Could you reexplain how to create a map?
In lay-mens terms. :tongue.gif:
I have python, got the download, i click what, in what order?
Assuming i want to map the entire island of Kauai and its western neighbor?
(farthest west hawaiin island)
ie
160-159 10' w
22 14' - 21 40' N
Wow, how lame. I just discovered that Hawaii's land cover information is stored differently so I have to modify my code. Pick somewhere on the mainland for now? Also, pick a smaller region, as this one is as big as Cape Cod and that ended up crashing my computer after eating all my RAM. Heh. :-)
Hello. I have been trying to use this package for a few days now. My first git clone was after your update on the 12th. I managed to acquire a dataset, built the images and world, and then copied the world data into my saves folder. I checked to make sure that there was a level.dat directly under the directory I copied out of my topomc/Worlds folder.
However, the game seemed to overwrite the two different datasets I tested. It showed a valid world, but upon choosing it it gave a "Building Terrain" message and after loading I couldn't see any resemblance. They were only 9 tiles each, is there a recommended minimum?
I realized the best way to test would be to use the test dataset, so I set about building it. I did a git pull first (about an hour ago), and now I get TypeError: 'int' object is not iterable on line 69 of dataset.py. I can give you a stack trace if that'd be helpful, but I'm unfamiliar with the syntax used on that line (does Python let you automagically assign multiple variables if a function returns a tuple?) and I haven't been able to find a fix.
Hello. I have been trying to use this package for a few days now. My first git clone was after your update on the 12th. I managed to acquire a dataset, built the images and world, and then copied the world data into my saves folder. I checked to make sure that there was a level.dat directly under the directory I copied out of my topomc/Worlds folder.
However, the game seemed to overwrite the two different datasets I tested. It showed a valid world, but upon choosing it it gave a "Building Terrain" message and after loading I couldn't see any resemblance. They were only 9 tiles each, is there a recommended minimum?
I realized the best way to test would be to use the test dataset, so I set about building it. I did a git pull first (about an hour ago), and now I get TypeError: 'int' object is not iterable on line 69 of dataset.py. I can give you a stack trace if that'd be helpful, but I'm unfamiliar with the syntax used on that line (does Python let you automagically assign multiple variables if a function returns a tuple?) and I haven't been able to find a fix.
Thanks for sharing the code in any event.
Heh! You caught me as I fixed the bug. :-) Do another "git pull" and you should be able to do what you will with it. :-)
Hello !
This program sounds great, this is exactly what I was looking for !!!
But I have a problem.... I never use Python before and I have no idea how to launch TopoMC...
Could someone do a tutorial for a Windows user from the begininng please ??
Thanks a lot !
I hope this program will grow up ! :smile.gif:
(sorry for my english, i'm french :tongue.gif:)
I am glad that you see a bright future for TopoMC! Once the software is feature-complete, I plan on writing documentation for new users, including how to install and run the software as well as a list of all dependencies. Stay tuned -- when that documentation is complete, I will post to this thread!
Indeed, everything works correctly now. Thanks again for all of the hard work.
EDIT: Also, if there's a simple answer, what is the effective resolution at the various stages of the processing pipeline?
The land cover data points are spaced 30 meters apart. The elevation data is by degrees, and is roughly 10 meters apart in the case of NED 1/3 arc second (90% of the coterminous US) or 30 meters apart in the case of NED 1 arc second (the entire country, including Hawaii and even parts of Alaska). The elevation data's accuracy is estimated at one meter. I put the elevation data into what's called an "inverse distance tree" which allows me to create a grid of one point every five meters and interpolate elevation values. The land cover uses the majority algorithm to do much the same thing. I find a 1:6 scale to be a nice mix between "bigness" and smoothness of data. It's also nice to have the same scale for the vertical so things look "right", but that's not always the case.
I use the land cover data from the 2001 dataset, which doesn't yet cover the entire country. The 1992 dataset, which does cover the entire country, uses different values so I need to write some code to translate them. When that happens, I should be able to use any part of the United States. There's also some parts of the country with 1/9 arc second coverage for elevation -- that's every 3 meters, and I'm totally excited about that.
Erm.
I activate the python thing, and all it does is exit out, without even allowing mke to type in what i need, the second i click it it exits
Edit:THIS DOESENT SEEM PHYSICALLY POSSIBLE!
I dunno why, it just does.
It doesn't work by clicking on it. You'll have to use the command line.
If you're on Linux, open up an xterm. If you're on Windows, open up cmd.exe or whatever it's called these days.
Change to the directory where the files exist, and type in the commands you see in the README.
If it complains about imports, you have additional modules to install. If it complains about other stuff, first do a "git pull" and try again, next complain here and I'll fix it when I can!
Erm.
I activate the python thing, and all it does is exit out, without even allowing mke to type in what i need, the second i click it it exits
Edit:THIS DOESENT SEEM PHYSICALLY POSSIBLE!
I dunno why, it just does.
It doesn't work by clicking on it. You'll have to use the command line.
If you're on Linux, open up an xterm. If you're on Windows, open up cmd.exe or whatever it's called these days.
Change to the directory where the files exist, and type in the commands you see in the README.
If it complains about imports, you have additional modules to install. If it complains about other stuff, first do a "git pull" and try again, next complain here and I'll fix it when I can!
Jack.
Command.exe---->Then use the commands in the readme?
Erm.
I activate the python thing, and all it does is exit out, without even allowing mke to type in what i need, the second i click it it exits
Edit:THIS DOESENT SEEM PHYSICALLY POSSIBLE!
I dunno why, it just does.
It doesn't work by clicking on it. You'll have to use the command line.
If you're on Linux, open up an xterm. If you're on Windows, open up cmd.exe or whatever it's called these days.
Change to the directory where the files exist, and type in the commands you see in the README.
If it complains about imports, you have additional modules to install. If it complains about other stuff, first do a "git pull" and try again, next complain here and I'll fix it when I can!
Jack.
Command.exe---->Then use the commands in the readme?
Yes. You will probably need "mkdir Worlds" before the BuildWorld.py line. But be warned: you will probably have to install some Python modules and I don't know how to do that on Windows so you'll end up having to search the web to learn how. :-)
It doesn't work by clicking on it. You'll have to use the command line.
If you're on Linux, open up an xterm. If you're on Windows, open up cmd.exe or whatever it's called these days.
Change to the directory where the files exist, and type in the commands you see in the README.
If it complains about imports, you have additional modules to install. If it complains about other stuff, first do a "git pull" and try again, next complain here and I'll fix it when I can!
Jack.
Command.exe---->Then use the commands in the readme?
Yes. You will probably need "mkdir Worlds" before the BuildWorld.py line. But be warned: you will probably have to install some Python modules and I don't know how to do that on Windows so you'll end up having to search the web to learn how. :-)
Jack.
Dot Dot Dot.
Perhaps you could code an exe that does all this?
like, it controls all the others, you just tell it the coords and it does this all without having to manually do one hell of a lot.
BOOOOO LINUX
BOOOOO MACs
(Not to start a flame war, its just that your ways to do things are so much more complicated/easier than windows)
Perhaps you could code an exe that does all this?
like, it controls all the others, you just tell it the coords and it does this all without having to manually do one hell of a lot.
BOOOOO LINUX
BOOOOO MACs
(Not to start a flame war, its just that your ways to do things are so much more complicated/easier than windows)
Alas, making an .exe is beyond my capabilities. I am idly fantasizing about a web service where people go to a website, fill in some fields, click a button, and download a .zip file holding a Minecraft level covering the desired location. Until then, using this software will probably be an intermediate-to-advanced task with some assembly required. :-(
Perhaps you could code an exe that does all this?
like, it controls all the others, you just tell it the coords and it does this all without having to manually do one hell of a lot.
BOOOOO LINUX
BOOOOO MACs
(Not to start a flame war, its just that your ways to do things are so much more complicated/easier than windows)
Alas, making an .exe is beyond my capabilities. I am idly fantasizing about a web service where people go to a website, fill in some fields, click a button, and download a .zip file holding a Minecraft level covering the desired location. Until then, using this software will probably be an intermediate-to-advanced task with some assembly required. :-(
Jack.
Angryish face, as i am unable to use this as it is beyond my simple, humble capabilities.
I am idly fantasizing about a web service where people go to a website, fill in some fields, click a button, and download a .zip file holding a Minecraft level covering the desired location.
That would be great!
---
I tried it but got these errors (on my own dataset and the example). On the next file I was missing a bunch of python modules and when easy install didn't work I gave up.
topomc$ ./WarpElev.sh --region BlockIsland
13793074 is not of the correct type
warning: 13793074.zip appears to use backslashes as path separators
98432704 is not of the correct type
warning: 98432704.zip appears to use backslashes as path separators
./WarpElev.sh: line 135: gdalinfo: command not found
./WarpElev.sh: line 138: gdalwarp: command not found
These error messages mean that you need to delete Datasets/BlockIsland and then re-check it out from github, and you need to install the gdal software -- there are a few places where the Python bindings were insufficient for my needs, so I had to actually make calls to the binaries. Are you on Linux or Mac or something else?
Jack.
(making notes to update the README etc to list all the required software)
Great! I'm glad someone else had this idea, too. Raster elevation data to minecraft. Awesome
The directory format is open enough that someone could make imaginary places such as Narnia or Middle Earth -- in fact, there's already a project to generate digital elevation models for Middle Earth at http://me-dem.org/ so now we just need some landcover and you can go mining with Gimli. :-P :-)
So I stared at this for a few minutes, and realized I have no idea how to use GitHub. Maybe it's late and I suck at directions, but I don't get it... :|
Huh.
Rollback Post to RevisionRollBack
I'm the bard. Always will be. And no one EVER respects the bard.
In lay-mens terms. :tongue.gif:
I have python, got the download, i click what, in what order?
Assuming i want to map the entire island of Kauai and its western neighbor?
(farthest west hawaiin island)
ie
160-159 10' w
22 14' - 21 40' N
Wow, how lame. I just discovered that Hawaii's land cover information is stored differently so I have to modify my code. Pick somewhere on the mainland for now? Also, pick a smaller region, as this one is as big as Cape Cod and that ended up crashing my computer after eating all my RAM. Heh. :-)
Jack.
However, the game seemed to overwrite the two different datasets I tested. It showed a valid world, but upon choosing it it gave a "Building Terrain" message and after loading I couldn't see any resemblance. They were only 9 tiles each, is there a recommended minimum?
I realized the best way to test would be to use the test dataset, so I set about building it. I did a git pull first (about an hour ago), and now I get TypeError: 'int' object is not iterable on line 69 of dataset.py. I can give you a stack trace if that'd be helpful, but I'm unfamiliar with the syntax used on that line (does Python let you automagically assign multiple variables if a function returns a tuple?) and I haven't been able to find a fix.
Thanks for sharing the code in any event.
Heh! You caught me as I fixed the bug. :-) Do another "git pull" and you should be able to do what you will with it. :-)
Jack.
I am glad that you see a bright future for TopoMC! Once the software is feature-complete, I plan on writing documentation for new users, including how to install and run the software as well as a list of all dependencies. Stay tuned -- when that documentation is complete, I will post to this thread!
Jack.
EDIT: Also, if there's a simple answer, what is the effective resolution at the various stages of the processing pipeline?
Using laymen terms, please. :tongue.gif:
The land cover data points are spaced 30 meters apart. The elevation data is by degrees, and is roughly 10 meters apart in the case of NED 1/3 arc second (90% of the coterminous US) or 30 meters apart in the case of NED 1 arc second (the entire country, including Hawaii and even parts of Alaska). The elevation data's accuracy is estimated at one meter. I put the elevation data into what's called an "inverse distance tree" which allows me to create a grid of one point every five meters and interpolate elevation values. The land cover uses the majority algorithm to do much the same thing. I find a 1:6 scale to be a nice mix between "bigness" and smoothness of data. It's also nice to have the same scale for the vertical so things look "right", but that's not always the case.
I use the land cover data from the 2001 dataset, which doesn't yet cover the entire country. The 1992 dataset, which does cover the entire country, uses different values so I need to write some code to translate them. When that happens, I should be able to use any part of the United States. There's also some parts of the country with 1/9 arc second coverage for elevation -- that's every 3 meters, and I'm totally excited about that.
Jack.
I activate the python thing, and all it does is exit out, without even allowing mke to type in what i need, the second i click it it exits
Edit:THIS DOESENT SEEM PHYSICALLY POSSIBLE!
I dunno why, it just does.
It doesn't work by clicking on it. You'll have to use the command line.
If you're on Linux, open up an xterm. If you're on Windows, open up cmd.exe or whatever it's called these days.
Change to the directory where the files exist, and type in the commands you see in the README.
If it complains about imports, you have additional modules to install. If it complains about other stuff, first do a "git pull" and try again, next complain here and I'll fix it when I can!
Jack.
Command.exe---->Then use the commands in the readme?
Yes. You will probably need "mkdir Worlds" before the BuildWorld.py line. But be warned: you will probably have to install some Python modules and I don't know how to do that on Windows so you'll end up having to search the web to learn how. :-)
Jack.
Dot Dot Dot.
Perhaps you could code an exe that does all this?
like, it controls all the others, you just tell it the coords and it does this all without having to manually do one hell of a lot.
BOOOOO LINUX
BOOOOO MACs
(Not to start a flame war, its just that your ways to do things are so much more complicated/easier than windows)
Alas, making an .exe is beyond my capabilities. I am idly fantasizing about a web service where people go to a website, fill in some fields, click a button, and download a .zip file holding a Minecraft level covering the desired location. Until then, using this software will probably be an intermediate-to-advanced task with some assembly required. :-(
Jack.
Angryish face, as i am unable to use this as it is beyond my simple, humble capabilities.
These error messages mean that you need to delete Datasets/BlockIsland and then re-check it out from github, and you need to install the gdal software -- there are a few places where the Python bindings were insufficient for my needs, so I had to actually make calls to the binaries. Are you on Linux or Mac or something else?
Jack.
(making notes to update the README etc to list all the required software)
The directory format is open enough that someone could make imaginary places such as Narnia or Middle Earth -- in fact, there's already a project to generate digital elevation models for Middle Earth at http://me-dem.org/ so now we just need some landcover and you can go mining with Gimli. :-P :-)
Jack.
Huh.