Welcome, Guest. Please login or register.

Login with username, password and session length

 
Advanced search

1411490 Posts in 69371 Topics- by 58428 Members - Latest Member: shelton786

April 25, 2024, 01:32:14 AM

Need hosting? Check out Digital Ocean
(more details in this thread)
TIGSource ForumsCommunityDevLogsProject Sumo
Pages: 1 [2]
Print
Author Topic: Project Sumo  (Read 4472 times)
Limne
Level 0
**


View Profile WWW
« Reply #20 on: October 25, 2011, 06:36:45 PM »

I haven't too too much to talk about today. Unfortunately, most of today had to be written off between a several-hours long commute, a dental work marathon and down time away from my tools. Fortunately, I brought my C++ textbook on the bus and was able to read up on various data structures.

As for what I have to show:


I was working on a cleaner version of the bird man. This one features a larger head and squatter proportions; also less larger lines and less detail. You can see another version of the scarf I did for comparison. I'm a real sucker for complex drapery like this but it just doesn't look any good shrunk down to a realistic raster. Often adding in a lot of detail like that will kill a work's overall contrast too if you don't balance it out with areas with little detail. I noticed here that I could probably use the same forearm piece for both arms because they're both at about the same angle and mainly feature a prominent ulna. It makes me glad when I can think of ways to use my technologies efficiently.

I'm still trying to get a feel for what looks good on a screen. On paper, resolution is never an issue. On a screen, if you want all those details like facial features and ribs and knuckles to come through and have to make sure they're big enough. Of course, you can only make your characters as big as the game design dictates. The obvious solution is to go with that "super deformed" look. I'm still thinking about it.

In retrospect, I don't like the way this design looks as much as my pen and paper one. It might just be my expressionistic tastes. I think what I might try is squishing my color sketches down to fit a realistic character size given 720p and then kind of paint something in at that size, then size it up to do clean-up.

I continue to dwell on the subject of animation. I think it would be really cool to have different animations for if a strike connects or not; like, if it does, you can see the force of the impact kill their forward momentum, and if it doesn't, you can see them try to redirect that momentum to keep on balance. Thai boxing yesterday made me think about stuff like that. Also, about how the last strike you've thrown effects how you thrown the next one due to position, balance, and inertia. Beyond just animation, stuff like that makes me think about mechanics and how I'd like for players to be able play intelligently, maybe through a system based on throwing counters and context sensitive moves. I think the only fighting games I've played with a really great system for that were the Punch Out games (Probably my favorites after the Technos brawlers). But now I'm just plain rambling.

I'm glad to see Jared's working hard on the editor. Just make sure not to kill yourself with overwork my friend Wink
Logged

- Laura Lake
- lauralake.ca
Uykered
Guest
« Reply #21 on: October 25, 2011, 07:23:11 PM »

That new bird guy looks a lot better than the older version sketches!
Logged
Limne
Level 0
**


View Profile WWW
« Reply #22 on: October 26, 2011, 04:26:43 PM »

That new bird guy looks a lot better than the older version sketches!

Ya? Do you mean it looks cleaner or do you prefer the design better?
Logged

- Laura Lake
- lauralake.ca
Uykered
Guest
« Reply #23 on: October 26, 2011, 05:31:24 PM »

Well ya its cleaner without the sketchy look, the line art just looks like a massive improvement. How are you going to do the shading?
Logged
Virion
Level 0
***



View Profile
« Reply #24 on: October 26, 2011, 08:47:32 PM »

Quote
Looks good. I like this style. Very solid looking. Smiley
Logged
Limne
Level 0
**


View Profile WWW
« Reply #25 on: October 28, 2011, 08:37:17 PM »

So, I spent a decent amount of time drawing tonight but I really don't have too much to show. Certainly not enought to justify the ftp shenanigans it takes to upload anything. But here's the low down:

First of all, I'm glad that the cleaner art work was well recieved. I suppose that's only natural. All things considered though, polish is just grunt work; straight-forward but time consuming. In my opinion, it's the deeper principals of design that really make an illustration.

Quote
How are you going to do the shading?

Well... Hard to say. This is something I did in Photoshop by mouse last year:

http://lauralake.ca/?webcomic_post=ancient-civ-cover-art

I now have a Cintiq. That was also my test run for applying reflective lighting, a principal I'd like to further refine. Mostly, I'd like to kind of "paint" stuff in by semi-transparent layers that help preserve harder shadow shapes. Of course, shrunk down to fit a screen, it'll look a lot smoother. I might try to make the exact colors brighter and more vibrant like I do in my sketchbook though (doing stuff like making highlights, midtones and shadows gradually shift from one hue into a fairly different one). I kind of like psychedelic stuff.

But really, for the time being I figure I'm going to concentrate more on design. The important thing for now is to know what I want to work with in the editor. I think the best way is to take an iterative approach; start out with some messy sketches of characters and then gradually add more and more polish to each of them. I think that makes more sense than doing on character at a time, becuase if I do it that way, I'm sure I'll end up setting a standard for myself that's unreasonable for an entire game's art assets, and then I'd probably end up improving over time on top of that so that character's would be of wildly varying quality and design.


Today I decided I'd try drawing something more generic-human to help give me a better idea of how well my designs implement during testing. I've been kind of drawing rough figures, shrinking them down (For right now, to test with, I'm looking at 128 x 192 characters on a 720p widescreen raster.) making design changes based on the smaller size (Like eliminating the kind of forced perspective I tend to add drawing characters big and close up that you wouldn't get small and in a long shot.) then sizing them bad up to tinker with the details and so on. One thing I can say is that I think the character's look a lot better with a variable width line; like especially having heavier countours on the outside and a smaller line width on the inside details. Implementing that using node based sprites is going to be "interesting" though. Particularly because my brushes in Photoshop will just randomly break and start behaving way different with thicker or thinner line width despite that nothings changed in the settings. Man, that's a bug I'd love to see fixed.
Logged

- Laura Lake
- lauralake.ca
eigenbom
Level 10
*****


@eigenbom


View Profile WWW
« Reply #26 on: November 10, 2011, 03:42:28 PM »

I'm really digging the art for this. Smiley

The creatures all look very frankensteinian, maybe you could procedurally join different body parts together to get random monster creation at runtime?
Logged

Pages: 1 [2]
Print
Jump to:  

Theme orange-lt created by panic