| |
Where is VERGE going? Displaying 1-11 of 11 total.
1
resident
|
Don't get me wrong, I have no doubt that regular progress is being made. I've just suddenly realised that I'm not sure where the bus is going.
Posted on 2009-07-29 10:36:38
|
Kildorf
|
I'm not sure that I would really describe the progress as "regular" but it's true that stuff is happening. Not to get anyone's hopes up but if things pan out we should have a new engine release VERY SOON, and the current version of the engine has some fairly nifty new stuff.
Where is the bus going? That's a very good question, and I think everyone has a bit of a different answer on that. I have some ideas about what I want to happen, but frankly most of where I, personally, want Verge to go falls into one of two categories:
- It makes it easier for me to work on Geas or it makes Geas run better.
- It makes the engine more viable as a general purpose, fast engine so that I can use it for more games in the future.
As I said though, that's really just my personal goals. Honestly, Verge has been sort of a shoot-from-the-hip sort of thing and I don't know that there ARE overall engine goals. There has been a bit of discussion back and forth in various venues on the net about this but nothing resembling a real direction.
So, let me turn this question around, right back at you resident, and anyone else on the boards (including engine devs, newbies, compo participants, ANYONE):
Where do you want to see Verge going?
We certainly can't please everyone but it would be handy to know what people in the community want so those of us with the free time, experience, and energy to work on the engine can know what we might do for other people, rather than just for ourselves.
Posted on 2009-07-31 09:21:58 (last edited on 2009-07-31 09:22:50)
|
mcgrue
|
Regular consumers of the engine get priority in my mind as far as feature development.
So... if you want to make feature requests and are passionate about them, it really helps if you have a fresh demo to act like a delicious carrot to the engine developers. :)
(hint hint)
Posted on 2009-07-31 12:47:21
|
resident
|
I'm thinking that, with all the new stuff coming in, like the Lua push, maybe it's time to strip some of that RPG-Engine baggage for Verge and intentionally push more towards making it a general Rapid Application Development tool?
So there's less hoops to jump through between getting something on screen? So you don't need to pre-process all your sprites with CHRMAK to use the entity system, etc.
Posted on 2009-07-31 15:49:11
|
mcgrue
|
Generally, I agree.
Specifically, I super-agree.
Posted on 2009-08-03 16:53:53
|
Overkill
|
Rapid Application Development is RAD.
Posted on 2009-08-03 21:12:49
|
Kildorf
|
I'm not sure I completely agree that we should toss all of the RPG stuff; personally I think that's kind of what makes Verge worthwhile over other rapid development game suites like LOVE. Instead, what I'd really like to see is to take the entity and map system and make it more flexible! Make it easy to fall back to a top-down RPG system but also easy to treat entities as physics objects or clickable hotspots or whatever. Does anyone else think that would be worth the work (which would be not-insubstantial)?
As for the image drawing, what would you like to see in terms of "how to put an image on the screen"? Do you just want a persistent "throw these images on the screen, and keep them there until I say otherwise"? (Which really is covered by the sprite system but that has its own hassles! Meaning we should take a look at it and figure out how to make it nicer to use and faster at runtime... (which is already on my list))
I'm not trying to tear your ideas apart or anything, just trying to get to the meat of the matter. :) Please, more ideas, everyone!
Posted on 2009-08-07 06:45:47
|
Overkill
|
We need to modify the map/entity/tileset formats, in a bunch of ways to make the engine more "general purpose".
Tilesets need to allow tilesizes other than 16. We already HAVE a value in the format headers that specifies this. We just don't use it. We need to get the editors to allow smaller tile sizes (like 8x8 for those of us who'd like to make a NES-esque game with similar ability to chunk up the map in tiny parts), or bigger ones (like 32x32 for those who are making hi-res things).
Maps and entities need to be capable of storing meta. Maybe even just a large string that scripts can then load in a custom manner. But there's some information baggage that certain games need, that the "RPG-friendly" core can't provide. Sidescrollers need their own info. Tactical games need their own stuff.
And it's much more convenient if you can tie it to the core data formats, rather than having to code everything.
And okay, entities need custom animations. If anything gets annoying, it's that besides basic walk cycles and idle frames, you have to write all the REST of your animations in VC/Lua code.
Action/RPGs, fighters, and sidescrollers need this for all sorts of complex stances that can't be expressed by the default data. RPGs could also use this to add scripted dialogue animations, like laughs, or your battle system's attacks.
It can be done. It just requires a few modifications to the CHR format, and adding a couple functions to trigger/process these animation scripts engine-side.
Okay, there's more too. Like there are some issues that need clear planning ahead if we want to allow it, because it requires lots of refactoring/maintenance work. I can't really talk about these right now since I'm heading out for a bit. But I'll try and explain when I get back.
Stuff like alpha blitting (very nice, needs further rewrite of all blitters), hardware accel (I think will never be added due to API changes and extra complexity all around), and other things. Anyway.
Posted on 2009-08-07 17:34:59
|
resident
|
Yeah, just throwing an image on screen is easy. But doing that essentially kicks out the entity system, and that's way too neat to ignore. No matter what kind of game you're making the entity system is pretty much always going to be useful (unless you're making a text adventure, I suppose) There's a lot of legacy baggage around creating and using entities though.
Idly, a generic collision detection system might be neat, too, though I'm more hesistant on that, since how you choose to have your game handle collisions is a large part of what makes it unique, and a generic collision system might just result in a load of generic games.
As I mentioned in another topic, having devtool functionality directly in the engine itself would be a biggie (having functions to allow you to save maps, for example, could go a long way to allieviating the need to have external devtools at all)
Ideally, I'd like to see just about everything editable by a text editor, short of graphical and audio files.
Posted on 2009-08-17 10:37:13
|
Kildorf
|
resident: Actually, Overkill went and added a couple of functions like that a little while ago. I guess we weren't particularly good at letting other people know!
FileWriteMap() -- http://verge-rpg.com/docs/view.php?libid=1&function=415
FileWriteVSP() -- http://verge-rpg.com/docs/view.php?libid=1&function=395
FileWriteCHR() -- http://verge-rpg.com/docs/view.php?libid=1&function=393
Now, are these really enough for building tools? They might be! I'm not sure. I'd love to see someone try though.
Posted on 2009-08-17 20:17:37
|
resident
|
Ooh, toys! It seems like you could do a lot with those.
Posted on 2009-08-19 02:31:26
|
Displaying 1-11 of 11 total.
1
|
|
