search

jnonline / ogstudio

Automatically exported from code.google.com/p/ogstudio
0 stars 0 forks source link

Check crashing save file #232

Closed GoogleCodeExporter closed 9 years ago

GoogleCodeExporter commented 9 years ago 
There's a magic combination of turns that isn't loadable at startup.
BronzeBeard sent me one:

Original issue reported on code.google.com by korn...@gmail.com on 9 May 2011 at 3:37

Attachments:

GoogleCodeExporter commented 9 years ago

Original comment by korn...@gmail.com on 9 May 2011 at 3:52

GoogleCodeExporter commented 9 years ago 
Under Linux it says that layout is empty. Under Windows it is not.

Original comment by korn...@gmail.com on 9 May 2011 at 4:15

GoogleCodeExporter commented 9 years ago 
There's one addition #13 byte at 45(dec) offset. After removing it the Linux 
version just crashed.

Original comment by korn...@gmail.com on 9 May 2011 at 6:12

GoogleCodeExporter commented 9 years ago 
BronzeBeared gave another save that crashes

Original comment by korn...@gmail.com on 15 May 2011 at 3:55

Attachments:

GoogleCodeExporter commented 9 years ago 
BronzeBeared said that upon Bill Gates if you exit, then you get broken save if 
you choose to save upon exit.

Original comment by korn...@gmail.com on 15 May 2011 at 4:54

GoogleCodeExporter commented 9 years ago 
And it's not always happening.
I suspect it has smth with the way we terminate under Windows.

Original comment by korn...@gmail.com on 15 May 2011 at 5:12

GoogleCodeExporter commented 9 years ago 
May be seed does not always define globally unique rand() numbers. Try to store 
all tiles in a save.

Original comment by korn...@gmail.com on 20 May 2011 at 9:01

GoogleCodeExporter commented 9 years ago 
I need to stress test layout generations. Just do a separate test for Linux and 
Windows which will be generating things each second, and let it run for several 
hours.

Original comment by korn...@gmail.com on 20 May 2011 at 9:06

GoogleCodeExporter commented 9 years ago 
i can actually use only 'labyrinth' layout for testing.

Original comment by korn...@gmail.com on 20 May 2011 at 9:17

GoogleCodeExporter commented 9 years ago 
started the small app. i will later need to regenerate with those seed.

Original comment by korn...@gmail.com on 20 May 2011 at 9:55

GoogleCodeExporter commented 9 years ago 
i do generate it once a second with current time. stupid.
fix start time. and just generate the whole day.

Original comment by korn...@gmail.com on 20 May 2011 at 10:13

GoogleCodeExporter commented 9 years ago 
may be i just need to check the erroneous seed?

Original comment by korn...@gmail.com on 20 May 2011 at 10:18

GoogleCodeExporter commented 9 years ago 
and i need to check that every second. under both Linux and Windows.

Original comment by korn...@gmail.com on 20 May 2011 at 10:24

GoogleCodeExporter commented 9 years ago 
find out how many memory will it take to store positions in save file.

Original comment by korn...@gmail.com on 20 May 2011 at 1:52

GoogleCodeExporter commented 9 years ago 
Each tile position takes 2-3 bytes. each shuffle saves the whole play field. so 
in the works case we'll have about 0.5M for a save file.
Consider MT199337 algorithm to generate random numbers.
Still, include a full set of positions in 0.8 just to find out how it all works.

Original comment by korn...@gmail.com on 20 May 2011 at 2:20

GoogleCodeExporter commented 9 years ago 
System rand() function is suspected to generate non-consistent results. There's 
no way to fix the issue thus. It will be recreated for 0.8 if the issue still 
remains.

Original comment by korn...@gmail.com on 21 May 2011 at 9:26