I've been working on a small Windows Mobile game for some time now, and over the past few days I've been working my way around Mappy, the popular, free tile-based map editor. I've discovered that Mappy has some, IMHO, serious shortcomings in regard to its flexibility:

• The Null Tile, tile 0, is annoying. It throws off carefully alligned tile sheets during import.
• Can't manipulate the tile palette, so you can't reallign them.
• Can't import from more than one tile sheet.
• There's no Brush Palette. You have to select them from a menu, and there's a limited amount of slots.
• There's no Objects. Everything's a tile and all tiles are the same size.
• There's no way to place a tile offset from the grid.
• Collision is coursely defined at best.
• And many more...

This seems to stem from the fact that the author of Mappy takes a very narrow view on what, exactly, a tiled map editor should be. At the same time, Mappy was and, to a lesser extent, is purportedly somewhat popular even for commercial 2D, tile-based games (and more-so in the homebrew/indie scene.) Is it possible even dispite these obvious short-comings? Do studios simply write additional tools to address the items I've listed above? I'm willing to concede that some of my concerns may be negated since I'm still new to Mappy and may have simply missed some trick to accomplishing the task. That said, I've written a few tile-based games in my day (mostly RPGS and Platformers) and Mappy really seems to be lacking based on my experience with those projects. Are there better Editors out there, commercial or otherwise? What do they do better? What shortcomings do they have? What features have you always wanted? How could the UI be made more usefull/efficient? I'm so disappointed with Mappy that I'm seriously considering writing my own, and making it available when its finished. [EDIT] Mods: If there's a better forum than Visual Arts please feel free to move the thread. I didn't see anything regarding level design specifically.

You sound like me when I went this route too.

I soon binned it and wrote my own. If I thought it would actually sell, I would've made it more data driven but in the end I just did it the way I wanted it done. Because I only wrote the features I wanted (though with some flexibility) it didn't take long at all.

If you go the self route, you'll also be able to expand it next time and also fix bugs and stuff, obviously.