JDog053 DevBlog

[JDog053]

Started to write some Tile stuff. Managed to simplify my Tile object aswell compared to my original version.

My old system looped through every tile and sets the ones at that location to Active
It would then loop through every tile and check the 4 neighbour tiles to calculate which frame to display + offset (if the tile sheet had multiple tiles)
It would then loop through every tile and check the 4 neighbour tiles to determine which edge to activate of the tile for collision purposes

My newer system loops through every tile and sets the ones at that location to active
It then loops through every tile and checks either 4 or 8 neighbour tiles and produces a bitmask value

Using that value I can set the frame + offset and the collision value of that tile. so now i'm using an int to store my collision instead of 4 bools. not a bit change, but in a big tile grid that could be a huge saving

The TileMap object is also written and contains both the regular Render method and the special Render method which culls every tile not visible on screen which will certainly help lower powered machines.

Its good making progress again!

[fog]

My newer system loops through every tile and sets the ones at that location to active
It then loops through every tile and checks either 4 or 8 neighbour tiles and produces a bitmask value

Using that value I can set the frame + offset and the collision value of that tile. so now i'm using an int to store my collision instead of 4 bools. not a bit change, but in a big tile grid that could be a huge saving

yeah I think using bits is the normal way of doing that for maps.  I've also used that method in a couple of puzzle games too.

[T_M_C]

Good progress Jdog.  

I remember getting huge speed savings when i wrote a cull routine for my shooter.
Before i just assumed that Blitz would cull any images not on screen, which it does, but it still processed the images.
Just putting in a simple check to see if an image was onscreen before drawing it really gave me a big speed boost.
Which i was really surprised about.

I also used a partition system as well for my collision level.
Which is pretty much essential for anything with large levels that isn't tile based.

My RPG does that too and allows for a world of infinite size, but with no speed penalty for the size of the world

TMC

[JDog053]

Thanks folks, yeah its going quite well so i'm pleasantly surprised.

I'm in the process of planning a scrolling/wrapping texture shader to help me do my parralax image layer behind the tile grid but over the background image. Due to the way its done, I should be able to tint and offset the parallax layer and redraw it multiple times to easily add a few layers without much extra effort.

[JDog053]

Well my side project of a visual twitter app using XNA is finished so I can return to focussing on bat game. Since I suddenly have alot of free time now, this project should really make some serious progress.

[TheKhakinator]

Well my side project of a visual twitter app using XNA is finished so I can return to focussing on bat game. Since I suddenly have alot of free time now, this project should really make some serious progress.

You should shoot it to my old uni, they run one on LCDs in the student areas and it's animated and fantastically slow. Stutters something awful, I don't know how they got it so wrong. XD

[fog]

Good to hear of progress JDog 

Well my side project of a visual twitter app using XNA is finished so I can return to focussing on bat game. Since I suddenly have alot of free time now, this project should really make some serious progress.

You should shoot it to my old uni, they run one on LCDs in the student areas and it's animated and fantastically slow. Stutters something awful, I don't know how they got it so wrong. XD

lol.  Probably because they employed a 3rd party developer who didn't give a toss.   

[TheKhakinator]

Good to hear of progress JDog 

Well my side project of a visual twitter app using XNA is finished so I can return to focussing on bat game. Since I suddenly have alot of free time now, this project should really make some serious progress.

You should shoot it to my old uni, they run one on LCDs in the student areas and it's animated and fantastically slow. Stutters something awful, I don't know how they got it so wrong. XD

lol.  Probably because they employed a 3rd party developer who didn't give a toss.   

I more suspect it was done in-house by a student, hence the quality

[fog]

Good to hear of progress JDog 

Well my side project of a visual twitter app using XNA is finished so I can return to focussing on bat game. Since I suddenly have alot of free time now, this project should really make some serious progress.

You should shoot it to my old uni, they run one on LCDs in the student areas and it's animated and fantastically slow. Stutters something awful, I don't know how they got it so wrong. XD

lol.  Probably because they employed a 3rd party developer who didn't give a toss.   

I more suspect it was done in-house by a student, hence the quality

Such a good advert for the uni 

[JDog053]

Made a bold move tonight, i've decided to completely finish off FAT BAT for the PC and Windows Phone first before I worry about porting the whole thing over to my C++/OpenGL engine. Getting it finished in full before porting it, instead of cobbling my engine together with no frame of reference to the game i'm creating

[T_M_C]

 

TMC

[fog]

Made a bold move tonight, i've decided to completely finish off FAT BAT for the PC and Windows Phone first before I worry about porting the whole thing over to my C++/OpenGL engine. Getting it finished in full before porting it, instead of cobbling my engine together with no frame of reference to the game i'm creating

lol  You're almost taking the same route as I have with 1982.  Pretty much finished in C#/XNA and the ported to C++/AGK.

I remember saying at the time this was a good idea as the game was complete and it would let me concentrate solely on learning the language.....and then I spent a year tweaking that "finished" game. 


Regardless of how yo do it, I want to play Fat Bat. 

[JDog053]

True, but it does make the most sense, you guys have already played a semi-finished version of the game and it makes sense just to push that version to completion before worrying about converting it etc. Atleast that way i'll know whether its worth my time porting it if the game is any good lol I should've done this months ago though.

EDIT: Making great progress, added in my newer swankier menu system into this version along with my state traversal system, to allow for stacked gamestates etc. all my file reading code from twitter app has been put into this making it really easy to add custom stuff without having to recompile. log file and replay system are now ported over to C# and implemented.