Creating World Maps
- Thread starter Flynn
- Start date
Dear Folks -
While I appreciate the "how-to's" posted by folk such as Stainless, this section bothers me:
Make your planet
================
(snip)
Choose "Map colour"
*** MAGIC HAPPENS HERE ***
Save image (map.jpeg)
(etc. etc.)
I think I need more help with the "magic" bit. Alternatively, could some kind soul convert the following image to a real map, and step through the process:
(Yes, there's considerable self-interest subtley hidden in this request of mine...
)
While I appreciate the "how-to's" posted by folk such as Stainless, this section bothers me:
To me, this reads more like:
Make your planet
================
(snip)
Choose "Map colour"
*** MAGIC HAPPENS HERE ***
Save image (map.jpeg)
(etc. etc.)
I think I need more help with the "magic" bit. Alternatively, could some kind soul convert the following image to a real map, and step through the process:
(Yes, there's considerable self-interest subtley hidden in this request of mine...
I would be interested to know what you mean by 'autogen sectors'. Have you generated an entire sector, with illustrations to this level of detail/quality? That sounds like a mammoth project.
I used H&E to gen the entire Spinward Marches with every planetary body and I think it took about 36 hours to run. The resulting data files are massive. About the only reason I keep them are to cover everything in the event of a missjump. For actual play I edit and/or regen the systems near the players, add more detail, better maps, trade goods etc etc.
No, by 'autogen' I refer to the 'SectorMaker.cgi' script I wrote that combines sec2pdf, gensec, mapsub, pdftk and enscript unix programs to generate an entire LBB in pdf form containing sector data, subsector and sector maps, polities, borders, trade routes, random world/sector names, custom cover text and back cover quote, and optional world UPP, animal encounters and Book 7 cargo. Along with the sector LBB pdf it can export a 'sec' file, an 'msec file', a list of polities and a text file containing the descriptions.
You can generate the sector's data, routes, polities and names randomly, or feed in existing sec/msec files for known sectors.
Takes 100 seconds for a 'brief LBB' (without animal and UPP descriptions) or 400 seconds for a full 'rich LBB' jammy.
Take a peek in the Classic Traveller forum for the thread, and check out
http://micki001.cnc.net/trav/SpinwardMarches.pdf for an example of the 'rich LBB'.
Note that I did not edit or modify the sector pdf in any way - it is entirely created by the software (with the exception of adding the 'worlds' cover image instead of the familiar black-and-solid color cover). Enjoy!
You can generate the sector's data, routes, polities and names randomly, or feed in existing sec/msec files for known sectors.
Takes 100 seconds for a 'brief LBB' (without animal and UPP descriptions) or 400 seconds for a full 'rich LBB' jammy.
Take a peek in the Classic Traveller forum for the thread, and check out
http://micki001.cnc.net/trav/SpinwardMarches.pdf for an example of the 'rich LBB'.
Note that I did not edit or modify the sector pdf in any way - it is entirely created by the software (with the exception of adding the 'worlds' cover image instead of the familiar black-and-solid color cover). Enjoy!
Wow .. I am blinded by your awesomeness!!
Bows before computer screen.
That's really impressive. At 1218 pages it's still a pretty big project but all the data is there at hand if you need it. Great work.
Just imagine how big it gets if you use H&E or World Builders Handbook to do every planetary body in every system. Encyclopedia Galactica anyone? The OTU available in a handy set of 762 Little Black Books!
Bows before computer screen.
That's really impressive. At 1218 pages it's still a pretty big project but all the data is there at hand if you need it. Great work.
Just imagine how big it gets if you use H&E or World Builders Handbook to do every planetary body in every system. Encyclopedia Galactica anyone? The OTU available in a handy set of 762 Little Black Books!
mike wightman
SOC-14 10K
Welcome aboard Alik
I'd like to second your request.
I'd like to second your request.
I mostly use LunarCell or one of a small collection of open-source programs in conjunction with Flexify to create maps. I usually create a Mollweide version in order to measure areas.
For pretty pictures, I use Glitterosity to create a background layer, sometimes SolarCell, if I want to do a sunrise, followed by a LunarCell planet. Occasionally, I add a lens flare to enhance the sunrise. This has lead to some quick beautiful pictures.
Lately I've been playing with the Greg Martin planet tutorial. The following resulted from variations on that idea.
This one followed the tutorial fairly slavishly. I added a Glitterosity starfield.
I think i made a mistake on this one. The mountains look like holes. Kinda strange.
This one came out okay. A slightly grassy savanna-world perhaps?
An experiment in putting land on a wet planet. I'm almost happy with the results.
LunarCell sometimes looks better, and it's definitely faster. On the other hand I think I can adapt the Greg Martin tutorial idea to make pictures out of maps I created with planet , or arbitrary bodies.
EDIT: Added link to Torben Mogensen's excellent planet program. Just click on the word "planet" above.
Or just click on the link below:
http://www.diku.dk/users/torbenm/planet.zip
For pretty pictures, I use Glitterosity to create a background layer, sometimes SolarCell, if I want to do a sunrise, followed by a LunarCell planet. Occasionally, I add a lens flare to enhance the sunrise. This has lead to some quick beautiful pictures.
Lately I've been playing with the Greg Martin planet tutorial. The following resulted from variations on that idea.
This one followed the tutorial fairly slavishly. I added a Glitterosity starfield.
I think i made a mistake on this one. The mountains look like holes. Kinda strange.
This one came out okay. A slightly grassy savanna-world perhaps?
An experiment in putting land on a wet planet. I'm almost happy with the results.
LunarCell sometimes looks better, and it's definitely faster. On the other hand I think I can adapt the Greg Martin tutorial idea to make pictures out of maps I created with planet , or arbitrary bodies.
EDIT: Added link to Torben Mogensen's excellent planet program. Just click on the word "planet" above.
Or just click on the link below:
http://www.diku.dk/users/torbenm/planet.zip
Building a Slushfolder
I only have one program(Fractal Worldmap Generator) that generates planet maps to a pre-determined hydrosphere rating; and that, fairly inaccurately. All my other programs(planet, Lunar Cell, etc.), generate maps with random fractions of ocean. Therefore, in order to get planets with specific hydrosphere ratings, I need to create an extensive "slushfolder" of planet maps. Hopefully, this will be of help to other worldbuilders out there. Also, hopefully, someone might figure out how to do this procedure in batch, and let me know about it
.
The Procedure
Part I: File Management
This part is fairly simple. I create a main folder. In my home computer, it's imaginatively named, "planets." In that folder is a text file with the name and calculated hydrosphere percentage of each planet mapped. There are also a series of folders with the names of the planet maps they contain, each dedicated to all the files associated with a specific planet.
That's all. Not really exciting so far.
Part II: The Map
This is the meat. For the example, I'm using photoshop and lunarcell. These aren't critical, but lunarcell removes the requirement for a bit of magicwand-fu.
II-A) Da Files
First create the planet folder. In this example, I'll name mine, "Vogel."
In that folder, create a new photoshop document. In this case, "Vogel.psd" The image size should be twice as wide as it is high. Again, in the example, I use 1280x640, with a white background. I like 'em big, and I gots the space. "Now save it. Save early, save often," I'd save myself a lot of problems if I followed that dictum more reliably
.
II-B) Create a Mollweide Field
Now, with your clean white page, go to (Filter>Flaming Pear> Flexify... or Filter>Flaming Pear>Flexify II... if you got it ). Set, "Input," to, "equirectangular," and, "Output," to, "Mollweide." Hit, "OK," and you get a nice white ellipse. Use, "magic wand," to select the white area. Now go to the menu(Select>Save Selection...). Name your selection, "Mollweide Field."
Look at your histogram(Image>Histogram...). The fourth field down on the left says, "Pixels," this is the number of pixels in your current selection, in my case it was 641,661. I usually write this down. You can unselect now.
We now have a saved selection, so we can do the next part right on top of our, oh-so-lovely picture.
II-C)Generate a planet
First go to the Filter menu (Filter>Flaming Pear>LunarCell...) to load up Lunar Cell.
Set the mode to, "map color." Play with the settings till you get something you like. I'd strongly suggest unchecking, "Real Luna," and, "Lunar climate." Save your settings in the same folder as your psd. In this example, ".../planets/Vogel/VogelLunarCell." Usually it's best to set, "Ice," to zero, at least for this part.
Once you have something you like, save it. I'll use, "VogelLunarCell." Then you can hit, "OK." Now it renders. Pretty...
Next part is to make an equal-are version of the map. Duplicate the layer(Layer>Duplicate Layer...). Clever, huh?
Now make it equal-area. Go to your version of flexify; the settings should be the same as before{equirectangular->Mollweide}. Play with your longitude slider if you want. Probably leave Latitude and Spin alone, otherwise, if you ever decide to make a version with ice caps, you might find them on the equator. Save your settings in the same folder as your other stuff. I saved mine as, "VogelFlexify 2."
II-D)Determine hydrosphere.
Here is where our friend, histogram starts to play.
Go to the Channels palette(Window>Channels. You'll see channels named, "RGB," "Red," "Green," "Blue," and, hopefully, "Mollweide Field." Create a new one([Little triangle in the upper right corner of the palette]>New Channel...), make the color black, choose, "Selected Areas," and name the channel, "Oceans."
Now go to Lunar Cell. Set the mode to, "ocean mask," and render.
Now go to your version of Flexify. Change the background color to white, and render. Select the, "RGB," channel.
Load Selection(Select>Load Selection...) from the, "Oceans," channel.
Now look at the histogram(Image>Histogram...), note the number of pixels, in my case I got 469,919. YMMV.
From this number and the number of pixels determined for, "Mollweide Field," (I will use that name for a starport someday.) we can determine the hydrosphere for our new planet.
(|Oceans| / |Mollweide Field|) * 100
I got 73.2%, again your mileage will likely vary. I create a new line in the text file I had made in my main folder. It will have the name of my planet and the hydrosphere rating I calculated.
"Vogel: 73.2% hydrosphere"
I prefer to sort these entries from least hydrosphere to greatest, in order to facilitate later reference.
Part III: Using the Slushfolder
I will go through the procedure I have detailed many times till I have a large and extensive slushfolder. Now it's time to start creating planets.
I'll use GT: First In to detail a new planet. It turns out to be an earthlike world with a diameter of 6,800 miles and a hydrosphere rating of 73%.
Well, now I examine my Hydrospheres text file, and look... here's Vogel with a hydrosphere of 73.2%, excellent! I'll put an asterisk next to that entry to denote that this map has been used.
I go to the Vogel folder and there's the map for my new planet. Yay!
[EDIT="I skipped a step"]
III-B: Surface areas
First determine the surface area of your planet(4*pi*radius^2).
For this planet, with a radius of 3,400 miles, we come up with an area of ~145,267,244 square miles.
To determine the area represented by each pixel, we divide the area into the number of pixels as determined by the mollweide field histogram: in this case, 641,661. So we get that each pixel represents ~226 square miles. Planets are big.
The pixels are small enough that I can assume they are on a flat surface with only a small loss of accuracy. So lets take the square root of the pixel are to get the approximate size of each pixel. That gives us squares about 15 miles on a side. This is adequate for short trips, but you can expect distances to be fabulously inaccurate over long trips.
[/EDIT]
Part IV: Notes
This procedure obviously works best once you have attained a large collection of maps. This can be done at leisure. With practice this can be done quite quickly. It took me about two hours to do this example, but most of that time was spent on documenting the procedure for yall's benefit. After doing scores of these things, Ican do it in less than an hour. Yeah, I'm a slow typist .
I hope this will prove of use to you.
Thank you, my friends, for your attention,
Colin
I only have one program(Fractal Worldmap Generator) that generates planet maps to a pre-determined hydrosphere rating; and that, fairly inaccurately. All my other programs(planet, Lunar Cell, etc.), generate maps with random fractions of ocean. Therefore, in order to get planets with specific hydrosphere ratings, I need to create an extensive "slushfolder" of planet maps. Hopefully, this will be of help to other worldbuilders out there. Also, hopefully, someone might figure out how to do this procedure in batch, and let me know about it
The Procedure
Part I: File Management
This part is fairly simple. I create a main folder. In my home computer, it's imaginatively named, "planets." In that folder is a text file with the name and calculated hydrosphere percentage of each planet mapped. There are also a series of folders with the names of the planet maps they contain, each dedicated to all the files associated with a specific planet.
That's all. Not really exciting so far.
Part II: The Map
This is the meat. For the example, I'm using photoshop and lunarcell. These aren't critical, but lunarcell removes the requirement for a bit of magicwand-fu.
II-A) Da Files
First create the planet folder. In this example, I'll name mine, "Vogel."
In that folder, create a new photoshop document. In this case, "Vogel.psd" The image size should be twice as wide as it is high. Again, in the example, I use 1280x640, with a white background. I like 'em big, and I gots the space. "Now save it. Save early, save often," I'd save myself a lot of problems if I followed that dictum more reliably
II-B) Create a Mollweide Field
Now, with your clean white page, go to (Filter>Flaming Pear> Flexify... or Filter>Flaming Pear>Flexify II... if you got it ). Set, "Input," to, "equirectangular," and, "Output," to, "Mollweide." Hit, "OK," and you get a nice white ellipse. Use, "magic wand," to select the white area. Now go to the menu(Select>Save Selection...). Name your selection, "Mollweide Field."
Look at your histogram(Image>Histogram...). The fourth field down on the left says, "Pixels," this is the number of pixels in your current selection, in my case it was 641,661. I usually write this down. You can unselect now.
We now have a saved selection, so we can do the next part right on top of our, oh-so-lovely picture.
II-C)Generate a planet
First go to the Filter menu (Filter>Flaming Pear>LunarCell...) to load up Lunar Cell.
Set the mode to, "map color." Play with the settings till you get something you like. I'd strongly suggest unchecking, "Real Luna," and, "Lunar climate." Save your settings in the same folder as your psd. In this example, ".../planets/Vogel/VogelLunarCell." Usually it's best to set, "Ice," to zero, at least for this part.
Once you have something you like, save it. I'll use, "VogelLunarCell." Then you can hit, "OK." Now it renders. Pretty...
Next part is to make an equal-are version of the map. Duplicate the layer(Layer>Duplicate Layer...). Clever, huh?
Now make it equal-area. Go to your version of flexify; the settings should be the same as before{equirectangular->Mollweide}. Play with your longitude slider if you want. Probably leave Latitude and Spin alone, otherwise, if you ever decide to make a version with ice caps, you might find them on the equator. Save your settings in the same folder as your other stuff. I saved mine as, "VogelFlexify 2."
II-D)Determine hydrosphere.
Here is where our friend, histogram starts to play.
Go to the Channels palette(Window>Channels. You'll see channels named, "RGB," "Red," "Green," "Blue," and, hopefully, "Mollweide Field." Create a new one([Little triangle in the upper right corner of the palette]>New Channel...), make the color black, choose, "Selected Areas," and name the channel, "Oceans."
Now go to Lunar Cell. Set the mode to, "ocean mask," and render.
Now go to your version of Flexify. Change the background color to white, and render. Select the, "RGB," channel.
Load Selection(Select>Load Selection...) from the, "Oceans," channel.
Now look at the histogram(Image>Histogram...), note the number of pixels, in my case I got 469,919. YMMV.
From this number and the number of pixels determined for, "Mollweide Field," (I will use that name for a starport someday.) we can determine the hydrosphere for our new planet.
(|Oceans| / |Mollweide Field|) * 100
I got 73.2%, again your mileage will likely vary. I create a new line in the text file I had made in my main folder. It will have the name of my planet and the hydrosphere rating I calculated.
"Vogel: 73.2% hydrosphere"
I prefer to sort these entries from least hydrosphere to greatest, in order to facilitate later reference.
Part III: Using the Slushfolder
I will go through the procedure I have detailed many times till I have a large and extensive slushfolder. Now it's time to start creating planets.
I'll use GT: First In to detail a new planet. It turns out to be an earthlike world with a diameter of 6,800 miles and a hydrosphere rating of 73%.
Well, now I examine my Hydrospheres text file, and look... here's Vogel with a hydrosphere of 73.2%, excellent! I'll put an asterisk next to that entry to denote that this map has been used.
I go to the Vogel folder and there's the map for my new planet. Yay!
[EDIT="I skipped a step"]
III-B: Surface areas
First determine the surface area of your planet(4*pi*radius^2).
For this planet, with a radius of 3,400 miles, we come up with an area of ~145,267,244 square miles.
To determine the area represented by each pixel, we divide the area into the number of pixels as determined by the mollweide field histogram: in this case, 641,661. So we get that each pixel represents ~226 square miles. Planets are big.
The pixels are small enough that I can assume they are on a flat surface with only a small loss of accuracy. So lets take the square root of the pixel are to get the approximate size of each pixel. That gives us squares about 15 miles on a side. This is adequate for short trips, but you can expect distances to be fabulously inaccurate over long trips.
[/EDIT]
Part IV: Notes
This procedure obviously works best once you have attained a large collection of maps. This can be done at leisure. With practice this can be done quite quickly. It took me about two hours to do this example, but most of that time was spent on documenting the procedure for yall's benefit. After doing scores of these things, Ican do it in less than an hour. Yeah, I'm a slow typist
.I hope this will prove of use to you.
Thank you, my friends, for your attention,
Colin
Colin: great instrux. As an aside you can adjust the percentage of ocean in LunarCell to customize a world as you mention. In addition, you can set its atmo %, cloud % and color (to represent tainted worlds, etc.) and you can vary the world's diameter as well.
Best of all you can save those settings to a file, so you can tweak 'em later and regenerate whatever images you want. It's worked very well for me, but the next step is to automate the whole thing - to write a script that analyzes a world's UPP and generates planetary imagery hands-off, in GIMP with LunarCell.
That will enable my 'LBB next generation' effort, to have a page or two for each mainworld in a subsector including graphics, etc. to accompany the existing, text-only per-world breakdowns.
Best of all you can save those settings to a file, so you can tweak 'em later and regenerate whatever images you want. It's worked very well for me, but the next step is to automate the whole thing - to write a script that analyzes a world's UPP and generates planetary imagery hands-off, in GIMP with LunarCell.
That will enable my 'LBB next generation' effort, to have a page or two for each mainworld in a subsector including graphics, etc. to accompany the existing, text-only per-world breakdowns.
True about the adjustments, but they aren't really setting the actual percentage of ocean, etc. They appear to adjust slightly more esoteric variables such as the threshold value that is considered to be ocean, or the basic steepness, etc.
I can reseed ten times with the same Ocean value and get ten different hydrosphere percentages. The terminology is a bit of a trap. That said, I can certainly tweak those values to increase or decrease the ocean percentage till I get what I want. But it is hit or miss.
So, I just generate a great mass of maps, record the hydrosphere ratings, and then drag 'em out as I need 'em.
I can reseed ten times with the same Ocean value and get ten different hydrosphere percentages. The terminology is a bit of a trap. That said, I can certainly tweak those values to increase or decrease the ocean percentage till I get what I want. But it is hit or miss.
So, I just generate a great mass of maps, record the hydrosphere ratings, and then drag 'em out as I need 'em.
Then I must respectfully disagree with your assertion that "the adjustments aren't really setting the actual percentage of ocean, etc. They appear to adjust slightly more esoteric variables such as the threshold value that is considered to be ocean, or the basic steepness, etc."
Sea level = % of ocean. 0=no ocean. 99=all ocean.
Sea level = % of ocean. 0=no ocean. 99=all ocean.
I agree on one point. Sea level is not esoteric. It apparently sets the threshold altitude below which water exists. Cool.
On the other hand, I respectfully disagree with your assertion that Sea Level = % of ocean. There is an obvious correlation: higher sea level -> more ocean. 0=very little ocean. 99=very much ocean(usually, all ocean). However, 50 != 50% ocean. 50 is usually about 1/2 ocean, but it does vary fairly widely.
That is my only dispute. I apologise if I am causing consternation.
Thank you for your kind attention and interest.
On the other hand, I respectfully disagree with your assertion that Sea Level = % of ocean. There is an obvious correlation: higher sea level -> more ocean. 0=very little ocean. 99=very much ocean(usually, all ocean). However, 50 != 50% ocean. 50 is usually about 1/2 ocean, but it does vary fairly widely.
That is my only dispute. I apologise if I am causing consternation.
Thank you for your kind attention and interest.
Alik: I believe you are wrong, but frankly I don't want to muddy this thread with detailed dissection of your argument. I think this thread should focus on practical world map construction.
I'm fully prepared to continue the discussion, so we can take it offline if you'd like, or you might want to create a thread with a discussion of your formulae, etc. for those who are interested.
I'm fully prepared to continue the discussion, so we can take it offline if you'd like, or you might want to create a thread with a discussion of your formulae, etc. for those who are interested.
Similar threads
