Graham-
|
|
« Reply #2980 on: November 11, 2012, 07:30:21 AM » |
|
Yeah, it's annoying.
|
|
|
Logged
|
|
|
|
InfiniteStateMachine
|
|
« Reply #2981 on: November 11, 2012, 07:32:24 AM » |
|
roger, I'd say one screen at a time rather than one line at a time would help immensely
|
|
|
Logged
|
|
|
|
impulse9
Guest
|
|
« Reply #2982 on: November 11, 2012, 07:50:38 AM » |
|
Yeah. I think it's mainly the amount of text that is annoying. If it's 4 or 5 lines of text (so 2 scrolls), it appears kind of cute.
InfiniteStateMachine: thanks for the tip, I might just do that.
|
|
|
Logged
|
|
|
|
Graham-
|
|
« Reply #2983 on: November 11, 2012, 07:54:29 AM » |
|
You can change the scroll pattern based on contents. This is something that isn't done enough in text games. It's one more way to add character.
|
|
|
Logged
|
|
|
|
impulse9
Guest
|
|
« Reply #2984 on: November 11, 2012, 07:59:54 AM » |
|
That's actually a pretty good idea. I doubt that a lot of text will be displayed in-game though. This is mainly for quirky NPC one-liners. Still, it's good to plan ahead.
|
|
|
Logged
|
|
|
|
rivon
|
|
« Reply #2985 on: November 11, 2012, 08:15:35 AM » |
|
Linus does Whoa, whoa. Fucking Linus. He understands everything. No, no... He understands it because he wrote it
|
|
|
Logged
|
|
|
|
Graham-
|
|
« Reply #2986 on: November 11, 2012, 08:37:07 AM » |
|
Yes, I know
|
|
|
Logged
|
|
|
|
|
nospoon
Level 1
|
|
« Reply #2988 on: November 11, 2012, 02:17:21 PM » |
|
I implemented per pixel collisions in libGDX today :D
|
|
|
Logged
|
|
|
|
Geti
|
|
« Reply #2989 on: November 11, 2012, 03:26:19 PM » |
|
03:30, the day before my uni presentation, after spending at least 5 hours trying to figure out what was wrong, I finally found the bug in my implementation of Perlin noise Glad you got it working; You're gonna want some erosion methods next, because that sure does look like a fractal noise terrain
|
|
|
Logged
|
|
|
|
eigenbom
|
|
« Reply #2990 on: November 11, 2012, 04:18:13 PM » |
|
03:30, the day before my uni presentation, after spending at least 5 hours trying to figure out what was wrong, I finally found the bug in my implementation of Perlin noise Glad you got it working; You're gonna want some erosion methods next, because that sure does look like a fractal noise terrain hybrid multi-fractal will make it look a lot better too, and its just a minor adjustment to normal fractal noise :D
|
|
|
Logged
|
|
|
|
dr.crow
Level 1
|
|
« Reply #2991 on: November 11, 2012, 04:25:25 PM » |
|
Do you have any particular algorithm to suggest? I tried gaussian blur, but I want to do better. I'm currently looking at thermal and hydraulic erosion in this paper.
|
|
|
Logged
|
|
|
|
eigenbom
|
|
« Reply #2992 on: November 11, 2012, 04:57:04 PM » |
|
Do you have any particular algorithm to suggest? I tried gaussian blur, but I want to do better. I'm currently looking at thermal and hydraulic erosion in this paper. Hybrid Multi-fractal as I said. The original C code by Musgrave is here. (Texturing and Modeling: A Procedural Approach is the best book on procedural terrain algorithms.) This is basically what it looks like: Note that some areas are smooth, which can be thought of as erosion. There are other methods, that e.g., simulate actual rain, but hybrid multifractals are good enough and super fast. There's also Ridged Multi-fractal for a different kind of mountainous terrain.
|
|
|
Logged
|
|
|
|
Geti
|
|
« Reply #2993 on: November 11, 2012, 08:15:35 PM » |
|
You can also use a larger scale lower resolution noisemap for mapping "biomes", or whatever you want to call larger terrain features, and then interpolate from one terrain generator to the next (eg the above multifractal for mountains to a softer ridged fractal (something tangent-based perhaps) with some areas of harsher variation for deserts). If you're doing your generation offline doing a simulated erosion pass for rain, and especially at sea edges (tidal erosion) is great for adding believability. Blurring isn't really what erosion is like, erosion can keep sharp features and helps to enhance channels in mountains. A simple flow-based erosion algorithm isn't too hard to write up: get the implied slope by sampling around each pixel (5x5 areas work well enough, and you can sample from a mipmap if you want to sample larger areas without wasting cpu time), and then move some of the height from this pixel to the pixel in the direction of the slope. Boom.
|
|
|
Logged
|
|
|
|
dr.crow
Level 1
|
|
« Reply #2994 on: November 11, 2012, 11:18:29 PM » |
|
Thanks for the suggestions! I did hybrid multi-fractal: Look very spider-webby, though, guess I'll have to play around with the parameters a bit. Next up is ridged multifractal and then I'm going to attempt some thermal and tidal erosion.
|
|
|
Logged
|
|
|
|
George Michaels
|
|
« Reply #2995 on: November 12, 2012, 03:25:57 AM » |
|
I fixed the bug in my collision detection code that's been driving me nuts for the last week.
Turns out I wrote "x = m * x + c" instead of "x = (y - c) / m" making it return an x that was out of range.
|
|
|
Logged
|
Yeah, that.
|
|
|
InfiniteStateMachine
|
|
« Reply #2996 on: November 12, 2012, 05:21:31 AM » |
|
As far as I know I won't have to do anymore Java/Swing coding for a while. Back to C
|
|
|
Logged
|
|
|
|
xhunterko
|
|
« Reply #2997 on: November 13, 2012, 11:34:25 AM » |
|
Just fixed a scoring feature. The player is now properly penalized for asking for help. Yay!
|
|
|
Logged
|
|
|
|
eigenbom
|
|
« Reply #2998 on: November 13, 2012, 04:24:54 PM » |
|
f&*k yeah, got the latest moonman running on a mac ... weeee ... ecstasy...
|
|
|
Logged
|
|
|
|
kiddRaddical
|
|
« Reply #2999 on: November 13, 2012, 05:51:54 PM » |
|
"Personal Best" times and scores now saved between games, show up in world map :D
|
|
|
Logged
|
posting (too) regularly on Twitter: @EthanRedd EthanRedd.com
|
|
|
|