[CC] Yeti game sprites
Printed From: Pixel Joint
Category: Pixel Art
Forum Name: WIP (Work In Progress)
Forum Discription: Get crits and comments on your pixel WIPs and other art too!
URL: https://pixeljoint.com/forum/forum_posts.asp?TID=24629
Printed Date: 23 October 2025 at 12:42am
Topic: [CC] Yeti game sprites
Posted By: scorched
Subject: [CC] Yeti game sprites
Date Posted: 12 October 2015 at 2:23pm
|
This is the revised yeti sprites for my little game (continuation of the Ludum Dare one). Yeti is the main character and lives in the icy forest, so the background is bright. The sprites will be highly scaled to ~1/5 screen height. What do you think?  |
Replies:
Posted By: jalonso
Date Posted: 12 October 2015 at 5:53pm
|
You need to work on that palette so it reads. ------------- |
Posted By: scorched
Date Posted: 13 October 2015 at 9:15am
Ok, here we go: I also changed the background to an actual one used at the moment |
Posted By: Mandrill
Date Posted: 13 October 2015 at 11:38am
|
Better! The body hair is still quite hard to read. I suggest removing the dithering and working in clusters with a colour that is closer to the skin tone. |
Posted By: scorched
Date Posted: 13 October 2015 at 1:51pm
@Mandrill, how about this one? |
Posted By: dpixel
Date Posted: 14 October 2015 at 6:17am
Tried my hand at hair texture and better readability. The dithering will be tough to make look right here. The yeti is mostly hair except for the face (so I've been told), so the dark hair on it's head just makes it look like some dude with dark hair. ------------- |
Posted By: jalonso
Date Posted: 14 October 2015 at 7:48am
|
Dpixel has the right idea because it looks like a yeti which I did not know its what you were making.
Readability is everything in pixelart. ------------- |
Posted By: scorched
Date Posted: 14 October 2015 at 3:06pm
Thanks, guys. Do I get it correctly now?   |
Posted By: jalonso
Date Posted: 14 October 2015 at 3:23pm
|
Better but use that medium blue you have on the sides to enhance the fur without doing the pixel pecking you did originally. Try to make it read as fur.
Change that almost black to a dark enough blue to be dark but light enough to read blue. ------------- |
Posted By: scorched
Date Posted: 15 October 2015 at 1:14pm
Did you mean this way? |
Posted By: jalonso
Date Posted: 15 October 2015 at 2:15pm
|
Well the darkest blue can be a little darker if you like.
I do think you've gone back to the first version where it just looks like pixels all over the place almost random. Use the colors you have to help define the creature as much as you can. Study dpixel's edit again. ------------- |
Posted By: scorched
Date Posted: 16 October 2015 at 12:23pm
Hmm... So should I define him something like that? |
Posted By: jalonso
Date Posted: 16 October 2015 at 1:00pm
|
Exactly. See how you used the darker shades to define and help the viewer read it better?
As the artist its always clear in your mind but you have to help the viewer too :) ------------- |
Posted By: StoneStephenT
Date Posted: 16 October 2015 at 1:17pm
|
Originally posted by jalonso As the artist its always clear in your mind but you have to help the viewer too :) To quote http://terribleminds.com/ramble/2015/03/16/writing-is-a-profane-irrational-imperfect-act/ - author Chuck Wendig : “The end result will never match the expectation. You will never get it just right. The idea is God: perfect, divine, incapable of repudiation, utterly untouchable. The result is Man: fumbling, foolish, a jester’s mockery, a bundle of mistakes in tacky pants.” Your goal isn’t to make a perfect sprite. Your goal is to make it as close to the idea in your head as possible while still making it clear enough for a viewer to grasp. My current WIP is a good example of this philosophy: I have an idea in my head, but regardless of how hard I try, I can’t get the pixels to match the idea—and that’s okay. The idea is God; the result is Man (or in my case, an imperfect representation of the male body  ). The people helping me out in that thread generally understand what I want to achieve and are helping me get closer to a…well, not “more perfect”, but “less imperfect” representation of the idea in my head.Think about how you can best represent the idea of a “yeti” or “abominable snowman” while making sure the idea can be understood by other people. You’ll never hit upon a “perfect” representation of your idea—but you can always make it less imperfect.  |
Posted By: scorched
Date Posted: 16 October 2015 at 2:25pm
|
Okay, now that I've got a direction, I'll work towards it. @StoneStephenT, I uploaded the last version to hosting like a 5 times, after each try a little @jalonso popped up in my head and said "it's still crap. do it again" |
Posted By: jalonso
Date Posted: 16 October 2015 at 3:11pm
|
Not crap. You really are trying hard and learning. All is good but do keep me in your mind always screaming to be a little bit better and trying a little harder till the day you die :p ------------- |
Posted By: scorched
Date Posted: 17 October 2015 at 1:02am
Applying further our "newly discovered" technique... |
Posted By: jalonso
Date Posted: 17 October 2015 at 6:21am
|
Are these 3 views of the same yeti in different angles?
If yes, why are they different height? Looks like the view is straight on front so all feet would be on the same line including front view where the toes would not angle down. ------------- |
Posted By: scorched
Date Posted: 17 October 2015 at 7:16am
|
The view is JRPG-style (not a platformer-style). Yes, it's the same yeti. Hmm, I suppose that the left one needs fixing then? |
Posted By: StoneStephenT
Date Posted: 17 October 2015 at 9:17am
Pardon the wall of text that I’m about to sling your way, scorched.  Take a look at all of these sprites from JRPGs (in order, from left to right: Suikoden II, Chrono Trigger, Final Fantasy VI, Shining Force II, Phantasy Star II, and Phantasy Star III). What do you see when you first look at them? What sort of ideas pop into your head as you look at these sprites? If you were viewing these as if viewing them for the first time (maybe you are!), what would be the first thing you'd say about them? The quick and easy answer is that they’re all clearly human. Even with the side-view sprite of Sabin from Final Fantasy VI, you get the idea of “human” pretty damn quick. A closer study of that particular sprite makes it easy to pick out which pixels represent which body part. Even with the sprite of Rolf from Phantasy Star II (objectively the worst-looking sprite of this selection), you can still get the sense that the sprite means to represent a human body. Here’s the kicker: your sprite is meant to do the same thing. Here’s the latest version of your yeti sprite (as of the time of this post). You can make out that the figure is supposed to be humanoid thanks to the way you've constructed the head, arms, and legs. On that basic level, you’ve accomplished your task. Kudos! But as far as everything else goes, you have a bit of a mess going on. Go back and look at each of those JRPG sprites I posted: you can make out individual body parts, clothing, and the anatomical proportions of each individual sprite. (Except maybe from looking at the sprite of Bowie from Shining Force II. His damn cape’s in the way.  ) You get a sense of how each character might look if drawn in a “lifelike” manner. Your sprite…well, it doesn’t quite do that.The back view is about the closest you get to getting on the same level as some of those sprites. The pattern of white fur running along his shoulders and back combined with the darker fur and his squashed-yet-readable anatomy makes it clear that this is supposed to be some sort of monster. The only real problem I have with this view is the coloring of that fur, but I'll come back around to that in a moment. The side view fares a little better, but only because the arms and legs are readable. The upper legs still look wonky, his torso has some anatomy issues going on (is he fat or do his upper legs run up to his chest?), and his head could use a little work, but the general idea is there. The front view is where your sprite suffers the most. Every anatomical issue with your other views is magnified here: his legs look a bit too long, his arms look too short for his body, his chest sits way too close to his hips/crotch, and his head is barely readable beyond its general shape and his eyes. And when I mentioned the coloring of his fur earlier, this is part of what I meant—because none of the darker fur here really comes off as fur. I’ve said all these mean things about your work not because I want to discourage you—heavens no!—but because you have an idea you want to see made a reality and I want to help you do that. One of the underlying philosophies of pixel art, regardless of whether even the most advance pixel artists realize it, is a form of minimalism: “How can I do more with less?” Go back to that set of JRPG sprites one more time, but this time compare what each sprite accomplishes within its own limitations. The Suikoden II sprite is larger than the other sprites and has a slightly higher color count than the others (I checked to make sure), but the other sprites are just as readable. They’re just not as detailed—and that’s fine. When working with smaller sprites, the challenge is not packing as much detail into them as possible, but how to best portray the details you choose to show. Therein lies the major problem with how you’re coloring your sprite: it looks as if you’re trying to pack too much detail into too small a space. Rather than trying to detail every strand of fur on the yeti’s body, try constructing large patches of fur that are detailed only where necessary. On the back view, for example, you could make the brighter fur covering his head one large patch of white(-ish) fur, then use some minor shading to separate it from the next patch (his shoulders and upper back), then separate that particular patch from the last one (his lower back). Those three big patches of white could still read as fur with some smart placement of both outline and shading pixels—and you’d sacrifice virtually no readability in the process. If you’re really eager to mimic JRPG sprites with your work, I’d suggest going to http://www.spriters-resource.com/ - The Spriter’s Resource , grabbing some spritesheets from your favorite games, and studying them with an eye towards how they used specific details to create the effect they were going for. (Given the size of your sprite, studying SNES-, Genesis-, and GBA-era JRPGs would likely prove more beneficial than, say, studying sprites from the PS2 era.) Get a feel for how a little bit of detail—even just a single pixel—can change how an entire sprite looks. Try editing a sprite to give it a different haircut or a new outfit while staying in the same style as the original sprite. Figure out how the same principles that guided those artists can best be used in your own work. I believe in you, scorched. I believe you can make a well-constructed yeti sprite that both matches the idea in your head and looks like a monster to the average viewer. But to do so, you need to be willing to put in some time studying the works of others—and put some serious time into constructing your sprite. Your general idea is good, and I believe you can make it even better. |
Posted By: scorched
Date Posted: 17 October 2015 at 10:02am
|
@StoneStephenT, thank you, I'm posting pics here exactly to hear some advice from experienced pixel artists. I don't want to mimic JRPGs, and don't much like the genre, but I do love graphics in them. I mentioned it because I wanted to describe the view angles :-) I understand though that all your post still applies to my WIP, and I don't mind to rethink it again :-) http://pixeljoint.com/forum/member_profile.asp?PF=42346&FID=8 - |
Posted By: scorched
Date Posted: 17 October 2015 at 12:25pm
Boo! |
Posted By: StoneStephenT
Date Posted: 17 October 2015 at 1:16pm
|
That’s actually a hell of an improvement, scorch. The new coloring is more readable, the anatomy on the side view looks better, and the head on both the side and front views looks far better than it did before (the addition of the jaw on the side view and the “fluff” on the front view did wonders). It ain’t perfect (what is?), but it’s far less imperfect than the prior version.
|
Posted By: scorched
Date Posted: 18 October 2015 at 7:04am
Thanks, I admit to not being attentive again, as the actual background colour was this: Do you think I should change blue colors to brighter ones, or it's still ok? Like this:  |
Posted By: scorched
Date Posted: 18 October 2015 at 11:22am
Here is also the game screenshot (updated soldier today): |
Posted By: dpixel
Date Posted: 19 October 2015 at 10:21am
|
The background is pretty weak. It needs some shadows to define the lay of the land or something. Assuming this is a snowy scene and not sure of the perspective you're going for I did an edit as a quick example how to maybe pull things together.  On the other hand, the yeti looks pretty good. ------------- |
Posted By: scorched
Date Posted: 19 October 2015 at 11:37am
|
@dpixel, thanks for you example, it's really inspiring! The view I'm aiming is pretty like the picture below (and after looking at it myself I realize I need some tweaking). It has no perspective:  The main restriction is that levels in this game are defined mainly by placing trees (obstacles) and soldiers, so the trees are part of the game, and really not the background. On the other hand, I'm free to place additional walkable things and make trees look more diverse. The main difficulty with your example is the ground. It has to be equally walkable and I can only imagine nearly flat land for this, no ups and downs (except as a part of an obstacle). If you have advice on any of the points, I'll be glad to hear! |
Posted By: eishiya
Date Posted: 19 October 2015 at 12:21pm
|
The 3/4 RPG perspective often uses terrain to create obstacles, while not actually having a 3rd dimension. For example, in the image you posted, the "cliffs" are really just obstacles that make the parts above them look elevated, when really it's all just a 2D layout. You could certainly do the same in your game. In dpixel's version, the larger dark blue areas (shadows on the snow) could serve as obstacles. |
Posted By: dpixel
Date Posted: 19 October 2015 at 12:26pm
|
I think that would be 3/4 top-down perspective. If that's the case, the yeti is wrong. You would be able to see more of the top of its head.
------------- |
Posted By: scorched
Date Posted: 19 October 2015 at 2:05pm
|
Ok, I need some time to make my idea whole from the pieces I have. Thanks for correction on the terms. I do understand the programming part and already have 8 working levels, planning 16 total. When speaking about the ground I meant not every level would allow me to make that large obstacles. In some cases, I can put shadow-obstacles only at level sides (which is not a bad idea still  ). |
Posted By: scorched
Date Posted: 26 December 2015 at 11:33am
|
It's been a while since I posted here... Now that levels are done, and I made the next Ludum Dare game, I'd like to return to pixel art practice. Here's the screenshot of the current progress:  |
Posted By: eishiya
Date Posted: 26 December 2015 at 12:37pm
|
Is the cave meant to be something you can enter? It looks too small, if it is. The scale overall is a bit odd - just how tall is the yeti? I get that they're big, but surely they're not 20ft tall. Scale issues are expected and forgiven in video games usually, but because you're using realistic proportions on the yeti, everything else feels out of place being "chibi". |
Posted By: scorched
Date Posted: 26 December 2015 at 1:12pm
|
I meant yeti to stoop when going in the cave. But I'll think about scaling, thanks! |
Posted By: eishiya
Date Posted: 26 December 2015 at 4:23pm
|
A stooping animation would look great for the cave! And the thing with caves is that people don't have a very strong expectation of their size. However, the mismatch of style (more so than the mismatch of scale) is still an issue with the trees. |
Posted By: Friend
Date Posted: 28 December 2015 at 12:55pm
| amazing improvements!!!! digging the colors. Keep workin hard |
Posted By: scorched
Date Posted: 12 January 2016 at 6:08am
So... is this scaling OK? |