Testing FET-6.0.0

[Liviu Lalescu]

Could you please from Settings -> restore to default settings and try again? (exit FET after you restore to default settings).

If you want, we could initiate a TeamViewer on your computer so I can see that. I cannot replicate (Windows 10 64-bit, Ryzen 2700X). I also cannot replicate on my usual openSUSE Tumbleweed GNU/Linux.

[YOUSSEF HOUIET]

I have recorded the screen
link of record (change quality to 720)
https://drive.google.com/file/d/1z791YqzRyg17nKPicYw6Eg4h7VgPO-TH/view?usp=sharing

[Liviu Lalescu]

Thank you!

It surpasses me    I don't know.

Your processor has 2 cores and 4 threads, I searched on the internet. But it should not crash.

[Liviu Lalescu]

I confirm: I have the same behavior on a friend's computer. Laptop i7 with 6 cores/12 threads. Sometimes it just quits, but depends on file and luck.

But why doesn't Windows say: "This program has crashed." etc...?

[Liviu Lalescu]

I tested on another laptop of another friend. Crash on the first simulation, no crash on the ulterior simulations.

I will need to check the code.

[YOUSSEF HOUIET]

Yes, the same for me;
crash just in 1st simulation

[Liviu Lalescu]

Great! Thank you for your help! I know where to look for the error.

Please test this: restore settings to default, then see if it crashes like first time, then if it doesn't crash second, third time. Only if you have time and possibility.

[Liviu Lalescu]

Mr. YOUSSEF HOUIET,

I just put a new snapshot: https://lalescu.ro/liviu/fet/download/test/  . Could you please test it? I hope it won't crash anymore on generate multiple. Test by leaving it finish, by interrupting it, with timetables complete or with no timetable complete.

Please let me know as soon as possible.

Edit: It still crashes on Windows 10 when generating multiple 

[Liviu Lalescu]

I made some more changes, I just put a new snapshot: https://lalescu.ro/liviu/fet/download/test/  . I tested on two computers with Windows 10.

Please let me know as soon as possible if the bug is gone or not.

I hope it won't crash anymore on generate multiple. Test by leaving it finish, by interrupting it, with timetables complete or with no timetable complete.

[Alex Chernous]

Version: 6.0.0-snapshot-7-apr-2021-16_23 (April 2021).
Multiply threads generation (20 min., 4 timetables in 4 threads).

1. When one or two (less then 4) timetable is complete, the CPU load remains high for all CPU's (screenshot);
2. When stop manually generation, only one highest timetable writing to the disk.

[Liviu Lalescu]

1. You forgot the screenshot. But no problem, this is the desired behavior, all processors will work to see which finds a timetable first.
2. Do you mean if no timetable is generated and you interrupt? Also desired behavior, only max placed activities timetables are written in highest-stage form.

[Alex Chernous]

1. I agree, but not understand how I need to learn the theory of threads control.
2. When I press Stop button. I understand. I expected highest timetable for each threads.

[Liviu Lalescu]

So is it OK?

[Alex Chernous]

So is it OK?

Yes, it is OK.

[Liviu Lalescu]

Thank you!