Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Sizik]

Toady's going to need to make a bigger default tileset sooner or later. The problem, of course, is that under the current system each tile only needs to call up one byte for shape and, what, four bits for color?

Does anybody have any idea how much, if at all, FPS would be affected if the default tileset were 32x32 (thus, four times larger than currently) instead of 16x16?

The big problem with expanding the tile let is that it's designed around ASCII, which only only has 256 values.  This is what allows it to run without graphics on text-only displays (think dfterm).  Adding more tiles breaks the ASCII and would change a LOT of things.  I immagine there is an aweful lot of code where things are defined as char type that would have to be changed to ints.  Code could rely on some of the special things you can do with chars.  I have to imagine that it would be a huge amount of work to change something so fundamental.  It also has a lot of potential to reintroduce bugs all over the codebase.  I agree that it would probably be the single best thing feature in recent memory, but the programmer in me cringes at the thought if doing it.

DF is also designed around terminal displays, which are only 80x25 characters in size. Yet, we still have the ability to resize the window (on the fly, even) to however big we please.

[hermes]

Yeah.  Graphical footprints are an interesting idea, but I'm not feeling it.  Maybe I'm unimaginitive, but I can only think of one purpose for this feature: establishing tension/suspense while tracking an unknown creature, a la countless horror movies etc.  Textual description could establish tension too, and without the player/character confusion that Askot mentioned.

Also, it just seems perverse that, of all things, footprints could be the first use of ASCII art (as opposed to procedural portraits etc).

Agreed that text descriptions could be just as good (if not better since I prefer written works), but the more I think about Brad's post, the more it makes sense to start procedural artwork with footprints.  It's a 2D medium made in a single material (soil), with a limited number of general cases, where the only variations from a "normal" print are (off the top of my head) size and the absence of toes/limbs.... the perfect place to start experimenting with ASCII art viability.  Faces and bodies are, in comparison, complete nightmares.

That is, of course, if this is how the images are created.

If they were procedural, and if the player was told in notes what kind of beast left the tracks, then the art could convey, besides the atmosphere of the hunt, some ancillary information like the size of the animal, if it's injured (blood spatters?), its dominant side, etc.  Lots of 'ifs' there, have to wait and see...

[BradUffner]

DF is also designed around terminal displays, which are only 80x25 characters in size. Yet, we still have the ability to resize the window (on the fly, even) to however big we please.

There is no fixed defined size for a terminal window.  I've seen 40x25, 40x24, and 80x24  as default sizes on some systems.  You are just thinking about the rendering system for the display, which is it's own compartmentalized system that is actually maintained by Baughn, not Toady.  Adding tiles would go far deeper than that, in to the core of the program and how things are fundamentally stored. ASCII by definition is 128 (or 256 if using extended) chars.  The "char" type in c++ is one byte, which holds up to 256 values which exactly matches ASCII.  It can't be extended. Adding more tiles would mean a fundamental shift away from using "char" and ASCII to store tile data.  I'm not saying it's impossible to move away from chars and ASCII. I'm just saying that it's a pretty HUGE task that could take a very long time and introduce a LOT of bugs.  I actually think it would be a great thing to do, but I wouldn't want to be the programmer doing it.

[Footkerchief]

It's a 2D medium made in a single material (soil), with a limited number of general cases, where the only variations from a "normal" print are (off the top of my head) size and the absence of toes/limbs.... the perfect place to start experimenting with ASCII art viability.  Faces and bodies are, in comparison, complete nightmares.

I was thinking more of portraits just for the humanoid races -- hardcoding the basic idea of a humanoid face, and using the existing facial feature attributes (how wide-set are your eyes, etc) to individualize the portraits.  Portraits for all species would be significantly harder, yes.

That is, of course, if this is how the images are created.

I strongly doubt it -- Toady has been emphasizing that this is in the hypothetical stage.

[eux0r]

ok, wait, what... what? im confused now..
does this mean we walk around the forest, and then this window pops up with some ascii-art-image and it says: "you found some footprints!"?
if this is the case, then im very unhappy:

-i hate windows/screens/images popping up just like that. they fuel my rage to the heat of a thousand stars! HEAR ME ROAR! the trade-discussions already annoy the hell out of me....

-we are supposed to look at the footprint and see: "oh, its missing its second toe, that must be the hamster-man i was fighting with just now!" my concern here is the same as what others already mentioned, if i as a player can see that, then where does my character come in play here? the same if i cant see the details im supposed to see while my character is some über-awesome tracker. we should get a procedurally generated text describing what our character sees, with the level of detail depending on the characters abilities/knowledge. just like we get to see more or less random/wrongly estimated prices when trading with a low-skill broker in place. im not totally against the images, i mean its awesome if we can learn some animal prints by playing df  just like were learning something about minerals, but its enough to put those images below the text, with the images generated with no more detail than the description.
in other words, if we have a skilled char tracking a deer, we should see an image of a deers footprint below the description of the footprint. when we track a deer with a low-skill char then we should only get a description talking of some kind of unidentifiable track and no image/an unidentifiable image.
all that in the normal description of the tile the prints are on with an asterisk on the ground animating the player to look at the tile if he wants to. only if we look at them. please, no more popups, give the player the choice to ignore the footprints without asking "do you want to look at the ground now?"

tl;dr: pretty much repeating complaints from 2 pages ago and ragin a bit

edit: i realized, no one here could possibly know how it actually will look like, apart from toady: therefore adding colour

[CLA]

my concern here is the same as what others already mentioned, if i as a player can see that, then where does my character come in play here?

Again, I think it's not that big of a problem. Combat already works like that. You, the player, make the decisions where to strike, when to flee, etc, and the character's skills determine how successful that will be.
To quote myself:

For sneaking this would mean, the player decides where to sneak, when to stop, etc, but sneaking/perception skill influence how much noise the character makes, how well and acute he can stop and, say, remain motionless, how accurate he can see where an enemy looks, who is alerted, etc.

For tracking, whatever perception/tracking ability he has, decides finding tracks in the first place, and possibly helping player memory with additional information like how old the track is, in which direction it goes, etc.

I really don't think this will be a problem. However, whether it should be displayed graphically or with text is worth to argue about I think.
Personally, I think it's a great addition.

[bombzero]

I kind of like eux0r's idea, if the images are procedurally generated that is, otherwise it would be basically impossible.

however I don't think it will be an obtrusive popup or anything like that.

[CLA]

Procedurally generated images based on information in the raws

Yeah, that would be really nice, especially if it can serve as framework for faces or even engravings at some point.

[Putnam]

does this mean we walk around the forest, and then this window pops up with some ascii-art-image and it says: "you found some footprints!"?

It'll probably be something like "closely (L)ook at the tile you're standing on" for tracks, with shift+l. The way you describe is extremely gamey and just as stupid as you think it is

[DG]

Some people had too many negative cornflakes for breakfast. I'm obviously more of a blind fanboy than most because the graphical footprints seem like a cool idea to me and I'm not worried that their inclusion will add X years to DF development.

Tracking will likely be a mode that you choose to enter. You will only see footprints when you want to, and you can probably filter the type that you're interested in so that you aren't spammed by chicken prints when chasing a colossus.

[EmeraldWind]

I think the footprint images could be used to denote groups of footprints for similar species.

IE: You get a text paragraph saying "the footprint is medium sized and has five toes". You look some more and find "the footprint is medium sized and has five toes."

Problem is one is humanoid and the other is ape-ish.

If you show a generic humanoid foot or a generic prime ape foot... the player will be like "okay I'm following that Chimp that stole my food; so I'll ignore anything that doesn't have the prime ape foot print". Alternatively if they are following a dwarf, they will ignore anything but humanoid prints. They might have to study the description more to tell specifics (ie, dwarf prints are medium and broad; humans are medium), but the generic foot prints could act as a visual filter. The player sees the image and knows whether or not they could just skip the description. The description can go into more detail for missing toes and other info that might help the player.

That being the case, the images could be limited to groups that make sense. IE: bipedal humanoid (human, elf, dwarf), padded quadruped (dog, cat, wolf, tiger), rodent trail (rat, mouse), drag trail (snake, alligator, zombie whale), talons (chickens, owls, turkeys), hoofed quadruped (boar, goat, horse), bipedal humanoid with shoes (human, elf, dwarf).

A couple more than that as well. Some special creatures could have unique foot prints for flavor (ie: dragons), but every single animal species shouldn't require a unique footprint.

I could see the value of using the graphic as a quick and easy way to determine whether or not you want to read the description in depth thus saving you a couple seconds. That may not seem like a good reason, but imagine someone lead you through an area trafficed by tons of creatures. You'll love having a simple indicator to help you out.

[Askot Bokbondeler]

there's also the thing that tracks are rarelly footprints, and even when they are, they're rarely perfectly delineated footprints, more like an amorphous hole in the ground from which you can get a rough size, and maybe a width\length ratio. it is easier to identify if an animal is limping(and more relevant) than to count its toes

[Quatch]

With the new skills coming, and the increasing relevance of the existing skills, will the number of points available in 'demi-god' increase to make that moniker meaningful?

[Cruxador]

With the new skills coming, and the increasing relevance of the existing skills, will the number of points available in 'demi-god' increase to make that moniker meaningful?

More than that, I think it would be good to have a "limitless" mode where you could build your character however you pleased and not worry about points at all. But this is more suggestion than question.

[tfaal]

Yeah, it's kind of silly to have any limitations in such a sandboxy game.