XlibSweeper-Final DL Available!

Floodkiller · Post by **Floodkiller** » Wed 31 Jan, 2007 2:06 pm

maybe it might be a problem with my calc. Recently I tried to optimize one of my older games, only to find that the group was corrupted. When I redownloaded my uncorrupted group from my computer (which I sent to ticalc.org), it had some program recognization errors even though the programs clearly existed.

And yes, I did download xLIB within the past 3 months

kalan_vod · Post by **kalan_vod** » Wed 31 Jan, 2007 11:58 pm

Floodkiller wrote:maybe it might be a problem with my calc. Recently I tried to optimize one of my older games, only to find that the group was corrupted. When I redownloaded my uncorrupted group from my computer (which I sent to ticalc.org), it had some program recognization errors even though the programs clearly existed.

And yes, I did download xLIB within the past 3 months

That would be TI Connect, an old version would corrupt groups...

Floodkiller · Post by **Floodkiller** » Fri 02 Feb, 2007 9:20 pm

Ok, update here:

-Greyscale has been redone quite a bit after a math test. Now, it runs butter smooth with barely and visible flickering, at the cost of about 133 bytes.
-Small optimizations have been made according to some stuff repeating inside the game that only needs to be run once
-Replaced X and Y variables when used in the program in order to prevent being trapped in the 0th row at the start.

Bugs still fixing:
-Greyscale version commits a very strange flicker on the first loading, but runs fine after pressing one button. Probably a problem in first display that I will have to check over.
-More optimizations to speed (maybe a for( loop running the program instead of a while) and memory (any other parts being repeated that dont have to be)

crzyrbl · Post by **crzyrbl** » Sun 04 Feb, 2007 2:56 am

Sounds pretty solid. Are you still planning on mixing it up eventually?

Floodkiller · Post by **Floodkiller** » Sun 04 Feb, 2007 7:47 pm

yea, after I release this version, ill either work on some of my other projects, or make a 3d version. I already have a small bit of groundwork on making a matrix that stores information three dimensionally without the aid of other matrices.

Floodkiller · Post by **Floodkiller** » Wed 07 Feb, 2007 4:19 pm

Floodkiller wrote:Ok, update here:

Bugs still fixing:
-Greyscale version commits a very strange flicker on the first loading, but runs fine after pressing one button. Probably a problem in first display that I will have to check over.
-More optimizations to speed (maybe a for( loop running the program instead of a while) and memory (any other parts being repeated that dont have to be)

Fixed the greyscale bug, for some reason the program creates blank screens (except for text) when storing the first picture of the program. So, I created a dummy picture, then overwrote it when actually storing the picture. About 2 more bytes for that fix.

Also, made a small speed optimization by replacing the while loop in the greyscale version with one that counts time at the same time, as well as commiting an endgame if your time goes over 999 (9990 when you count in the fact that the process is 10 times faster than the counter should be, at least about that). Still wondering how to fix that into the B&W version to increase some speed, but in that version I already have the while loop pulling an arguement instead of a 1, so it might be a little more difficult.

Other than that, I am also working on implementing the "can't lose on first try" routine, which will take a little longer to do, but I'll get it done before release.

Floodkiller · Post by **Floodkiller** » Thu 08 Feb, 2007 9:39 pm

Small progress, but at 98%, small progress means almost there!

I have a "version" of not losing on the first try, which means that instead of bothering to code out an entire routine for changing the mine around and rewriting two big parts of the matrix (as well as the fact that the routine that I "had" worked out took about 3-5 seconds to execute, requires an extra variable, and will always have an extra check that slows stuff down for the rest of the game). So, instead, I made it so that the upper left corner of the matrix (also where your cursor starts) is safe no matter what inside the matrix.

I also made some optimizations to the B&W version for some repeated code. I also made it so that the time limit of 999 is applied inside the code, as well as the same effects will happen as in the greyscale version to show you either quit the game or ran out of time.

Another good thing is that I added a random board mode, which means that instead of doing the same three settings, or bothering to pick dimensions and mines for yourself, the game will randomize a board height of 7-30, a board width of 12-30, and a mine density ratio of 15% to 30% of the area of the board (I also made sure that the game can be finished in time by testing on the highest settings, and it is quite possible for me to get it with over half the time to spare, so if you get 999 time and lose, its your fault for being slow

).

Finally, I increased some of the mine count on medium and hard in order to have an increasing difficulty of mine density rather than just increase of board with same mine density all across.

All I have to do is make a few more optimizations that my skill level in coding right now sees, then I will make it a final version.

Edit: 2 triple posts and a double post or more in this thread! Is that a bad thing?

Floodkiller · Post by **Floodkiller** » Mon 12 Feb, 2007 3:51 pm

I had accidentally deleted the B&W version that I had optimized, but I recoded it back to its optimized state (as far as I can remember). I also took out the loading time for changing the matrix into a decimal format and now runs through it in one formula.

Also, random mode is finished, some for loops were taken out and optimized, as well as some other speed and memory boosts. I think that it is pretty much ready for a final release. Any other bugs can be reported then.

And its about time too, I am done with minesweeper for a while, as I am completely bored of looking through the code over and over for optimizations

.

Floodkiller · Post by **Floodkiller** » Thu 22 Feb, 2007 3:38 pm

Well, because I found out there was an error in the greyscale minefield creation program (probably made by accidentally deleteing a required variable), I was forced to fix that, and while I was at it, decided it would be pretty cool if you could have less than a 7x12 matrix. So I threw that in with some small adjustments

.

Also, I am doing a final search as well through the program for any missed optimizations.

Floodkiller · Post by **Floodkiller** » Wed 14 Mar, 2007 10:25 pm

The final download is here

!
http://www.savefile.com/files/553484
No screenshots at the moment. When I do get the time to make them, I will upload it to ticalc.org

Tell me if you spot any bugs that I might have missed.

Post by **tr1p1ea** » Wed 14 Mar, 2007 11:15 pm

Pretty cool, here is a screenie for you

:

Floodkiller · Post by **Floodkiller** » Thu 15 Mar, 2007 12:41 pm

Awesome!! Thanks a ton, tr1p!

Floodkiller · Post by **Floodkiller** » Mon 23 Apr, 2007 8:35 pm

Version 1.1 Update Log (75% done):

-Fixed the greyscale text on the greyscale version.
-Added a timer into the info bar.
-Added a pause button.
-Made optimizations.
Still in Progress:
-Time attack mode (as suggested by crzyrbl).

As soon as I get into a programming mode again, I should be able to finish up the time attack mode, and then I'll submit an update to ticalc.org.

Post by **tr1p1ea** » Tue 24 Apr, 2007 7:17 am

Coool! Time attack mode sounds great, lets hope you can find some programming time to add such features. Will be awesome once its done and dusted, and up on ticalc.org

.

* looks at FloodKiller's other projects in his sig --- gets excited

*

Floodkiller · Post by **Floodkiller** » Thu 26 Apr, 2007 3:16 pm

Well, my calc decided to pull a full out crash. I suspect that some of the games that I was playing may have clashed with each other or something. Anyway, my calc kept crashing over and over in a cycle, so I wiped the whole memory after backing it up on a comp. It seems to be running fine right now, so I am suspecting that it was at least one or more of the programs that must have messed it up. I am slowly introducing a program at a time to see what did it.

What does this have to do with xLIBSweeper? Well, the crash corrupted both groups in the main program that you have to run. I am hoping that it wasn't corrupted before I backed everything up, but I can't be sure until I try it again on my calc. If so, I will have to start over from 1.0. This won't affect my other projects, though, as they were just sprite work and map work, and I have all that on paper to redo in case they were corrupted.