Timeline for Any workarounds to help DSolve solve this first order standard d’Alembert differential equation? (no initial conditions)
Current License: CC BY-SA 4.0
12 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Aug 9 at 10:47 | history | edited | Michael E2 | CC BY-SA 4.0 |
Responded to comment
|
| Aug 9 at 10:38 | comment | added | Michael E2 | @lotus2019 Interesting. The same happened to me on V14.1/V14.0. I was about to concede, when on the 3rd try, the first one succeeded! Hacks are unstable. It turns out the length of the time constraint seems significant. Somehow 1 sec. was short enough when I ran this a couple of weeks ago. Today 0.2 sec. works, but 0.3 fails V14.1 fails on 0.2 but succeeds if 0.1. Hacks generally are not the way to fix things. I just use them mainly to check and illustrate the bug. Thanks for the notice! | |
| Aug 9 at 7:48 | comment | added | lotus2019 |
Hello, the result of running both of these code snippets is that Solve is unable to compute. 14.0.0 for Microsoft Windows (64-bit) (December 13, 2023)：Solve::nsmet: This system cannot be solved with the methods available to Solve. General::stop: Further output of Solve::nsmet will be suppressed during this calculation.
|
|
| Jul 21 at 3:34 | comment | added | Michael E2 |
@Nasser I coaxed (bullied?) the singular solution out of DSolve. A pretty severe hack! Not really of general use, I'm afraid.
|
|
| Jul 21 at 3:31 | history | edited | Michael E2 | CC BY-SA 4.0 |
Added addendum
|
| Jul 20 at 19:11 | history | edited | Michael E2 | CC BY-SA 4.0 |
Responded to comment
|
| Jul 20 at 18:57 | comment | added | Michael E2 | @Nasser BTW, it looks like there's a bug in the singular solver. It gets a list of candidates and tests them. It throws out all of them if any of them fail. In your case, it gets your solution and a spurious one, $y=x$. Consequently, it throws out the one works. | |
| Jul 20 at 18:55 | vote | accept | Nasser | ||
| Jul 20 at 18:55 | comment | added | Nasser | My second run was fast, 2.4 seconds. I only tried it twice, each from Fresh kernel. Yes, it looks like caching issues, where first time, some code gets loaded first time. Ok. | |
| Jul 20 at 18:52 | comment | added | Michael E2 | @Nasser No, I'm using V14.0.0 (Mac ARM). It looks like some internal cache gets set. When I time it from a fresh kernel, it's 84 sec. for me, too. Second time, it's 2.4 sec. Probably something was cached when I tried your DSolve code without any changed. I can't explain why your 2nd run is still slow. I doubt it's a system thing. | |
| Jul 20 at 18:38 | comment | added | Nasser | are you using internal version of 14.1? Because for me on V 14. on windows 10, your code takes 84 seconds to complete. Not 2.4 seconds? Here is screen shot FYI i.sstatic.net/BHTK8Luz.png tried it 2 times., same timing. From clean kernel also. | |
| Jul 20 at 18:14 | history | answered | Michael E2 | CC BY-SA 4.0 |
