next »

[Thaumaturge]

Well, all of ability, time (or other resources, such as money), and inclination, I'd say. There may, after all, be other elements of the project that take precedence.

That said, I was speaking to the general case--and how common is it that a given developer (especially a small or solo indie developer) is in a position to implement such a feature, and do so without taking a significant amount of time away from other elements of the game, or money from their budget?

As I said, I have little argument with this specific case; I'm primarily cautioning, I think, against someone reading the above posts as indicating that it's generally true that it's better to make one's own engine rather than using extant middleware.

[qMopey]

Yeah, that's why people that tend to write their own little physics engines in 100 lines or so were at some point professional devs. They got those skills at a time when they weren't making an independent title. The implication is getting that kind of skill is time consuming, whereas having that kind of skill on-hand is really valuable.

[Thaumaturge]

Yeah, that's why people that tend to write their own little physics engines in 100 lines or so were at some point professional devs. They got those skills at a time when they weren't making an independent title.

Heh, I wouldn't be surprised if that were true, indeed. ^_^

As a minor counter-point, I myself once worked in a (small, admittedly) game studio, and I've even made a simple 2D physics engine for J2ME mobile devices. That said, unless I had an extremely good reason, I'd prefer to stick to middleware--more time spent on the game, less on technical matters, in general. (As seen on a later mobile project, for which I used a middleware engine.)

Of course, physics is not my forte as a developer, and I daresay that it would take me more than a hundred lines, and a fair bit of time, to make a decent 3D physics engine. (I daresay that I could do it; it would just be a fair investment.)

That said, I suppose that I'm not really arguing against you here. Rather, I want to add a cautionary note in case someone reads your quote above and infers that any one-professional dev. should be able to make a 3D physics engine in a hundred lines and a short time-frame. Different devs have different skills, and different levels of experience.

The implication is getting that kind of skill is time consuming, whereas having that kind of skill on-hand is really valuable.

True, I imagine. And I think that that's true of pretty much any applicable skill: physics, gameplay, art, rendering, and so on.

[GrumpyGiant]

Been experimenting with the houselegs this week. I wasn't happy with the beam; visually it seemed OK but I didn't like how it went right through things.

The main idea of the game, in-between solving puzzles, is that the houselegs is supposed to be searching for you. Now in order for that to work, the player needs to be able to hide behind things. But how does the player know if they're suitably hidden or not? If the beam just goes straight through walls, there's no indication to the player whether they're visible. The player needs to be able to see the beam approaching and keep out of its way.

What I need is a way to demonstrate visibility. So I added a dynamic shadow map to the game, rendered from the point of view of the houselegs. With this I can shine an actual spotlight circle onto the ground. Furthermore, I can do some hacky volumetric effects on the beam itself to make the beam stop once it hits something. This lets the player stay behind a wall and know that they can't be seen.

I'm really pleased with how this turned out. It's starting to feel like the right kind of gameplay now.

You may notice, however, that I've also switched the lighting from sunny daytime to a darker, overcast look. With the daytime, I just couldn't get the beam to actually look like any kind of decent searchlight. But with a darker background, it can now really stand out.

I'm not sure if this is the right direction visually, but I'm liking it so far. I did kinda like the bright sunny look, so I'm going to try some experiments to see if we can't get some kind of compromise between the two -- possibly having areas start grey, and become clear once you complete them.

I added some stormy clouds overhead too, just for good measure. Because they're dynamic and not just baked into the skybox, I should be able to remove them as needed to turn the dark sky into a clear one.

[CaptainBlueSkies]

I dig it! I can't tell if the images are more foggy, or blooming with light, but I'd be interested to explore that world!

[Thaumaturge]

That volumetric spotlight looks rather good, to my eye! ^_^

[Pawl]

Just found this game, looks cool! Any updates? :-)