Re: need help to find an octave crash

[Tatsuro MATSUOKA]

I have tested octave-3.0.0 on mingw by me octave-2.9.19 by Michael.

The following results are the former. But the latter gives the same results and 
step(sys,1,3,100) command seemed to show out a reasonable graph.

I have enough time to purphase your origin. But at least latest two octaves for 
windows have no
problem for your issue.    

Regards

Tatsuro
*****************************************
octave.exe:2> n=poly([ -11 -10 ])*10000
n =

     10000    210000   1100000

octave.exe:3>
octave.exe:3> n=poly([ -11 -10 ])*10000
n =

     10000    210000   1100000

octave.exe:4>
octave.exe:4>
octave.exe:4>
octave.exe:4> d=poly([ -2-3j -2+3j -2.1-3j -2.1+3j -3-3j -3+3j -3-3j -3+3j -5.5
-5 -5 -1 ])
d =

 Columns 1 through 3:

  1.0000e+000 + 0.0000e+000i  3.6700e+001 + 0.0000e+000i  6.4241e+002 + 
0.0000e+000i

 Columns 4 through 6:

  7.1043e+003 + 0.0000e+000i  5.5257e+004 + 2.2737e-013i  3.1819e+005 + 
4.6612e-012i

 Columns 7 through 9:

  1.3899e+006 + 4.0359e-011i  4.6356e+006 + 1.9998e-010i  1.1701e+007 + 
6.3653e-010i

 Columns 10 through 12:

  2.1745e+007 + 1.2637e-009i  2.8077e+007 + 1.2915e-009i  2.2285e+007 + 
5.0022e-010i

 Column 13:

  7.7664e+006 + 0.0000e+000i

octave.exe:5> n=poly([ -11 -10 ])*10000
n =

     10000    210000   1100000

octave.exe:6>
octave.exe:6>
octave.exe:6>
octave.exe:6> d=poly([ -2-3j -2+3j -2.1-3j -2.1+3j -3-3j -3+3j -3-3j -3+3j -5.5
-5 -5 -1 ])
d =

 Columns 1 through 3:

  1.0000e+000 + 0.0000e+000i  3.6700e+001 + 0.0000e+000i  6.4241e+002 + 
0.0000e+000i

 Columns 4 through 6:

  7.1043e+003 + 0.0000e+000i  5.5257e+004 + 2.2737e-013i  3.1819e+005 + 
4.6612e-012i

 Columns 7 through 9:

  1.3899e+006 + 4.0359e-011i  4.6356e+006 + 1.9998e-010i  1.1701e+007 + 
6.3653e-010i

 Columns 10 through 12:

  2.1745e+007 + 1.2637e-009i  2.8077e+007 + 1.2915e-009i  2.2285e+007 + 
5.0022e-010i

 Column 13:

  7.7664e+006 + 0.0000e+000i

octave.exe:7> sysout(sys)
error: `sys' undefined near line 7 column 8
error: evaluating argument list element number 1
octave.exe:7>
octave.exe:7> sys=tf(n,d)
sys =
{
  num =

       10000    210000   1100000

  den =

   Columns 1 through 3:

    1.0000e+000 + 0.0000e+000i  3.6700e+001 + 0.0000e+000i  6.4241e+002 + 
0.0000e+000i

   Columns 4 through 6:

    7.1043e+003 + 0.0000e+000i  5.5257e+004 + 2.2737e-013i  3.1819e+005 + 
4.6612e-012i

   Columns 7 through 9:

    1.3899e+006 + 4.0359e-011i  4.6356e+006 + 1.9998e-010i  1.1701e+007 + 
6.3653e-010i

   Columns 10 through 12:

    2.1745e+007 + 1.2637e-009i  2.8077e+007 + 1.2915e-009i  2.2285e+007 + 
5.0022octave.exe:8>
sysout(sys)
Input(s)
      1: u_1

Output(s):
     1: y_1

transfer function form:
10000*s^2 + 210000*s^1 + 1.1e+006
----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
-----------------------------------
1*s^12 + 36.7*s^11 + 642.41*s^10 + 7104.3*s^9 + (55257 + 2.2737e-013i)*s^8 + 
(3.1819e+005 +
4.6612e-012i)*s^7 + (1.3899e+006 + 4.0359e-011i)*s^6 + (4.6356e+006
+ 1.9998e-010i)*s^5 + (1.1701e+007 + 6.3653e-010i)*s^4 + (2.1745e+007 + 
1.2637e-009i)*s^3 +
(2.8077e+007 + 1.2915e-009i)*s^2 + (2.2285e+007 + 5.0022e-010i)*s^1
+ 7.7664e+006
octave.exe:9>
octave.exe:9> step(sys,1,3,100)



--- Doug Stewart <address@hidden> wrote:

> I am running qtoctave on windows XP with octave 2.9.15
> 
> Would someone try this script and see if the new ver. of octave crashes 
> or not.
> 
> My system has a seg fault
> 
> n=poly([ -11 -10 ])*10000
> 
> d=poly([ -2-3j -2+3j -2.1-3j -2.1+3j -3-3j -3+3j -3-3j -3+3j -5.5 -5 -5 
> -1 ])
> 
> sys=tf(n,d)
> 
> sysout(sys)
> 
> step(sys,1,3,100)
> 
> 
> 
> 
> 
> 
> Doug
> _______________________________________________
> Help-octave mailing list
> address@hidden
> https://www.cae.wisc.edu/mailman/listinfo/help-octave
> 


--------------------------------------
New Design Yahoo! JAPAN  2008/01/01
http://pr.mail.yahoo.co.jp/newdesign/