Animated Poly

[nesnl]

Is there anything special to this? I was unable to figure out how to use it when I messed around with it briefly? Does this allow you to create animated images within graphics of the level? Would you be able to make something moving and also have it so the 'hull' is also moving along with it? If so, is the object limited to being in one place or can you make something animated that also can move throughout the game level?

Any help or guidance on how to use this function would be greatly appreciated.

Also on a side note, what programs do you use to create the graphics for your current levels? Any tips, tricks, or advice in creating a level in a more streamline way as far as creating the graphics goes?

[lamster]

For now the Animated Poly editor is only for my own internal purposes -- there is no support for moving objects or animated graphics in the current map editor. I'd like to add support for this eventually, but it will be part of a larger feature expansion if and when that happens. Unlikely to happen for several months.

For the official levels in the game right now everything was drawn in Photoshop and then each "View" (Game, Background, etc) is sliced up (using Photoshop's slice tool), the slices are exported to the map's image directory, and placed in the appropriate positions using Place Sprite (Ctrl+E).

In a slightly more ideal world, we'd have a tool that analyzes your Photoshop PSD, slices it up automatically to remove transparent blocks, guesses appropriate collisions bounds based on alpha, and imports the slices as objects into the editor. I may write this tool when I get a chance to focus on the editor, but for now you must slice by hand.

For developing new maps, my suggested workflow:
(Optional) Sketch your concept on paper, capture the overall intended aesthetic for the map and decide how you're going to deal with various logical/graphical hurdles (for example, in lost city we use destroyed sections of the buildings for floating tunnels)
1) Draw the map in the editor using simple opaque polygons. Test the map. Improve it. Test some more -- get all the collision bounds and layout perfect. It is much easier to make changes and improve gameplay without worrying about graphics.
2 -- if you're an artist) Zoom the editor out so that the entire level is visible and take a screenshot of your level layout. Paste this image (and resize it to original pixel dimensions) in your favorite graphics program and use this as a blueprint over which you'll paint beautiful, final map art. Slice up the final image and import the sprites into the editor, placing them over your blueprint and adjusting collision bounds as apprpriate.
2 -- if you're not an artist) Grab a few tileable textures off the internet (one for grass, one for metal, whatever you need) and assign those to your polygons as appropriate. This is really easy and can yield decent results. If the map becomes popular, you can always convince an artist to paint it later.

[PodBot]

I use photoshop for any grafic work i need done, however as time developed in my early alt-editor carrier i found the built in poly tool became more and more useful, i personaly try to build my maps with the intention to get a map balanced i meanings of "apperance" vs "size" (in kb). Fair enough most people dont have any restrictions with either bandwith or cpu/gfx resources. However i find the challange in it. making it look good without any unnecasary textures.

What im trying to say is, its easier to (as lams said) making a "painting" of the map and then integrating it all at once. Rather than "importing" bits and pieces.
9 out of 10 attempts to get a cool area on the map with fair advantages for both teams (or players) when using pre-made textures (rocks and stuff) i had to tab out to photoshop, edit the texture, export it, and refresh the editor and finaly move it or resize it to test it in game again. when you do this to many times it just calls out for errors to be made heh, keep it alt-editor as long as possible, you´ll save alot of hours doing it.

I made a bunch of 32/32 px sprites (textures) and use the "place texture" function in the editor. I will put up a texture pack later this month with a bunch of themes and stuff. With it a small tutorial in swedish if we ever get more of those heh.

[nesnl]

I am having another problem that is unrelated to the topic of this post (I apologize if this gets posted twice as I just typed this all out and when I clicked post it said I wasn't logged in and then deleted the message I typed!) Anyway, so I am using a Mac to make my maps. We had talked before and I know you told me that the test run function wasn't working on the mac version for the time being. So that left me with the only option of exporting my map, launching the game, and creating a server each time I want to test the map. That is fine except that sometimes when I do it it goes through the same process (preparing map resources, acquiring remote ip) and then it just comes back and says "No response from Master Server." Is that something I am doing wrong on my end or no?

Also I was wondering if there was any benefit to slicing up images like you had talked about. I am currently working on a map that has a continuous graphic for the upper and lower limits of the map. I saw no need to slice it up so I left it as it is (over 3000 pixels wide). Does this affect performance? Or are there any other benefits to slicing into smaller pieces? I ask because I notice on the map tbd_cave that even though the outer border is continuous, it is still sliced into many smaller pieces.

Again, thanks for all your help and quick replies!

Update: So I have tried a few different maps and it looks like when I load some maps it is fine and then if I go to load other it gives me that "no response from master server." so I am not sure what's going on there.

[protest boy]

Originally Posted by nesnl

Also I was wondering if there was any benefit to slicing up images like you had talked about. I am currently working on a map that has a continuous graphic for the upper and lower limits of the map. I saw no need to slice it up so I left it as it is (over 3000 pixels wide). Does this affect performance? Or are there any other benefits to slicing into smaller pieces? I ask because I notice on the map tbd_cave that even though the outer border is continuous, it is still sliced into many smaller pieces.

I'd like an answer to this as well. Also, is there any reason to slice/not slice up a background image?

[lamster]

Oops didn't notice the last post. Regarding slicing: the reason is storage and runtime performance. An enormous image (say 3000 tall where the middle 2800 pixels are transparent) will eat a tremendous amount of video memory, take more space to store on disk, and as far as the video card is concerned, every pixel of your rectangular image (whether completely transparent or solid blue, for example) that is within the camera's view has to be drawn to the screen. For users with Nvidia or ATI graphics cards, this may not be noticeable, but computers with integrated graphics will have trouble filling the same pixels multiple times per render (eg drawing a bunch of overlapping transparent pixels each frame). Hence by slicing away transparent regions you increase the performance for low-end cards.

I know it's really annoying to slice things this way. The good news is that the engine will eventually be smart enough to slice away these transparent pixels for you behind the scenes. The bad news is I haven't had a chance to implement that yet so for now you must slice by hand for optimal performance. If you're not experiencing any slowdowns just worry about the really huge/easy cases -- if your map becomes popular you can optimize it more later.

[lamster]

Also, is there any reason to slice/not slice up a background image?

If there are no transparent pixels, there is usually no benefit to slicing it up. For example, a full-level hand-drawn background is perfectly fine as a single image. If there are repeating chunks (eg a gradient fill that repeats across the whole map) and only a few dynamic chunks (eg a thin border around the edges) it may be worthwhile to chop these up separately -- but generally speaking, slicing is only useful for removing transparent pixels.