Welcome, Guest. Please login or register.

Login with username, password and session length

 
Advanced search

1411274 Posts in 69323 Topics- by 58380 Members - Latest Member: bob1029

March 28, 2024, 03:43:01 AM

Need hosting? Check out Digital Ocean
(more details in this thread)
TIGSource ForumsCommunityDevLogsM.I.N.T (Mecha, Infantry, and Tactics)
Pages: 1 2 3 [4] 5 6 ... 26
Print
Author Topic: M.I.N.T (Mecha, Infantry, and Tactics)  (Read 70034 times)
Gregg Williams
Level 10
*****


Retromite code daemon


View Profile WWW
« Reply #60 on: January 30, 2013, 02:58:53 AM »

I'm not sure if you're aware of this or not, but why not use TileEd instead of creating your own tool.  As far as I can see it should support all the features you need.
Do you mean for building tile info/properties, or as an actual map editor?

I could possibly see it working for assigning tiles in a tile set various properties and saving that information out, though it would certainly be less than optimal given its a generic system. Just as an example here, to my knowledge TileEd has no way to automatically generate tile height information per tile like the above little tool does. So this would be another property we would have to generate with a 2nd tool, and then manually enter for no doubt hundreds of tiles.

As a map editor though, its far from what we desire. To my knowledge it doesn't even support staggered isometric maps, but only diamond shaped ones. Its also again a very generic setup, and not customized to meet the game's specific needs, and as such is a less than ideal experience for anyone trying to make maps for the game. Which ultimately leads to the other issue. Could we really spend multiple development years building a game, battle.net like online system, likely charge 20+ dollars for a copy of the game, and then bundle it with some free third party generic tile map editor?

I think not. Map editor wise, our goal is a very simple and easy to use editor that makes it pleasant for people to make maps, and is highly specific to the game's needs. Looking at the Starcraft map editors compared to a generic tile map editor would be a good comparison.
Logged

motorherp
Level 3
***



View Profile
« Reply #61 on: January 30, 2013, 03:25:12 AM »

I meant for both building tile info/properties and as a map editor.

As an internal tool for creating your own maps I'm sure you'd find TileEd more appropriate than you imagine with a little creative thinking.  For example, staggered maps are really nothing special which require their own unique tool.  Its simple to create a diamond map which encompasses the area of your staggered map and simply not fill out the tiles which fall outside of the desired staggered map area.  All you'd then need to do is clamp the camera to the desired area and maybe remap your tile co-ordinates if your game logic already works with staggered co-ords.  Tiles, layers, maps etc can be tagged with any additional parameters you wish within TileEd which simply get spit out as xml and thus could be used to convey pretty much any additional game information you require.  Finally if you do need additional automated functionality, eg. your tile height calculator, TileEd is open source so you could incorporate this functionality into the tool yourself.  That would be faster than writing your own tool from scratch.

I wasn't aware however that your intention was to release the map editor as an external tool for players to create their own maps.  In this case I agree, better to go with your own solution in order to provide a more focused and intuitive user experience.
Logged
Gregg Williams
Level 10
*****


Retromite code daemon


View Profile WWW
« Reply #62 on: January 30, 2013, 03:39:03 AM »

I wasn't aware however that your intention was to release the map editor as an external tool for players to create their own maps.  In this case I agree, better to go with your own solution in order to provide a more focused and intuitive user experience.
Yeah thats the big issue really. Frankly we've always developed our own tools, even for internal use, and we do have a nice map editor currently but its for regular tile work, and not isometrics. Its also much like TileEd, completely unsuitable imo for actual gamer use, and also in our own editor's case lacks cross platform support.

Making everything very focused for actual users is part of what is really taking a long time to do, and will continue to take a long time to do. Its like the road tiles, we don't want a user to ever see a sheet of tiles and have to be picking tiles one by one, and so forth. They should just be able to click on the road brush, and paint out roads to their heart's content. Which means the tiles need to work well with randomized layout vs requiring careful placement, and we may also need special logic to support that cause. (Like variety pieces which are meant to work in 2x2 configurations, etc)

Likewise things like buildings, which really encompass ground, shadow, object, and even overhead tile layers, should just be a building in a nice little list that can be selected, snaps to the tile grid when moved around with the mouse, and can be placed with a single click.

This also leads into all of the auto tiling functionality and other junk desired. Did you just drop some other terrain type over your roads? Best that everything magically re-configures itself to transition nicely.

Then of course you also have all the scenario information, and game modes, and potentially scripting and so forth to integrate as well.

Its a pretty daunting task admittedly. It wouldn't surprise me, if it doesn't work out to be more difficult to create the editor to the level that we wish it, than to create the game, and even all the online servers and associated services.
Logged

motorherp
Level 3
***



View Profile
« Reply #63 on: January 30, 2013, 03:46:20 AM »

Its a pretty daunting task admittedly. It wouldn't surprise me, if it doesn't work out to be more difficult to create the editor to the level that we wish it, than to create the game, and even all the online servers and associated services.

Haha, yeh I can see that  Cheesy.  It's so much harder to put a tool in the hands of a player than in the hands of a developer.  I went down that road once with one of my projects and in the end got so sick of having to monkey proof everything that in the end I abandonned the idea.  I probably spend about 90% of my time making the tool interfaces and such and only 10% actualy making a game  Yawn.

Good luck with this project anyway sir, you've certainly got my attention, it looks ace  Gentleman.
Logged
Gregg Williams
Level 10
*****


Retromite code daemon


View Profile WWW
« Reply #64 on: January 30, 2013, 04:04:51 AM »

Its a pretty daunting task admittedly. It wouldn't surprise me, if it doesn't work out to be more difficult to create the editor to the level that we wish it, than to create the game, and even all the online servers and associated services.

Haha, yeh I can see that  Cheesy.  It's so much harder to put a tool in the hands of a player than in the hands of a developer.  I went down that road once with one of my projects and in the end got so sick of having to monkey proof everything that in the end I abandonned the idea.  I probably spend about 90% of my time making the tool interfaces and such and only 10% actualy making a game  Yawn.

Yeah.. thankfully I've spent many years previous to this doing professional tools coding, though not for games. As such I'm sadly well aware of the hurdles and frustrations it can bring Cheesy

Honestly if it wasn't for being in large part MP focused, I'd totally skip out on shipping a commercial grade level editor. I just don't see a good way to really build and keep alive an online community, without players being able to make and share their own maps though.

Quote
Good luck with this project anyway sir, you've certainly got my attention, it looks ace  Gentleman.

Cheers Smiley
Logged

Gregg Williams
Level 10
*****


Retromite code daemon


View Profile WWW
« Reply #65 on: January 30, 2013, 05:28:42 PM »

Started looking into GWEN gui stuff more, I like how easy so far it is to build. (Can just chuck the files into your source directory.. really the best kind of building imo) The gui skin is pretty minimal, but it might just be more than enough for a map editor.

The biggest issues so far is really I need to write my own renderer, and there seems to be no open/save file dialog support built in. I'm also having to look into unicode since it seems to use that.. *curse you unicode*

Logged

Gregg Williams
Level 10
*****


Retromite code daemon


View Profile WWW
« Reply #66 on: January 31, 2013, 04:45:43 AM »

So managed to integrate gwen without to much trouble. Still need to build a custom renderer, but the basic OpenGL one is at least letting me draw GUIs made from just colored rectangles at the moment.

There also seems to be some bugs in the way input handling is done on some of the controls. Like you can push down on a button, slide the mouse off of it, release the mouse, and the button will still trigger. Thats pretty bad, but the source code was simple enough I could just drop a work around into the code myself fairly quickly.

Not sure how far spread this type of problem is though since I've only tested a few control types.

Anyways I really like how light weight everything else, I'll probably move forward with gwen.
Logged

Gregg Williams
Level 10
*****


Retromite code daemon


View Profile WWW
« Reply #67 on: February 01, 2013, 09:25:51 AM »

Some more progress with the gwen gui library, events now work properly.

Next up map editor window with scroll bars.
Logged

Nektonico
Level 2
**


Time traveller graffiti


View Profile
« Reply #68 on: February 03, 2013, 05:25:28 PM »

Looks delicious.
Logged

He was built by the worlds finest surgeons to drive the fastest car ever designed and nothing can stop him now. ಠ_ృ
“The greatest misfortune is when theory outstrips performance.” - Leonardo DaVinci
Gregg Williams
Level 10
*****


Retromite code daemon


View Profile WWW
« Reply #69 on: February 04, 2013, 01:04:53 PM »

So sadly not a ton of progress, been bogged down with contract work still.

However, I've redone the base isometric engine rendering system.

Before the engine was requiring all tiles to be the same basic width and height.  So for instance previously all tiles where 100x100 even through most ground ones only consumed 50-54 pixels of height. How the engine was setup all those extra transparent pixels were being drawn, and if you suddenly wanted a tile that was 150px high, every tile would need to be 150px high.

Now the engine uses the height data of tiles outputted by our earlier utility tool, to only draw the actual visible part of the tile image, and as such skips a ton of useless alpha pixels. Likewise the render now offsets the tiles on the Y axis to account for the variable rendering heights.

In the end the map rendering can now support arbitrary tile heights on a per tile height basis, and render these tiles in an optimized fashion.

This was an important step, as when I started moving into multiple tile layers, tile height requirements where changing rapidly. Some buildings are way over 200px tall, while shadows for instance are rarely over 50px. I certainly didn't want to be drawing every tile in the game as if it was 100x250 or something, when thats a rare exception.

I also managed to get scrollbars working via gwen, though ran into more bugs in the gwen code in the process of doing so. I was able to fix them, and also report them, but this is a troubling trend of constantly running into broken code.

Logged

Gregg Williams
Level 10
*****


Retromite code daemon


View Profile WWW
« Reply #70 on: February 06, 2013, 09:30:38 AM »

So a bit of progress. I've successfully sliced our first building and have managed to get it rendering on the map correctly.


In the end this building came out to be 11 tiles, across three different layers. (foundation, shadow, and, building/object)

At any rate the next steps is to slice and import some more buildings, come up with an initial map file format, and get units able depth sort in the scene correctly. Really starting to need that level editor though Lips Sealed sadly the gwen gui library is really turning out to be less than ideal.
Logged

oodavid
Level 8
***


Discombobulate!


View Profile WWW
« Reply #71 on: February 06, 2013, 09:49:15 AM »

Looks gorgeous, especially the dinosaur unit on the first post :-)

Just read this on your blog:

Quote
The isometric engine also underwent some heavy re-work so that it can support tiles of varying height, and now renders tiles in a more efficiently. Which will start to pay off in spades no doubt, once all of the other layers like foliage are in for instance.

The next major step will be to get units moving across the map and depth sorting behind and around buildings correctly.

I tackled that one a while back on a similar looking system, pretty straightforward with your perspective:

Z-sort = Y

That is, the lower the tile on the grid, the higher the z-index. You might need to make adjustments if you have ground of different height etc, and naturally you'd need 2 layers: ground and sprites (inc. buildings etc)
Logged


Button up! - Out on Android and iOS

latest release: 13th March 2015
Gregg Williams
Level 10
*****


Retromite code daemon


View Profile WWW
« Reply #72 on: February 06, 2013, 08:28:45 PM »

Looks gorgeous, especially the dinosaur unit on the first post :-)

Cheers Smiley


Quote
I tackled that one a while back on a similar looking system, pretty straightforward with your perspective:

Z-sort = Y

That is, the lower the tile on the grid, the higher the z-index. You might need to make adjustments if you have ground of different height etc, and naturally you'd need 2 layers: ground and sprites (inc. buildings etc)

Yeah it shouldn't be to bad, since I'm going through the pain of splitting all the buildings and such into single tile size pieces. Yet these things still tend to get messy when you throw in partial translucency, and various overlays. Hopefully we will see how it turns out real soon though Smiley
Logged

impulse9
Guest
« Reply #73 on: February 07, 2013, 01:06:03 AM »

Beautiful graphics. Great choice of colors. I'm just stunned at how beautiful it looks. Smiley

I like the concept as well. I'm betting the AI will prove quite challenging to implement.

Good luck. Really wish to see this finished. Smiley
Logged
Gregg Williams
Level 10
*****


Retromite code daemon


View Profile WWW
« Reply #74 on: February 07, 2013, 01:52:53 PM »

Beautiful graphics. Great choice of colors. I'm just stunned at how beautiful it looks. Smiley

Cheers Smiley

Quote
I like the concept as well. I'm betting the AI will prove quite challenging to implement.

Yeah AI will certainly be interesting to do. I'm not sure if we will have AI players for MP games or not, so some of the AI work maybe made easier if its just for the single player campaign. I have a hard time thinking about writing an AI that would work properly in MP with the number of options that will likely exist on a per map or even per game basis. (Points for the match, possibly limited units available, time limit per turn, than customization of said units, etc.)

I like doing AI work though, so we will see how it goes, once we get to that point. I remember one of the first AIs I actually did was just a set of rules and conditionals on paper for playing a specific Magic the Gathering deck some 19 years ago. We use to write logic for decks out, and then play them against each other. Good times.
Logged

toipot
Level 0
**


View Profile
« Reply #75 on: February 07, 2013, 05:32:47 PM »

sadly the gwen gui library is really turning out to be less than ideal.

Sorry to hear that, I feel bad for recommending it now  Sad

It's just full of bugs?
Logged
Gregg Williams
Level 10
*****


Retromite code daemon


View Profile WWW
« Reply #76 on: February 07, 2013, 06:19:40 PM »

sadly the gwen gui library is really turning out to be less than ideal.

Sorry to hear that, I feel bad for recommending it now  Sad

It's just full of bugs?

Yeah that seems to be the case.

I mean I haven't explored it to far, but just every basic thing I try to do, I end up running into bugs.

1. Just building the included OpenGL renderer was impossible, as it ended up relying on ancient libraries like free image, which won't build correctly on recent versions of Mac. So I had to gut a bunch of the code in it, to just get a basic OpenGL (no skin/texture) support renderer working to see how the library would work, before writing my own renderer.

2. Buttons, and everything that is derived from a button/uses a button (like checkboxes, and windows with close buttons, scrollbars, etc) don't really handle mouse input properly. You can click down onto the button, hold the mouse button down, slide completely off the button, release the mouse button, and it still acts as if you clicked it.

So like say you start to click a window close button, but then change your mind and mouse away from it after having clicked it, but before releasing the mouse? Button is still clicked.

Even more oddly, after doing this, you can continue to click the mouse button up and down and it will keep triggering clicks on the button, despite being no where near it. You have to essentially move the mouse, with the mouse button released, for the button to stop thinking its being hovered over.

3. Scrollbars don't have any events for when you click the little arrow buttons at the end, nor can you hold down an arrow and have the bar continue to scroll.

Which when simple things like this don't work properly, I really start to get concerned when I think about much more advanced controls like tree controls, multi-line text boxes, and other junk.

It wouldn't be so bad if I popped over to the github, filled out an issue, and then got a response, maybe even a fix, however its been a week and I've received zero in the way of responses. It seems like the library isn't really maintained or worked on very consistently.

Which in the end is indeed a shame, since I really am liking how simple and lightweight it is. I just don't think I want to start trying to re-write bunches of it, to fix problems.
Logged

Gregg Williams
Level 10
*****


Retromite code daemon


View Profile WWW
« Reply #77 on: February 08, 2013, 08:26:26 AM »

So progress is a bit frozen at the moment. Have reached the point where I really need to start in on a map editor, in order to be able to nicely layout maps to test buildings, depth sorting, and so forth.

Unfortunately, I want the map editor to be cross platform. This means I've spent hours again examining GUI libraries. I'm now looking into MyGUI, which from an hour or two of looking at it, seems to have good potential. Though most of the documentation is in russian... except the bits on OGRE3D's wiki.

I may end up just rolling my own UI widgets at this rate. Something like wxWidgets or QT is also circling back around as a possible fallback, I just really prefer an OpenGL based solution, as I need GUI elements in-game as well.
Logged

toipot
Level 0
**


View Profile
« Reply #78 on: February 08, 2013, 10:32:21 AM »

Which when simple things like this don't work properly, I really start to get concerned when I think about much more advanced controls like tree controls, multi-line text boxes, and other junk.

It wouldn't be so bad if I popped over to the github, filled out an issue, and then got a response, maybe even a fix, however its been a week and I've received zero in the way of responses. It seems like the library isn't really maintained or worked on very consistently.

Which in the end is indeed a shame, since I really am liking how simple and lightweight it is. I just don't think I want to start trying to re-write bunches of it, to fix problems.

Fair enough, it's a killer if there are obvious issues like that lying around and no active maintainer. As you say though it's a shame, it looked to me like a nice solution.

With my projects, homebrew GUI code has been a massive timesink actually, and of course it's a pretty uninspiring thing to spend a lot of time on. In hindsight I'd probably be better off having gone with a library. Although now I feel like i've got a much better understanding of how to put together a good Widget UI system.
Logged
Gregg Williams
Level 10
*****


Retromite code daemon


View Profile WWW
« Reply #79 on: February 09, 2013, 02:56:51 PM »

So I started to take a stab at MyGUI which so far i'm liking despite all the comments and documentation being in russian, and it having taken like 6 hours to get something rendering. Well see how far this relationship lasts.

Kawe has also been busy Smiley



Evolution, options, and some color schemes.


Units in M.I.N.T will have quite a bit of customization options, along with various point costs to balance these.
« Last Edit: February 13, 2013, 06:38:11 AM by Gregg Williams » Logged

Pages: 1 2 3 [4] 5 6 ... 26
Print
Jump to:  

Theme orange-lt created by panic