Kohlenklau Test Dummy Log for learning Butschi's New Map Elevation Tool

[Aragorn2002]

14 hours ago, kohlenklau said:

oh dang, I wish I was NOT scratching my head. Hair is gone. Scalp skin gone. Cranium bone almost gone. I better stop before I hit brain tissue...or a hollow void?

 

 

Well, I'm closely following the progress, but I will wait with experimenting for a while, I think, until Butschi and you are completly done. But what an interesting development!

I'm also re-reading Jon Sowden's superb Scenario Design DAR/AAR. @JonSGreat job, especially the part on map making.

Edited December 16, 2022 by Aragorn2002

[Butschi]

3 hours ago, Aragorn2002 said:

Well, I'm closely following the progress, but I will wait with experimenting for a while, I think, until Butschi and you are completly done. But what an interesting development!

I'd ask you to get involved as soon as possible because "completely done" will never happen.

Maybe I should have made this clearer: From my point of view this tool (at least the elevations part) is done already and has been for a while. It does everything it is supposed to do. I'm perfectly willing to help and improve things to make it easier for you guys. But I can only do that if I get feedback.

In your case, everything is there already, btw, there is nothing to wait for.

 

[Aragorn2002]

3 hours ago, Butschi said:

I'd ask you to get involved as soon as possible because "completely done" will never happen.

Maybe I should have made this clearer: From my point of view this tool (at least the elevations part) is done already and has been for a while. It does everything it is supposed to do. I'm perfectly willing to help and improve things to make it easier for you guys. But I can only do that if I get feedback.

In your case, everything is there already, btw, there is nothing to wait for.

 

Ha! Well, in that case there's no time to lose. I'm actually glad to hear that, because I want to begin with my map, after doing a lot of research in the past year and I will have some time in the coming weeks. I must start with the elevation map, especially since the landscape around Goldap is remarkably hilly. I knew there is quite a big hill to the south of Goldap, but there are many more smaller hills as well.

Edited December 16, 2022 by Aragorn2002

[Aragorn2002]

Too optimistic, as always. It turns out I don't know the first think about map making (overlays and stuff), so I've spend most of the day reading instructions and experimenting. Time well used of course, but rather overwhelming and frustrating to realize one is below the dummy level. Anyway, enough for today. This is going to take weeks, if not months to master for me. I'll get there though.

Edited December 17, 2022 by Aragorn2002

[Combatintman]

1 hour ago, Aragorn2002 said:

Too optimistic, as always. It turns out I don't know the first think about map making (overlays and stuff), so I've spend most of the day reading instructions and experimenting. Time well used of course, but rather overwhelming and frustrating to realize one is below the dummy level. Anyway, enough for today. This is going to take weeks, if not months to master for me. I'll get there though.

Of the two that I know you're looking at - I'd do Schirwindt first.  As someone who's made a map or two, I don't fancy Goldap at all.  You've got a whole bunch of closely spaced contours in your built up area plus what at a very quick look appear to be some quite tricky changes of elevation to get right.  Your problem there - even if you use the elevation script is doorways and windows sunk into the ground in the town and all sorts of other shenanigans.  Also, I find built-up areas quite a chore to do and Goldap has more of that type of terrain than Schirwindt.  Schirwindt, elevation-wise is a doddle by comparison and is a neat and compact town.

I'm sure you've sort of looked at this already but here are the two side-by-side to illustrate the ugghhhh factor for Goldap vs Schirwindt:

If I was to do Schirwindt, I'd rotate it like this so that you've got the main streets in the town in vaguely straight lines:

Edited December 17, 2022 by Combatintman

[Butschi]

57 minutes ago, Combatintman said:

Of the two that I know you're looking at - I'd do Schirwindt first.  As someone who's made a map or two, I don't fancy Goldap at all.  You've got a whole bunch of closely spaced contours in your built up area plus what at a very quick look appear to be some quite tricky changes of elevation to get right.  Your problem there - even if you use the elevation script is doorways and windows sunk into the ground in the town and all sorts of other shenanigans.  Also, I find built-up areas quite a chore to do and Goldap has more of that type of terrain than Schirwindt.  Schirwindt, elevation-wise is a doddle by comparison and is a neat and compact town.

I'm sure you've sort of looked at this already but here are the two side-by-side to illustrate the ugghhhh factor for Goldap vs Schirwindt:

If I was to do Schirwindt, I'd rotate it like this so that you've got the main streets in the town in vaguely straight lines:

Well, doing the elevations for both maps doesn't cost much, just a few hours of CPU time. So from that point of view I'd give it a go, just out of curiosity.

The built up area chore is a thing, though. I mean, the-currently-still-not-quite-ready-enough new version of CMAutoEditor will do buildings for you but only if they are on the map. With aerial imagery it is easy enough to "paint" the buildings onto a map but if you only have a topographic map without any building outlines you'll probably end up placing them yourself in the Scenario Editor.

I also have to make a note to prioritize adding map rotation as a feature...

[Aragorn2002]

5 hours ago, Combatintman said:

Of the two that I know you're looking at - I'd do Schirwindt first.  As someone who's made a map or two, I don't fancy Goldap at all.  You've got a whole bunch of closely spaced contours in your built up area plus what at a very quick look appear to be some quite tricky changes of elevation to get right.  Your problem there - even if you use the elevation script is doorways and windows sunk into the ground in the town and all sorts of other shenanigans.  Also, I find built-up areas quite a chore to do and Goldap has more of that type of terrain than Schirwindt.  Schirwindt, elevation-wise is a doddle by comparison and is a neat and compact town.

I'm sure you've sort of looked at this already but here are the two side-by-side to illustrate the ugghhhh factor for Goldap vs Schirwindt:

If I was to do Schirwindt, I'd rotate it like this so that you've got the main streets in the town in vaguely straight lines:

Sound advise, as always. I must admit I realize that too.  Goldap is a nightmare to make a map of. Another problem with Goldap (but also Schirwindt) is that the town is in the middle of the fight, but most of the fighting took place to the North and to the South of the town, because the Germans tried to encircle the town, with 5th Panzer as the Northern pincer and the 50th Infantry as the Southern pincer. So the map will have to be 4 by 5 kilometres or so. Or make two seperate battles. 

Edited December 17, 2022 by Aragorn2002

[Aragorn2002]

4 hours ago, Butschi said:

Well, doing the elevations for both maps doesn't cost much, just a few hours of CPU time. So from that point of view I'd give it a go, just out of curiosity.

The built up area chore is a thing, though. I mean, the-currently-still-not-quite-ready-enough new version of CMAutoEditor will do buildings for you but only if they are on the map. With aerial imagery it is easy enough to "paint" the buildings onto a map but if you only have a topographic map without any building outlines you'll probably end up placing them yourself in the Scenario Editor.

I also have to make a note to prioritize adding map rotation as a feature...

Buildings I actually prefer to place myself. Elevation, roads and rivers are another matter.

[Combatintman]

8 hours ago, Butschi said:

The built up area chore is a thing, though. I mean, the-currently-still-not-quite-ready-enough new version of CMAutoEditor will do buildings for you but only if they are on the map. With aerial imagery it is easy enough to "paint" the buildings onto a map but if you only have a topographic map without any building outlines you'll probably end up placing them yourself in the Scenario Editor.

I also have to make a note to prioritize adding map rotation as a feature...

Therein lies the problem with Schirwindt - this is what it looks like 70-odd years later:

However, please don't take that as criticism of your efforts - anything that gets more people into the scenario game is always supported by me.

[kohlenklau]

I always have advice that is worth 2 eurocents but it is what it is.

To all scenario dreamers, resist the urge to go "grand and big" and instead try to actually finish a very small scenario, some flavorful small aspect of the overall operation you are enamored with.

Not everybody is a GeorgeMC. Just try and zoom in to some small tactical encounter with a few tanks and some "accomplishable" PORTION of these towns.

[Aragorn2002]

 

3 hours ago, kohlenklau said:

I always have advice that is worth 2 eurocents but it is what it is.

To all scenario dreamers, resist the urge to go "grand and big" and instead try to actually finish a very small scenario, some flavorful small aspect of the overall operation you are enamored with.

Not everybody is a GeorgeMC. Just try and zoom in to some small tactical encounter with a few tanks and some "accomplishable" PORTION of these towns.

 

5 hours ago, Combatintman said:

Therein lies the problem with Schirwindt - this is what it looks like 70-odd years later:

However, please don't take that as criticism of your efforts - anything that gets more people into the scenario game is always supported by me.

All true and wise. But the fact is that the number of released scenarios (for all CM games) is getting less and less. Personally I think the reason for that is that making a scenario is an exhausting process. And map making is the most exhausting part. Against the time the map is finished most scenario makers are suffering from a CM burn out. I know, because I have made a lot of rough and ready maps/scenarios for personal use, but never came close to releasing one of them, also because the attention and appreciation for newly released scenarios is limited. At the same time I enjoyed making and playing my own scenarios tremendously, sometimes even more than the most sophisticated and detailed scenarios from the small number of scenario masters of this forum (for which I feel nothing but admiration and gratitude btw).

I still think the only way to get more scenarios is to join forces. Hence my request to map makers who enjoy making maps, to help me with elevation maps of Goldap and Schirwindt. A map is a funny thing. It's like a painting, which can easily be 'overdone'. Layer after layer of paint and spending far too much time, until the joy is gone and it starts to feel like work, not hobby. My ambitions are far too large for my limited skills. The closest I ever came to a finished scenario was build on an elevation map which Kohlenklau/Phil made for me years and years ago. Still grateful for that, Phil.

Anyway, the only way to get more people into scenario making is collecting a 'Kampfgruppe' of people who enjoy research, making maps, experimenting with AI plans or another part of the process to produce a good scenario. Personally I will probably struggle on, not talented enough, a scenario dreamer indeed, but full of ambition. Goldap, Schirwindt, Lauban, Bautzen. Still so much interesting battles that haven't been covered enough. 

I'm delighted to see gifted modders like JM, NPye, Lucky Strike, RockinHarry and other mod magicians join forces and stimulating each other to perform miracles at the moment, not to mention the people taking part in this thread, who are simply brilliant and very helpful. The same would be possible with regard to scenario development.

Personally I will soldier on and I will get there eventually. And I know I have to learn myself, because we all have things to do. I will also learn to master this great elevation tool.

 

Edited December 18, 2022 by Aragorn2002
Far too long

[Butschi]

16 hours ago, Combatintman said:

However, please don't take that as criticism of your efforts

No worries. If I was able to write magic code I'd not bother with this tool. I'd get filthy rich, buy Battlefront and make them add all those features into the Scenario Editor.

Alas, I'm stuck with writing ordinary code that has a lot of limitations.

What I will do, once I find the time, is make a tutorial about how to make an OSM like map from old topographic maps or something and feed them into the tool.

[kohlenklau]

20 hours ago, Aragorn2002 said:

Anyway, the only way to get more people into scenario making is collecting a 'Kampfgruppe' of people who enjoy research, making maps, experimenting with AI plans or another part of the process to produce a good scenario.

Your new idea is my old idea, and apparently the idea of several other folks. I do hold fast to the desire to keep them from "snowballing down the hill of grandiosity"...getting bigger and eventually too big to finish.

 

 

Edited December 19, 2022 by kohlenklau

[kevinkin]

Looks like the CMAutoEditor produces wonderful looking topography. Quick Question. Does the game engine adjust for the high granularity produced in terms on LOS and hull down status? Does the answer involve some controlled comparison testing? New to the thread but it's very interesting. 

[Butschi]

3 hours ago, kevinkin said:

Looks like the CMAutoEditor produces wonderful looking topography. Quick Question. Does the game engine adjust for the high granularity produced in terms on LOS and hull down status? Does the answer involve some controlled comparison testing? New to the thread but it's very interesting. 

CMAutoEditor does nothing you couldn't do in the Editor yourself. So, while I didn't test it, I don't see why it should have any effect on LOS and the like. I mean, there is an elevation value for each square whether you change it out not, so in principle the engine would have to do the calculations anyway. That said, I don't know if the engine takes shortcuts to speed up calculations. But if so that would affect "normal" maps, too.

[kevinkin]

Got it, I think, thanks. Your software more precisely places the elevation numbers so instead of tracing a 20 to a 23 level change by hand/eye, the software places a 20 to 21 level change which produces a smoother result especially when the effect is map-wide. Nice. 

[Butschi]

1 hour ago, kevinkin said:

Got it, I think, thanks. Your software more precisely places the elevation numbers so instead of tracing a 20 to a 23 level change by hand/eye, the software places a 20 to 21 level change which produces a smoother result especially when the effect is map-wide. Nice. 

Actually it's really just a way to get external elevation data into the editor. You get the detailed relief by using high precision lidar data where you have one datum for each square on the map.

It should also be noted that on a sub-meter scale the terrain may be less smooth this way because the editor seems to have floating point precision interpolation where the user can only set integer values.

[kohlenklau]

I am done installing 20 roof trusses and 30 sheets of 3/4" plywood so I can return to this effort and see if I can "narrate by text" to show how to use Butschi's CMAutoEditor tool.

Case 1 was my Buda map so I will go there first...

~~~~~~~~~~

rule 1, TRY to actually read and heed what Butschi says. Guilty before... I didn't recall his clear guidance on finding the EPSG number.
 

Quote

 

geotiff2cm.py:

--input-file: Select input geotiff
--output-file: Set name of output file (will have csv format)
--bounding-box: Set a box in which data is extracted. Some geotiffs are really large and you should always set a box. Usage: EPSG no., left, bottom, right, top (e.g. 23700, 647058.157, 240190.841, 647609.303, 240446.250)

Note: For epsg numbers see e.g. here: https://epsg.io/
They are used to specify the meaning of the coordinates of the box, i.e. the coordinate reference system.

 

ok, here is my lower left corner for my Buda bounding box...

19.013407 47.507867

and my right corner

19.013407 47.507867

well, it was the same so the page is odd to try and use

another thing to try and figure out

Edited December 19, 2022 by kohlenklau

[kohlenklau]

I guess I misunderstood and was in the wrong part of the page up at the top "Get position on a map" trying to use the page to find my 2 points

I zoomed out and now I see lower on the page to get the epsg numbers. I think

 

[kohlenklau]

I will pick my 2 points using Google Earth as I am used to that!

GE uses WGS84 and the epsg code for WGS84 is 4326

that puzzle solved for case 1...

my python line will use these numbers

4326 47.504545  19.008712 47.508774  19.017503

whoops I must switch

This was Butschi's example.

python geotiff2cm.py --input-file merged_ddm.tif --bounding-box 4326, 19.008865, 47.504562, 19.016790, 47.508557 --output-file buda.csv

So I do

python geotiff2cm.py --input-file merged_ddm.tif --bounding-box 4326, 19.008712, 47.504545, 19.017503, 47.508774 --output-file buda.csv

My 2 specific points for the bounding box were the ones I wanted for my scenario. Butschi had just roughed it when he sent me the "gift"...but he was damn close!

 

 

 

Edited December 19, 2022 by kohlenklau

[Butschi]

1 minute ago, kohlenklau said:

I will pick my 2 points using Google Earth as I am used to that!

GE uses WGS84 and the epsg code for WGS84 is 4326

that puzzle solved for case 1...

 

 

My mistake, I guess. Added the link to epsg.io just so you could read up on those obscure numbers if you are interested.

Using WGS84 and getting the points from Google is perfectly ok. Just keep in mind: the orientation of the coordinates axes is usually slightly different in longitude/latitude than in the a projected coordinate system. So the rectangle selected by the two points is a little different after projection. For this reason you also didn't go too close to the edges of a geo-tiff. You will have a small region with invalid values there.

Note to self: make the tool write out a picture with the actual selection...

[kohlenklau]

1 hour ago, Butschi said:

My mistake, I guess.

I wouldn't say that at all! I am just showing my floundering for all to see and hopefully at the end, I have it figured out enough to create a bit of a guide for myself for later when I forget it all! AND for others who might want to try it

 

Edited December 19, 2022 by kohlenklau

[kohlenklau]

I guess this is like being in a band and practicing a song? "Take it from the top boys!"

Case 1 is my Buda map using Google Earth coordinates for the TEM data that Butschi bought me from Hungary. 1 square kilometer and I just want for this scenario a certain rectangle.

I have it all loaded so I will go open my command window AND CMRT scenario editor and give it a try.

OH, HOLD ON, this is a new step to make my csv for the actual map making. Got it!

Standby!

EDIT: whoops, I got to load the new requirements. hahaha

look at the date on the requirements.txt file in the folder and make sure it is the NEW one, hahaha

python geotiff2cm.py --input-file merged_ddm.tif --bounding-box 4326, 19.008712, 47.504545, 19.017503, 47.508774 --output-file buda.csv

EDIT: I must not have all the latest version of files. I get a red error in the python window.

 

 

Edited December 19, 2022 by kohlenklau

[Butschi]

51 minutes ago, kohlenklau said:

EDIT: I must not have all the latest version of files. I get a red error in the python window.

Or it's an actual bug. Do you have a screenshot?

[kohlenklau]

26 minutes ago, Butschi said:

Do you have a screenshot?

Edited December 19, 2022 by kohlenklau