
>  unwind_protect_cleanup
>    close (hf);
>  end_unwind_protect
> !!!!! test failed
> ASSERT errors for:  assert (axis (),[1e-5, 10, -10, -1e-5])
> 
>   Location  |  Observed  |  Expected  |  Reason
>     (1)         1e-005       1e-005      Abs err 5.0822e-021 exceeds tol 0
>     (4)        -1e-005      -1e-005      Abs err 5.0822e-021 exceeds tol 0
> 
>>>>>>  processing 
> C:\octave\Octave-4.0.3-x64\share\octave\4.0.3\m\testfun\assert.m
> ***** test
>  x = [-40:0];
>  y1 = (10.^x).*(10.^x);
>  y2 = 10.^(2*x);
>  assert (y1, y2, eps (y1));
>  fail ("assert (y1, y2 + eps*1e-70, eps (y1))");
> !!!!! test failed
> ASSERT errors for:  assert (y1,y2,eps (y1))
> 
>   Location  |  Observed  |  Expected  |  Reason
>     (11)        1e-060       1e-060      Abs err 4.1454e-076 exceeds tol 
> 1.3818e-076
>     (13)        1e-056       1e-056      Abs err 3.3959e-072 exceeds tol 
> 1.132e-072
>     (14)        1e-054       1e-054      Abs err 4.3467e-070 exceeds tol 
> 1.4489e-070
>     (19)        1e-044       1e-044      Abs err 3.7338e-060 exceeds tol 
> 1.2446e-060
>     (22)        1e-038       1e-038      Abs err 2.6101e-054 exceeds tol 
> 1.3051e-054
>     (23)        1e-036       1e-036      Abs err 3.341e-052 exceeds tol 
> 1.6705e-052
>     (27)        1e-028       1e-028      Abs err 2.2421e-044 exceeds tol 
> 1.121e-044
>     (29)        1e-024       1e-024      Abs err 3.6734e-040 exceeds tol 
> 1.8367e-040
> 
> I do not see FAIL on ichol.m.
> 
> Tatsuro


Also I executed a test with --enable-windows-64 --enable-64.
The results are the same as those with --enable-windows-64 as expected.

Tatsuro



reply via email to

[Prev in Thread] Current Thread [Next in Thread]