World Reversed

My work time has been getting compromised by doing chores and trying to start jogging again, so I haven’t made as much progress in my event system rewrite as I would like.  The task is proving to be much larger than I initially thought, but I am making progress, if you define progress as breaking everything.  I’ve got it to the point where the main menu works again though.

The short version is that I created a base event manager component, that reads in an XML file with very basic event node data, mostly just an id, a type, and a subtype id.  It then checks the type, and sends it to the correct subcomponent manager to process.

I sort of have the map processor partially working now, but it has large chunks of legacy code in it where it’s trying to process all of it’s own nodes in one go, instead of passing an exit id back up to the main event manager.  Moving forward at least, should hopefully have the base rewrite wrapped up by tomorrow, and can actually move on to basic dialog.

Felt like I was hit by a bus.  Having really bad allergies, so I didn’t get anything done.  Will get back to it Monday.  For reals.  Unless I’m dying of the flu.

Okay, I lied yesterday, I still don’t have anything to show, but it’s not from lack of trying this time.  The core rewrite of the events system is proving to be a much larger task than anticipated.  I started by breaking out the initial couple screens in the demo out into the maps subsystem, and had to rewrite the manager to read in data in a new format for that.  That part is working again.

What happened next still has my code base a broken mess however.  I ended up breaking out the events system from it’s subcomponents, and the main game loop, into it’s own manager.  I’m still dealing with the fallout from this, and right now it won’t transition properly into a combat event.  I should have this mostly fixed tomorrow, and hopefully have the basics of the dialog system up and running.  Crossing my fingers but making no promises this time.

So a friend basically smacked some sense into me today.   I’d been getting mentally stuck on working on the minutiae of the combat engine, and missing the overall big picture that I might ultimately just end up scrapping or significantly rewriting some of the features I was stuck on, since I still don’t even have a full picture of what I’m going to do there yet.

In that light, I put that aside for now, and started work on the dialog and map subsystems.   Started adding basic stubs for both, but I have to rewrite my events system slightly to move some of the data around into their own compartments.  Should hopefully have something rudimentary to show tomorrow, and I actually mean it this time, damn it.

So I’ve been in a bit of a slump the last few weeks.  I’ve been having a fairly large amount of trouble getting myself in the correct mindset for working on this project.  No one to blame but myself really.  Been fiddling with getting basic collision detection in for the last week or so but didn’t get anywhere, so I sat down tonight and forced myself to get most of it implemented.

I mostly implemented it straight from the transformed sprites example on the app hub, but with modifications to work with my objects.  Right now it’s a very rudimentary implementation that tests one texture against another, which ultimately won’t work for my final solution, but it works for figuring out how to go about this for now.  I still don’t have it working properly, so I don’t have anything to show, but I should in the near future, after I figure out how to debug why it’s not working properly.  It’s part of the way there though.

Okay, I got the movement mostly working again, and have them locked in 4 directions at a 45 degree shift to point them in isometric angles.  I started working on the collision today, and went through one sample, only to find out I was working with the wrong one that didn’t deal with rotating objects.  Oops.  I’ll pick it back up again tomorrow and hopefully get it working properly, and then video the whole thing.

Okay, I’ve gone far too long without updating.  The last week, I’ve consolidated most of my squad handling code into it’s own class where it always mostly belonged, cleaning up my combat handling module nicely.  My primary focus has been reworking movement, to lock it in four directions, and break it down into units of movement.  This will allow me to more easily set a static range the units can move, and also easily implement wait commands in the middle, since combat will happen simultaneously.

I’ve also been working on how to handle collision a bit more properly, and looked up a bunch of per-pixel collision solutions, since that’s what I’m going to have to work with, since I’m dealing with 2d isometric, without a flat 2d grid stored  in memory, at this time.  I may have to step back and rework that idea later, but with my current implementation, I don’t need that level of detail for what I’m doing.

Sadly, I haven’t been  nearly as productive as I would have liked to have been this week though, and haven’t made nearly as much progress on this as I would have liked.  I am nearly done though, and should have it finished up by next week.

Well, made some moderate progress today.  Now that I have movement somewhat working, it was time to move on to collision.  To even get started on this, I had to modify the level data to take in level fixtures and collision rectangles.  With these added, I now have rocks being drawn in the level.  Exciting.  I don’t actually have collision figured out yet, because since the arrow rotates, and I’m basing initial collision on that, the solution is not as easy as it could be.  I did some basic research on it, and I might have to do a pixel by pixel compare, which seems like it might be ugly, but I’ll dig into it more in the next few days.

Made a large amount of progress today, although it does not look that impressive.  The first thing I did was refactor the combat engine to handle sprites in squad groups instead of individual characters.  Right now it’s only doing squads of one character each, but it’s a start, and helped clean up the code a great deal.

After that, I finally got around to mostly finishing up movement.  I added a range limiter on the movement arrows, and figured out how to get the sprites to follow the paths set down by the player.  It’s still not completely perfect or finished, but it’s a start.

Finally got back to fixing the broken movement arrows.  Made a bunch of progress in making them link together properly.  I still don’t have range limiting back on in on it, nor do I have the sprites following the paths, but that’s up next.