Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[cerapa]

There actually already are rain shadows. I am currently "dragging" humidity to the east, reducing it by 1 per every tile and difference in height.

[Techhead]

Ah, I see that now. That's probably a better way of modelling it. I didn't notice initially because I was mainly looking at the mountains instead of just wide strips of land. Two nitpicks/suggestions:

First, you should modify it so that humidity travels westward in equatorial regions. The trades blow east to west.

Second, you might want to base it on just elevation instead of a combination of elevation and distance. The Amazon river continues to be humid 1000 miles west from the Atlantic, while small mountainous strips like Madagascar and Baja California have very noticeable rain shadows. Northern and Southern Baja California actually display rain shadows on opposite sides due to differences in prevailing winds.

Disclaimer: I am not a climatologist. I just remember my Earth Science and Geography classes and did additional research as necessary.

[10ebbor10]

Distance is important for the amount of rain, but by far not as important as mountains. Temperature also influences.

[cerapa]

Wasn't able to get much done today, cept for splitting a couple of things up into different functions.

So, gonna do a bit of a humidity rewrite:
No distance loss.
Wind direction changes on latitude. (how to make a smooth change without having distance loss?)
Temperature determines generated/lost humidity(?)

EDIT:

Initial humidity, after removing distance loss and doubling height loss to compensate.

EDIT2:

Spoiler (click to show/hide)

Seems I have some work to do.

I also need to redo the terrain a bit. It currently has too many seperate islands and not enough continents. Its okay when the water level is low, but once it gets higher you stop having good landmasses.

[GalenEvil]

Something I did to get more continents was have a bias map alter the noise values a little based on how far they are from the center of the map or some other control point(s). You could also instead apply a low granularity simplex map over the terrain and normalize the end values back to the range you are using. EG: Your most detailed map is 100x100 in size, your low grade map might be 10x10 with each value within applied to a 10x10 area of the detail map. The lower detail map might also be biased so that it gives a higher average value, which will cause more land masses to form instead of island chains.

[10ebbor10]

Note, for the landmasses, you could try to do what Earth did. Ie, start with a single pangea continent, then randomly place some hotspots around, and then apply tectonics.
(That is probably to complicated though).

Oh, and a note, that map you got there isn't that accurate, it might be better just to set up a athmospherical pressure system based on temperature, and draw your wind speeds for that. That one would be much more fun when terraforming.

[cerapa]

Oh, and a note, that map you got there isn't that accurate, it might be better just to set up a athmospherical pressure system based on temperature, and draw your wind speeds for that. That one would be much more fun when terraforming.

Very true. Needs a bit of thinking though.

Initial thought with that is to calculate the differences in temperature, blur with weights according to those differences, and have height remove humidity every blur cycle. Stir until stabilization. And of course to add the coriolis effect into the proceedings. But this has a major problem in that wind will always blow in the direction of the sea(at least with the current heat system), so humidity wont be properly distributed.

Then again, perhaps the equator being warmer would sufficiently change the wind direction to make it blow onto land.

This is quite complicated. Would be helpful if someone could explain to me what exactly causes wind to act like it does. And if temperature is identical to atmospheric pressure for practical purposes?

[10ebbor10]

There are 3 major forces influencing wind direction.
      -First of all is air pressure. Hot temperatures cause air to expand and rise, resulting in a low pressure area. (Air moves away through the higher parts of the athmosphere). Low temperatures cause pressure to increase as the air shrinks and drops. This is generally true, though winds and other phenomena can cause local anomalies. Nothing to worry about though. Now it's only a point of connecting the dots.*
      -Second is the Coriolis, which causes High and low pressure cores to rotate, so that the air can't follow straight lines. Another interesting effect is the fact that air heading towards the equator will turn towards East.
      -Third is local topography. Ie, mountains stop winds.

Then there's some other complicated systems involving storm fronts and such, but those are more local phenomena rather than continues trends.

*(Note: The sea will often function as a hightemperature area in winter, and a low temperature area in summer, creating rain seasons)
Edit: This might become a bit to complicated. Especially with changing temperatures. (Worst case scenario you just make 2 heatmaps, one for winter and one for summer.)

[cerapa]

I think I'm gonna leave that system for later. I arent even sure of the particular timescales that I'm gonna be dealing with yet. Turn based though, possibly with a year being a turn, but "a year" means very different things on different planets. Will see sometime later.

But for now, I did a bunch of housekeeping. My coding practises are pretty bad, so I tend to leave important bits and pieces where they dont belong, like height generation also being responsible for a piece of humidity generation. Without crap lying around, I should be able to work much faster now. Plus not having the same function for both planet generation and map rendering is just a good thing if you want to see humidity without changing the source code.

Spoiler (click to show/hide)

If these aren't self explanatory, then I have failed as a designer.

[10ebbor10]

Standard
Humidy
Heat
Clouds?

[cerapa]

Clouds?

Bah, knew you would say that.

Supposed to be a height map. Trying to show the difference between sea and land is a problem.

This shows everything clearly, but it isn't very evocative of a heightmap.


Ah, much better, blends together really well. Is this okay as a heightmap?

[10ebbor10]

Kinda, maybe you could add heightlines to drive the point home?

[cerapa]

Needs some work. And I really oughta make it so that the coastline gets drawn as a heightline, and then the others are just 30 height units away from that.

[Valid_Dark]

cool project, posting to watch.

[10ebbor10]

Looks nice.

Just a question, what's the actually ingame significance of the maps.