4
$\begingroup$

Update: I added small movie showing the crash, since others who tried it so far could not reproduce it on their system.

enter image description here


Original question

I am finding new problems with DSolve in 10.3.1. Here is another crash. could someone see if they can duplicate this? This happens on windows 7, 64 bit. It is reproducible. I was not sure if I should add this to my earlier finding here or make new question about it since it is different problem and do not know if they are related internally or not. Will also send email to support@wolfram.com to investigate. ps. It takes 3 minutes to get the crash. PC running intel fast CPU with 16 GB ram.

First one

$Version
Clear[x, y];
eq = x^3 + y[x] y'[x] + x y'[x]^2;
DSolve[eq == 0, y[x], x]

enter image description here

With version 9, it does not crash, but also it is still running after 20 minutes. So it looks like it hanged. But no kernel crash ! (which is better, a hanged kernel or one that crashes? :)

second one

Here is a second crash on new problem I just found: (all with new fresh Kernel) This one took about 5 minutes.

$Version
Clear[x, y];
DateString[]
eq = D[y[x], x]^3 - 2*y[x]*D[y[x], x] + y[x]^2;
DSolve[eq == 0, y[x], x]

Mathematica graphics

Third one

This one is very similar in pattern to the above, small variation. Thought to put it here for reference.

$Version
Clear[x, y];
DateString[]
eq = D[y[x], x]^3 - y[x]*D[y[x], x]^2 + y[x]^2;
DSolve[eq == 0, y[x], x]

Mathematica graphics

$\endgroup$
  • $\begingroup$ How long did it take to see the kernel crash on V 10.3.1? $\endgroup$ – Moo Jan 9 '16 at 8:12
  • $\begingroup$ No crash on linux with both 10.3.0 and 10.3.1 . $\endgroup$ – b.gates.you.know.what Jan 9 '16 at 9:38
  • $\begingroup$ @b.gatessucks thanks for checking. It must be a windows only issue then. I am using windows 7, service pack 1, 64 bit. $\endgroup$ – Nasser Jan 9 '16 at 9:45
  • $\begingroup$ @Nasser: For me, the first example was running for hours, with no signs of anything. Certainly that is very unusual, so the thing was hung, but just in a different way. $\endgroup$ – Moo Jan 9 '16 at 14:28
  • 1
    $\begingroup$ Actually, looking closely at the solution to the first, i.stack.imgur.com/ggBj3.png, one can see an invalid integral, !Mathematica graphics $\endgroup$ – Michael E2 Jan 10 '16 at 0:38

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.