Ultimate Game Idea!!!

[TradeMark]

The symbols are all distinct from one another. Attaching meaning to the symbol is a simple single relation, psychologically. It is actually a much less intensive process to connect the basic ASCII characters to content than it is to deconstruct and assign complex images. This is not debatable, this is simple psychology, specifically visual orientation and relational processing.

maybe if youve played those games all your life, to me its just gibberish and its really hard to remember which letter stands for which thing, especially for letters like á and à ... looks almost same, 1 pixel difference...

[Neddie]

Virtually nothing uses accents in that manner, and technically, that isn't a one pixel difference. It is a two pixel difference, the lack of one from the former, the addition of another.

o

What is "o"?

o is a circle, denoting the letter "o" (oh) or the number zero in basic usage, representing perhaps a tunnel, a hole, a head, goatse, the wall of a city... but that identification occurs subconsciously and contextually. It is easy, and you do it all the time.

[JohannesH]

o

Looks like an orc.

[knorke]

What is "o"?

[TradeMark]

What is "o"?

o is a circle, denoting the letter "o" (oh) or the number zero in basic usage, representing perhaps a tunnel, a hole, a head, goatse, the wall of a city... but that identification occurs subconsciously and contextually. It is easy, and you do it all the time.

yeah... you need to look at the letters and think at them, instead of just seeing everything at once and knowing where is everything because they look exactly like what they represent. this is also why we use 3d models instead of ascii chars to play BA...

you cant see the whole screen with one glance in ascii graphics... you need to focus in small area to see what letters are there, in other words, you need to read it... if it was like BA, you dont need to read anthing, you just see where is your stuff without moving your eye focusing point anywhere from middle of the screen. this is why graphics pwn ascii.

[Teutooni]

Modeling in text based games is never just single characters, all objects have descriptions of varying length. Depending on the description, most information would be hard to impossible to model accurately in 2D, not to mention 3D or 4D (animated). Simple things sure would be more intuitive graphically, but simplicity is not what most text-based games are about.

[Neddie]

You're missing the point. Your mind does this subconscious processing based on context automatically, with both complex and simple images, but complex images take more cognitive load to process and associate because they have more elements. How is a simpler object going to be harder to process. It isn't. If you're sitting there consciously trying to think about it for more than the first few times you see it, no wonder you can't process it effectively. I'm surprised you can adapt to anything.

[TradeMark]

well, my mind likes BA graphics more than ascii graphics, i think youre missing the point here.

i cant remember something that doesnt represent logically what it really is... if it makes no sense, i cant remeber it. unless i really try to remember... maybe im just retarded then. special in other words <33

[SinbadEV]

Personally I agree on both counts... since a k is easier to process as kobold then a pixely image of a kobold I think that makes sense... though it only has that meaning because we have been told it has that meaning... so what I'm thinking is something like this,,, you will have a list of creatures/object visible to translate... mousing over these entries would highlight the associated creature



Then the user could either click on the actual character and then type a shortcut action (perhaps there would be a default actrion, or they could type 1A to attack the dragon or they could type M (which would bring up a pup up menu of spells) and then type the associated letter and then type the number of the target... or they could click attack on the creature they wanted to attack or they could choose the other menu, select attack and then click the target... or something I decide later

anyhoo... here are some more examples of how things might look

3D with characters as textures:


Isometric with characters as sprites...


In these cases it might make sense in the long-run to provide user support for advanced graphics to replace these ascii textures... (3D models or sprites) and then servers would make these graphics available as an optional download.

[SwiftSpear]

You do not just "render some model" and it's magically iso in the right aspect ratio and lighting without any problems.

Plus, if you go that way... you're building the entire art content for a game... in 3D. Talk about a total waste of time.

No, because you don't have to solve nearly any of the mathematical problems that come with working with a 3D universe as soon as you have the basis completed.

[SwiftSpear]

What is "o"?

o is a circle, denoting the letter "o" (oh) or the number zero in basic usage, representing perhaps a tunnel, a hole, a head, goatse, the wall of a city... but that identification occurs subconsciously and contextually. It is easy, and you do it all the time.

yeah... you need to look at the letters and think at them, instead of just seeing everything at once and knowing where is everything because they look exactly like what they represent. this is also why we use 3d models instead of ascii chars to play BA...

you cant see the whole screen with one glance in ascii graphics... you need to focus in small area to see what letters are there, in other words, you need to read it... if it was like BA, you dont need to read anthing, you just see where is your stuff without moving your eye focusing point anywhere from middle of the screen. this is why graphics pwn ascii.

it depends how busy the screen is... Dwarf fortress is pretty easy to manage, but there's the occasional thing. Still, I stand by my original point. If you have 20,000 items in your game, you're wasting time creating graphics manually. If you can't come up with a procedural way to do it, your game almost isn't worth doing any more, at least that is to say a one man team will spend more time doing art than actual game creation. Animation is even worse.

[SwiftSpear]

well, my mind likes BA graphics more than ascii graphics, i think youre missing the point here.

i cant remember something that doesnt represent logically what it really is... if it makes no sense, i cant remeber it. unless i really try to remember... maybe im just retarded then. special in other words <33

Because with BA graphics you get information about how the object works from what it looks like most of the time. If an object looks big and clunky and has a giant gun, it's probably a slow powerful tank. If the object looks sleek and agile and has a little lazer it's probably a scout or fast attack vehical or something. Most importantly, we get information about what direction the parts of the object are facing. which is fundamental to any type of real time tactical combat situation that we're supposed to be micromanaging.

Alot of the customizations and mods break this pretty throughly, but ideally you are gaining multiple points of information just by looking at the object on screen.

In my DF type game, all I have a "k". And that's a kobalt. I can't see what weapon he's holding, what kind of clothes he's wearing, if he's angry or happy or anything, what action he's preforming (there are ways around that)... However, if there are maby 400 different types of weapons, 200 different types of clothes, 12 actions, 10 emotions... there are 9600000 possible combinations. If I'm working with a 20x20 pixle sprite I'm probably not going to be able to set up an easy procedural way of doing that, and it's impossible for all of those combinations to have a unique graphic designed by me.

For a one man game development team, why waste the time? Just have the player able to click the object and learn all those things from a diaglog box instead. They will know pretty quick that "k" = kobalt, and if a kobalt in that area isn't normal they had better check up on him and see what he's doing.

[KaiserJ]

You do not just "render some model" and it's magically iso in the right aspect ratio and lighting without any problems.

Plus, if you go that way... you're building the entire art content for a game... in 3D. Talk about a total waste of time.

well, if he were to make all of the models in 3d, it probably would be.

however, if he were to grab some IP-free models, and start constructing graphics out of 2d renders from them, then he might well end up saving a ton of time (not to mention that programming a 2d engine for a game requires a lot less effort than a 3d engine)

and no, you dont just "render some model" and expect it to look good; you set up some basic lighting and camera depth information, and render so that your output provides you with a double-sized sprite image, and then reduce it using photoshoop (i could do this in about 3 seconds in fireworks using guides)

in fact i've seen a "2d sprite generator" plugin for max that will automate the creation of images using different perspectives and keyframed animation...

done foolishly, yes, lots of time can be wasted, but i think if he had a collection of models to make the images from, it would end up being a lot faster than drawing everything by hand, especially if he wants things like walk animations etc.

[Pxtl]

Considering the amount of free MD2 content out there, and how simple it is to get an MD2 model onto the screen, I think the easiest approach would be a simple tile-based 3D game from a 3/4 perspective (not isometric) using MD2 and maybe MD3 models.

You could even simply code your tiles as MD2 models if you didn't care about performance - just make each one a cube and swap out the textures for the different tiles.

[knorke]

Isometric with characters as sprites...

why make the terrain isometric if objects are just letters? Does go well together imo and makes things just more complicated to program.
for example to program on which tile the mouse pointer is positioned. Or draw the objects in a way they obstruct each other correctly. (would be easier if the objects graphics are so small they do not go over the borders of the tile they are on)
I have no idea about your programming experience but for a beginner a 2d ascii view like nethack is easier. You said you just want to make a quick prototype. So keep everything that isn't related to your ideas as simple as possible.

[Argh]

You can do ASCII worlds in Spring without breaking a sweat. It takes very little effort:

Write unitDef. Probably for a project like this, most of them would be virtual clones of one another- the real values would be customParams, since you'd be ignoring most of Spring's RTS side... which will work just fine.

Build some Lua that reads the unit's ASCII value from a customParam, and draws it using gl.Text. Or you could "get fancy" and use the new text system, whereupon you can use <gasp> multiple fonts to convey more information to players at a glance.

Blank BOS.

One S3O, consisting of a point with a hitsphere. The hitsphere's just for visibility, though, as you won't be using real projectiles, etc.

Now write the gamecode and UI with LuaRules. Obviously everything would use Lua to control movement; there would be zero need to use a lot of Spring... but, and this is the big "but"... if you went graphical later, voila, you're ready, and you're not performing a massive graft on a program that's utterly not designed for the task.

[SinbadEV]

Stuff

I'm not making it in Spring... sorry.


Also:

Here is what the "overlap" would look like in an isometric view as I have it in mind.



There is a small amount of overlap, but not enough to say that it would interfere with a user's ability to select the correct item... I'll grant that the mouse-capture is significantly more difficult, but this problem really only needs to be solved once... while I have not finished programming "anything"... I have programmed quite a bit and do so professionally now so working out a bit of complex pixel math shouldn't be too hard even though I'm still a beginner.

I think the visual advantages to an isometric view outweigh the added difficulty.

[SinbadEV]

I found an ANSI C SVG rendering engine... how hard is it to use C code in a C# program?

[Neddie]

SVG? FFFFF-