5
$\begingroup$

Bug introduced in 13.2 or earlier and fixed in 14.1

Save your work before attempting.

I realise this may get fixed in later versions – I am more interested in the why.

On OSX 14.4 and Mathematica 13.2 the code below will crash the kernel if FullSimplify is uncommented.

Block[{α0, β0, ϵ, γ,  δ, θ}, 
 1/8 α0 (8 + 4 ϵ Sin[θ]^2 + 
    4 (δ + (δ - ϵ) Cos[2 θ]) Sin[θ]^2 - (8 α0^2 (δ - ϵ) * 
  Cos[θ]^2 (δ + ϵ + (δ - ϵ) Cos[2 θ]) Sin[θ]^4)/(α0^2 - β0^2) - 
 (δ + ϵ + (δ - ϵ) Cos[2 θ])^2 Sin[θ]^4)(*//FullSimplify*)
]

If I evaluate the expression and run something like %//FullSimplify after, it just runs endlessly.

What is more bizarre is I could swear this code was working (with the FullSimplify!) a few weeks ago but my time machine is broken and can no longer verify this statement...

Marking this as a bug since I got independent confirmation this is not to do with my setup. Either way, if someone can offer some plausible reason why this happens I would really appreciate it. This has been reported to Wolfram Research.

---EDIT---

Since first posting, this has been confirmed by WR tech support and carries on to v14 on Mac and Windows.

Improve this question
$\endgroup$
10
  • 5
    $\begingroup$ I can confirm the crash happening also in 14.0.0 for Microsoft Windows (64-bit) (December 13, 2023). As usual – kernel crash is something that certainly isn't supposed to happen, so you should report this to the technical support. $\endgroup$ Commented May 27, 2024 at 13:19
  • $\begingroup$ Thanks, justified to mark with bugs tag $\endgroup$ Commented May 27, 2024 at 13:49
  • $\begingroup$ With v14 on a Mac, it doesn't crash if you use // TrigToExp // FullSimplify; however, you end up with a higher LeafCount than when you started. $\endgroup$ Commented May 27, 2024 at 17:39
  • 2
    $\begingroup$ I am voting to close this question, since you already got the confirmation and the answer from the WRI – the kernel crashes because there is something wrong in the kernel code. $\endgroup$ Commented May 30, 2024 at 10:29
  • 4
    $\begingroup$ I’m voting to close this question because it is already solved and OP agrees the closing of the question. $\endgroup$ Commented Sep 30 at 1:20

0

Reset to default

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.