Post you comments to this announcement here.
Discuss: A Change of Direction
Post you comments to this announcement here.
Last edited by RonHiler; 02-10-2009 at 08:49 AM.
"They laughed when I said I was going to be a comedian ... They're not laughing now." - Bob Monkhouse
quick response
I've been kinda surprised at how long you've been going at MD... but I can't say too much since the project I've been working on has been in "gamma testing" since before I joined in 2003.
[When did you start it, anyway? I remember finding you my first year in college, which would have been... late 1998. That makes MD's incarnations ten years old, yes?]
So, while it's sad to see MD get shelved until someone else wants it... I've had enough exposure to game engine design/coding to have some idea how much it is not game design/coding :) One of the reasons I'm thinking vaguely about pursuing a PhD in Software Engineering is so we can teach computers how to do more of the coding part of things.
I don't think I've heard of the C4 engine, but if it does what you need it to, then there you are! And very good news to hear that you've gotten so far with it so quickly.
Yeah, sounds about right.Originally Posted by Strike
You willI don't think I've heard of the C4 engine, but if it does what you need it to, then there you are! And very good news to hear that you've gotten so far with it so quickly.I misspoke in the announcement, the engine's only been around since 2005, not 2002 (I'll edit that)
If you want to check it out:
The engine's Wiki
http://www.terathon.com/wiki/index.php?title=Main_Page
Release notes (all of them since 2005)
http://www.terathon.com/c4engine/notes.php
The forums
http://www.terathon.com/forums/index.php
(this is where you will find all my stupid newbie questions, sans the licensed area stuff)
And, if you are really adventurous, the demo
http://www.terathon.com/c4engine/download.php
"They laughed when I said I was going to be a comedian ... They're not laughing now." - Bob Monkhouse
I was wondering about that date :)
I'll have to peek at the wiki, but first I need the time. Last couple weeks have just been overcrowded.
Splitting hairs here, but it was actually early '98, perhaps even late '97
Alas I'm sad to see it go, but understand this completely. Although they never got the multiplayer simultaneous turn system right in Civilization IV, at least they've implemented most of the stuff we discussed back then, most importantly multi-ethnic cities----- Original Message -----
From: "Ron Hiler" <bndwgn@pacbell.net>
To: "Innocence" <innocence@earthling.net>
Sent: Wednesday, April 08, 1998 1:36 PM
Subject: Re: Regarding Manifest Destiny
I've been planning on putting up a web-page about the game for a couple of months now, but I just haven't gotten around to it. Been busy coding and all that.
Tell you what. You're just the catalyst I need (I need a break from coding anyway, for a day or so). I'll get working on that web page, and I'll drop you an e-mail with the address as soon as I'm done. Should take a day or two.
Man, 1998... back then we were still waiting for the J-patch and Stars! Supernova
I look at game engines myself from time to time, although I realize I'll never get the time to get back into programming. There's a pretty extensive list of engines here, but then again I take it you've already done some browsing
Wow, there's a blast from the past, hehe.
I sincerely hope the next project is a) more successful and b) takes significantly less than 10 years, heh. I think both will be true
"They laughed when I said I was going to be a comedian ... They're not laughing now." - Bob Monkhouse
Last I checked, neither of those were ever delivered.Man, 1998... back then we were still waiting for the J-patch and Stars! Supernova
Regarding the whole ridiculous amount of time to produce a world thing, I should mention Dwarf Fortress, which can take 20 minutes to generate a small world on a relatively slow computer.
Having said that, I'm wondering if there is anything that could be done to cut that time down. Have you profiled the creation process to figure out where the bottlenecks are? I'm wondering if you can use or increase use of things like interpolation, and also possibly if you can ditch some of the iterative stuff in favor of synthesized generation.
An idea--and I haven't thought this out let alone prototyped it--would be the basics: generate tectonic plates, and assume they've been around for a while and have moved. Generate some land on top of them in more or less final position, and then synthesize features based on what would've happened had you iteratively generated them. Oh, you've got a landmass that straddles two continents? Figure out if there's a subduction zone or the kind that forms mountains, and if it's the latter, just slap down mountains roughly along the plate boundaries.
Sure I did. I used every trick I could figure out, hehe. The problem with interpolation is that you can only take it so far. If you get too carried away with it, you will end up with blurry maps. I *think* I'm already interpolating as much as I can (1 real texel per 64 displayed texels). If you go any higher than that, it's going to look bad (it may not even look that great at that level).
For a somewhat unrelated project, I was looking at L3DT (over at http://www.bundysoft.com/L3DT). This is a really nice terrain creation package that I played around with a bit (and it comes up with some pretty nice terrains, I have to say).
I thought about the idea of incorporating it into MD, but there are problems. One, it doesn't do Mollewiede maps (only rectangular ones), which is what MD needs to wrap the sphere with. Two, it doesn't come with source code (so I'd need to get it). Three, I don't think the license allows me to just dump the program into MD wholesale (so I'd need to get special permissions). And, four (and worst of all), a map the size we need would take just about a week to produce with that program (I tried) on MY computer (which is a quad core with 8 gigs of ram). Forget about it on anything less powerful, hehe.
All this is my way of saying that a package that does this exclusively (and admittedly produces nicer maps than I was doing, mostly due to the light mapping) still takes WAY longer than I was to make those same size (roughly) maps (I got mine down to hours rather than days).
I'm not saying it's an *unsolvable* problem, but I ran out of ideas. I did everything I knew how to do, and couldn't get it to where I thought it was acceptable. Perhaps it just needs someone more intellegent and/or a better programmer than I. I guess you could reduce the world size, but if you do that too much, it reduces the scope of the game, and it becomes something I'm not as interested in, to be honest.
Anyway, if you want the code, Pix, I can send it to you. Maybe someone with a fresh perspective can figure it out
"They laughed when I said I was going to be a comedian ... They're not laughing now." - Bob Monkhouse
When I was talking about interpolating, I was thinking, I guess, of an earlier stage. I forget now how big the planet's surface is, but I assume it's huge. To, for example, generate a set of tectonic plates, you don't need (say) a 4000x4000 map, I would think. Let's say 1/10 the size on each side, then at some point you stretch the map, and that's where you would interpolate. Draw a somewhat wandering line from each point defining the perimeter of the plate to the next point. I doubt you need a lot of resolution for that.
At some point I will probably take you up and look at the code but right now I'm busy playing with basic concepts, just for the heck of it, in world creation, and I don't have much free time for coding.
It might be interesting, if you have or can easily generate data showing the relative times of the various stages of your world building process.
Posting Permissions
Reply With Quote
