Print

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

[Manveru Taurënér]

As someone who plays SS13, sprites are not exactly representative of the true size of something.

If you want everything scaled true to size it'd be hard to see (Somethings would be super tiny, others super large multi tile monstrosities), sizes would be all over the place and it'd no doubt be a massive pain to do that. Obviously the chicken and goose aren't as big as the horse for example.

I think (or hope) that everyone is well aware of that by now, but still, suspension of disbelief only works so far. And everyone has a different threshold for how far things can be scaled one direction or the other before their brain automatically rejects it. This is also very dependant on how familiar one is with the creature in question and their true size as well as whatever is displayed next to it (muskox are actually much smaller than they seem for instance and are probably okay like that, whereas I'd agree the cat could probably use being shrunken down a tad).

[Mike Mayday]

Blazerules pretty much describes my approach to the problem.
For example, the big cat bothers you but the dog being over 3/4 the size of a cow is ok? Well - there's gonna be 100 people for whom the dog is the bigger problem. And then to satisfy everyone, we'd have to make all relative sizes accurate. And then we'd have elephants that take up about 12 tiles...

I won't be surprised if Meph prepares a mod that makes the relative sizes more accurate, but I chose clearer representation of the creature and more interesting details in favour of accurate sizes, because I believe the latter is a lost battle already, what with every possible creature occupying just one tile right now. If that ever changes? I'm all for it.

I'm open to a few small corrections, I'll try to do the cat somewhere down the line, but probably only after we have all other graphics ready.

[mko]

If you want everything scaled true to size it'd be hard to see (Somethings would be super tiny, others super large multi tile monstrosities), sizes would be all over the place and it'd no doubt be a massive pain to do that. Obviously the chicken and goose aren't as big as the horse for example.

I am pretty sure that noone wants this. See size of a dragon and what would happen with linear scaling (even if dragon can take a bit more than a single tile).

So I personally think it's perfectly fine but I'm also used to all things being in 32x32 pixels as a max limit regardless of their real size. What matters most is that you can make out what the thing is, even just having a basic "small things are 10x10, medium are 32x32 large are..." type thing would still be a pain.

It is probably also hard, but cat, dog, goose are really weirdly sized in this example. Note that it may turn out that smaller animals are not much better - we see higher static images, in actual gameplay it will work differently.

But lion-sized housecat seems to be unfortunate.

And then to satisfy everyone, we'd have to make all relative sizes accurate. And then we'd have elephants that take up about 12 tiles...

Accurate would not work. But ensuring that things drastically smaller than something else have at least a bit smaller image may cover nearly all real complaints.

[Manveru Taurënér]

Blazerules pretty much describes my approach to the problem.
For example, the big cat bothers you but the dog being over 3/4 the size of a cow is ok? Well - there's gonna be 100 people for whom the dog is the bigger problem. And then to satisfy everyone, we'd have to make all relative sizes accurate. And then we'd have elephants that take up about 12 tiles...

I won't be surprised if Meph prepares a mod that makes the relative sizes more accurate, but I chose clearer representation of the creature and more interesting details in favour of accurate sizes, because I believe the latter is a lost battle already, what with every possible creature occupying just one tile right now. If that ever changes? I'm all for it.

I'm open to a few small corrections, I'll try to do the cat somewhere down the line, but probably only after we have all other graphics ready.

While I agree overall, and think you've done a really good job with sizes so far despite my occasional nitpicking, I also think it's a good idea to (if possible without losing essential identifiers) slightly adjust those things that stand out the most as straying from the loose relative size spectrum everything is haphazardly crammed into by the game limitations. Like, the important bit I'd say is rather that things (where possible, as I know it's often harder in practice than theory) stay close to their relative size order than the actual relative size, so as long as the cat looks and feels smaller (even if just a tiny bit) than most of the the things it should be smaller than then it still feels okay, regardless of if the scales are way off still on the whole. It probably depends a lot on what animal it is too, dogs for instance have wildly varying sizes IRL so are probably easier to instinctively justify being off one way or the other (same with horses or even cows) unlike a cat which is generally pretty small, or something that's generally perceived as really big.

[mko]

Good point about dog breeds.

---

And for me cat vs pig issue was striking because I am familiar with this animals.

Goose size issue that someone notice was not so clear for me, probably because I had limited contact with this animals.

And I am unlikely to be bothered by gorlaks, even if their size will be wildly misrepresenting RAW value.

[MoltenIdol]

I actually have to agree on specifically the cat being a tad too big. I get the whole limitations thing, and you are doing such a good job with it. But the cat has been bothering me from the start, and since somebody else just voiced their concern, i had to do it as well. Not trying to be nitpicky and annoying, its just a thought. The slightly abstract size works better for the dog an the others, the cat Should just be like, 1 pixel slimmer in each direction, if that makes sense.

[MoltenIdol]

Well, I don't have a tiger mane atm, but I can give you the formerly mightiest of all creatures:

BEWARE THE GIANT SPONGE!

Now, it feels important for me to mention that I'm not talking about animations and redrawing of sprites, but more like, in the case of the zombie breathing example, taking the sprite, cutting it up by chest, legs and head, and disjointing them from each other, designating them as seperate parts, to be able to move them a bit around and create "animations" with one sprite. You could then generalize this method trough all sprites , to make a general idle animation

This is still an amazing amount of work. There are close to 3000 creatures sprites. Take them apart into pieces, lets say legs, arms, torso, head, means 6x3000.

Please don't get your hopes up on any fancy animations.

EDIT: Toady is really fast. I'm impressed.

With that edit, you kinda got my hopes up a bit, what does that mean ?

Never mind the cutting of sprites, but more regarding something like a simple system that could disjoint a sprite from JUST its tile, in order to be able to place it wherever on a tile you want, and make a smooth slide between tiles, for boppy-walking.
And Maybe even turning sprites to lay down? Or mirror them to face east and west according to walking/attacking/interaction direction, and bopping towards targetted "element"? You read all that. Im just trying to convey the concept as clearly as possible. I of course would Know Nothing about the time Even just that would take.

Feel free to school me

Never mind the cutting up of  the sprite For the animations, Thats only dreaming on my part if thats a Lot of work, dont bother wasting time thinking on it<3

The suggestion i made thats most important to me, is definitely the walking bopping up and down while sliding thingy. A Lot could be done with only that as a system.

Ill stop wasting more page space on this now

Ps: Where are those sponges supposed to.. grow i guess?
Nicely done tho.

[mko]

Ps: Where are those sponges supposed to.. grow i guess?

See DF Wiki - http://dwarffortresswiki.org/index.php/DF2014:Giant_sponge - it has location info and gems such as

the combat mechanics of v0.40 have made it invulnerable no longer, due to the mechanics of pulping. Due to the sponge's weak material properties (even weaker than flesh), pulping sponge tissue with blunt weapons is surprisingly not too difficult. However, their large size means they can shatter bones, articulations, bruise organs or even kill a dwarf via headshot using their default push attack. They may also themselves pulp body parts beyond recognition, as the push attack is also a blunt attack.

(...)
If some hapless dwarf appears near their water, giant sponges may feel suddenly threatened and charge (!) the hapless dwarf and engage in combat - the only occasion when they move at all. They may also become enraged or unconscious or feel pain, as utterly improbable as that sounds.

[MoltenIdol]

Ps: Where are those sponges supposed to.. grow i guess?

See DF Wiki - http://dwarffortresswiki.org/index.php/DF2014:Giant_sponge - it has location info and gems such as

the combat mechanics of v0.40 have made it invulnerable no longer, due to the mechanics of pulping. Due to the sponge's weak material properties (even weaker than flesh), pulping sponge tissue with blunt weapons is surprisingly not too difficult. However, their large size means they can shatter bones, articulations, bruise organs or even kill a dwarf via headshot using their default push attack. They may also themselves pulp body parts beyond recognition, as the push attack is also a blunt attack.

(...)
If some hapless dwarf appears near their water, giant sponges may feel suddenly threatened and charge (!) the hapless dwarf and engage in combat - the only occasion when they move at all. They may also become enraged or unconscious or feel pain, as utterly improbable as that sounds.

"Without a nervous system, the only thing it can feel is ANGER"

Hehe

Theres a picture at the bottom of the wikipage, Saying the smoke represents the ANGER. Are giant sponges gonna emmit smoke particles? Thqt would be cool, the more "animated/moving" tiles there are (like waves and flowers blooming), the better. (I mean, this is the graphical release, right ;P? (No pressure hehe))

And, this is not really that relevant, but I always wished for the civilized/lesser civilized races to be able to smoke stuff, and also adventurers, emitting smoke particles, maybe with a pipe in hand, it could look really good with the new graphical release, making the smoke particles translucent, different sizes, and be as an overlay on a tile, instead of only displaying smoke on a tile, where theres also a dwarf smoking.
Would Also be cool to grow sum Good ol pipeweed.

[Cruxador]

Just wanted to update you guys on the ramp situation. Weird corner/edge cases looking much better now.

This looks great, but now I'm imagining how much work it will be to make water look good flowing down all these crazy procedural angles.

No new animations are planned for now; those were all mock-ups showing Tarn how it would look ingame. We first have to do everything else; animations would be possible additions we might tack on at the end. That being said, if he adds a bit of code support for the community to create 2-3 frame animations using ALT_TILES for more objects/creatures, I'll be the first to make you guys a set with that.

Animations are a big effort relative to the payout, but making sprites slide from one tile to another should be pretty simple, it only requires sprites to be displaced dynamically from the center of the tile, and the only big requirement for that is that they be able to overlap the edges of the tile which it seems is already in the cards for big creatures. Then, it needs to be set up different for each action that uses it, but unlike proper animations with frames, it doesn't need to be done differently for each creature and equipment has no effect on it. Of course, it's Toady working on it and not you guys, so I know you can't definitively say anything, but I would expect that he can do it pretty easily, and would be willing to. Aside from stuff that's already all but confirmed, I think this would have by far the biggest impact on the look of the game relative to work put in. In fact, I think the only things more impactful are having different dimensions on the text and map tileset, seeing the background tiles behind creatures and items, and having sprites accurately represent equipment, wounds, and distinctive characteristics, the latter of which is a huge effort from everyone involved. Even the multi-level view is, I think, slightly less of a big deal aesthetically. It just feels bigger since we've already become accustomed to it in TWBT. And it will, of course, be handy to new players who aren't accustomed to z-levels.

Blazerules pretty much describes my approach to the problem.
For example, the big cat bothers you but the dog being over 3/4 the size of a cow is ok? Well - there's gonna be 100 people for whom the dog is the bigger problem. And then to satisfy everyone, we'd have to make all relative sizes accurate. And then we'd have elephants that take up about 12 tiles...

I won't be surprised if Meph prepares a mod that makes the relative sizes more accurate, but I chose clearer representation of the creature and more interesting details in favour of accurate sizes, because I believe the latter is a lost battle already, what with every possible creature occupying just one tile right now. If that ever changes? I'm all for it.

I'm open to a few small corrections, I'll try to do the cat somewhere down the line, but probably only after we have all other graphics ready.

I think part of the problem is that the smaller end of the size range is spread a lot less carefully than the larger range. The coney is pretty much exactly the right size compares to the cavy, and the duck and chicken also have pretty much real life size, while the goose is only slightly smaller than it realistic is compared to them. And that puts the goose at a size comparable to somewhat bigger stuff. I'm not saying that you should go back and redo all those or anything, and I'm also not sure that's entirely a bad thing (might as well have more sprites use more pixels, and thus be more recognizable, compared to the alternate possibility) but I reckon that's the methodological thing which lead to it.

But yeah, no matter how you slice it, the cat is too big and that happy little piglet sprite should probably be accompanied by a full grown boar/sow sprite. Although it's weird, since even though it's drawn as a piglet, the size isn't wrong for an adult if compared to something like a horse.

[Rekov]

I think the area marked in magenta here is supposed to be a single flat side, but the shading makes the 3 marked triangles look like they are all angled differently. The transition between the two tiles in the orange box don't transition seemlessly like the other tiles.

Nope, all three magenta triangles are at a different angle (the top one is supposed to be floor). Due to my mistake, the sprite is missing some shadow indicating this (I forgot to enable a layer).
The orange rectangle- there's an edge there, hence the lack of smooth transition.

I'm not sure you have the best approach for that magenta tile. You currently seem to favour an approach where ramps take up the whole tile they occupy. This makes sense when you plan ahead to multi z-level view because it allows slopes to continue smoothly across z-levels. For lack of better terms, lets call these the sharp corner (left) and the beveled corner (right)



In both the magenta and orange highlighted tiles, you have chosen a 'beveled' solution when a 'sharp' solution exists.

Here is my understanding of the magenta tile as you've described it. Forgive me if I misunderstood. The solution on the right, which is as jecowa describes it fits better into continuous slopes across z-levels, and is more consistent with the way you've chosen to do corners elsewhere.



In the case of the orange tile, I don't necessarily see any potential problems across z-levels. I think the current configuration actually helps to break up some of the regularity in the ramp system. However if you really wanted to be consistent, here would be the geometry.



EDIT:
Here is a potential third solution for the magenta tile. A 'best of both worlds' where the slope across z-levels doesn't have that flat platform, but it also slightly breaks up the regularity of the ramps:

[Meph]

I think Vettlingr had a nice solution for that. Mike seems happy about changing it again. ^^

[Mike Mayday]

This forum has a general no-profanity attitude, right?

Rekov: Vettlingr has indeed proposed a similar approach. I like it! Thanks for the input, I'm probably gonna implement it.

[Rekov]

I've thought about this some, and I've come up with a procedural solution for every possible ramp scenario.

Picture each tile as having a point in the center, and eight points around the edges, four at the corners and four at the midpoints. Triangles radiate out from the center.



Rules for creating ramps:
1. Each of these points only moves along the z axis.
2. The center point always ends up exactly halfway between z-levels.
3. Move remaining points up or down to connect with their equivalents on neighboring tiles.

To demonstrate this, here is the little scene with the magenta and orange markings recreated by this method



Here is another demonstration. In this case, my procedure creates a significantly different geometry than the one implied in the example:



Especially in the top left corner the solution is different.


Each triangle has one of only four orientations. All that remains is to smooth the shading across transitions between these orientations and add shadows as desired.

All of that having been said, ya'll are artists, and certain tiles like the orange marked one differ from this solution and look much better because of it. However if you guys ever end up stuck, this is a sure way of arriving at a viable geometry.

[Mike Mayday]

In general I like it. As for specific applications - the viability of the geometry is only one side of the coin. The other is that the resulting geometry needs to be readable from 100% top-down view. Which is probably why you feel like some of my solutions look better. I will play around with these shapes when I get back to ramps.