CMx1 to CMx2 Map converter

[Lt Belenko]

Remember the Pyewacket Map Converter Tool. I converted a few CMBO scenarios to CMAK using it.

Is there an equivalent for CMx1 to CMx2?

Could one be developed or are the game engine differences too great?

[ChrisND]

Gonna go out on a limb and say it won't happen. Cmx1's code was not based off of Cmx1, it is entirely new.

[Field Marshal Blücher]

Gonna go out on a limb and say it won't happen. Cmx1's code was not based off of Cmx1, it is entirely new.

Think you mean CMx2 for the first one, mate. Although, all in all, definitely agreed.

[Redwolf]

I would certainly be possible to OCR out the terrain in CMx1 into a XML file (or whatever) and then make a rough first pass at automatically building it in CMx2.

It is a lot of work, though, and any kind of controlling programs by simulating mouse clicks is very fragile. Not most programmers' favorite kind of thing to implement.

Now, if BFC would choose to provide im/export of the map data in any kind of documented format things would be a lot different...

[PaulMG]

Pyewacket's converter took the x,y,z from one file and copied them into the other. The file extensions were different and you couldn't just change the extension to go from one game to the other but he figured out something that worked.

[MikeyD]

Might as well be a Donky Kong to CMx2 map converter for all the similaries the two games have. CM:Normandy isn't even compatible with CM:Afghanistan maps, which is the nearest generation game engine before Normandy.

[ChrisND]

I would certainly be possible to OCR out the terrain in CMx1 into a XML file (or whatever) and then make a rough first pass at automatically building it in CMx2.

It is a lot of work, though, and any kind of controlling programs by simulating mouse clicks is very fragile. Not most programmers' favorite kind of thing to implement.

Now, if BFC would choose to provide im/export of the map data in any kind of documented format things would be a lot different...

I just don't see it saving much work at all for the amount of effort it would entail.

[Redwolf]

I just don't see it saving much work at all for the amount of effort it would entail.

Depends on how many maps you want to convert.

The remote control by mouseclick required to get it into CMx2 is the kicker, though. Very nasty. If that could be solved by either BFC providing im/export or by somebody figuring out enough of the file format that would be good. I could provide a OCR solution to get the map out of CMx1 (assuming that the other converter doesn't already do it, or had snooped out the file format).

[Battlefront.com]

As folks have suspected, there will be no conversion tool for CMx1 to CMx2. Way too much work and the results probably wouldn't be all that great. For one thing CMx1 uses 20x20m single type terrain while CMx2 uses 8x8m multiple type terrain. Even if you simply upped the number of CMx2 Action Spots per CMx1 Tile you'd still have problems because you'd cover less terrain than CMx1 (16x16m vs. 20x20m) or more (24x24m vs. 20x20m).

Steve

[hoolaman]

Having done this a couple of times, I think getting the elevations right is the most time consuming and pertinent part of the task.

An automatic conversion of the terrain itself would look really weird anyway, the terrain elements are not exactly comparable, and CMBN is capable of making so much more complex and nicer looking maps that swapping like for like would be quite silly. The grids actually do align over 40 metres or 5x5 CMx2 tiles = 2x2 CMx1 tiles but nothing quite lines up perfectly doing it this way.

Achieving a map that plays the same and is recognisable is the main thing IMO. When I did it, I printed the maps from CMBO for quick reference, then put in the elevations first, then roads and walls and rivers and fields. The forests and foliage I put in rough freehand.

[Redwolf]

That doesn't change the fact that a first cut with interpolated titles that are roughly comparable would be great help.

You'd get the elevation, you'd get vegetation in the right place and the roads might or might not be in the right place after interpolation but they'll be there at least. I wouldn't tackle villages but fact is random sprinkling some houses where you had some randomly sprinkled houses before is better than flat, empty space.

Panzer Command seems to have no problem providing im/export for an external ASCII based format. The result was that an editor that is very powerful (ever wanted to drag things in CM map making) and can pull in data from public map sources (think google maps) appeared within weeks of the release.

[Battlefront.com]

We're not interested in having external Editors. That's not our thing. Never has been, never will be.

Steve

[moneymaxx]

We're not interested in having external Editors.

I do not know the reasons for this position, and i do not need to know. I only wonder why the editor does not allow the import of a basic csv-file (comma seperated value) as Redwolf suggested. The file could be organized like this x-coordinate, y-coordinate, terrain type, height, x-coordinate, y-coordinate, terrain type....... and so on. Even without the terrain-type it would save a lot of work.

[Scipio]

Normandy will have a working QB system. That's much more important (for me) then any map tool.

[MikeyD]

Its not as though CMx1 is a da Vinci drawing in need of loving reproduction. Rebuilding your favorite map from scratch, an afternoon's entertainment. Or a better idea, take a screenshot of a real Google Earth location in Normandy and build that instead. Converting CMx1 maps is like trying to retrofit a Mustang carburator to a Prius. Sometimes nostalgia can go too far.

[Battlefront.com]

Reading in data requires a parser and various error handling routines to rectify problems and/or inconsistencies between the expected syntax and the available in-game options to cross link with. This isn't a quick or easy feature to add. In fact, it has the potential to be quite a hassle. In theory, however, we don't mind the ability to import map data. We just don't have time to implement it.

Steve

[Renaud]

"Rebuilding your favorite map from scratch, an afternoon's entertainment."

Really? If this means that the editor has evolved since CMSF, that is welcome news indeed. I never could get into CMx2 due to the extreme time requirements for building a map of any significant size (my main interest is building historical scenarios - rather a moot point considering the fictional CMSF setting, I know). I started a huge project of exactly recreating the GW1 battle of 73 Easting, but building the map was just impossible. Particularly elevations, which about drove me blind. Just never could get the hang of it. And this is coming from someone that put a lot of time into painstakingly replicating 1:50k topo maps into 3km x 2km CMx1 maps - time consuming, but just barely doable in CMx1. And quite rewarding as the maps really looked great and just exactly like pictures of the real place. Many others have built incredibly good looking and realistic, yet gigantic, cmx1 maps as well. Notably of the Ardennes region.

CMx1 is probably a game people will be playing for another ten more years or more. If an import/export tool had been built for maps and unit data...possibly even far longer. As we know a noble attempt was made to build such a tool, but a really ugly scope creep put that in it's grave, despite heroic efforts. I would have been happy with just a simple import/export gui and a .csv file specification.

A generic import/export tool for map and unit data for CM:BN and the following WW2 modules would be a godsend to campaign managers and scenario designers and extend the longevity of the game perhaps to decades (at least for a small minority of hardcore historical/military simulation freaks). Of course, such longevity is often not a strong point from a business perspective. Also it appears that the CMx2 map specification has changed with most if not every module release...perhaps the unit specification format as well (?). This would require rejiggering the import/export tool and spec for every module.

[ChrisND]

Really? If this means that the editor has evolved since CMSF, that is welcome news indeed. I never could get into CMx2 due to the extreme time requirements for building a map of any significant size (my main interest is building historical scenarios - rather a moot point considering the fictional CMSF setting, I know). I started a huge project of exactly recreating the GW1 battle of 73 Easting, but building the map was just impossible. Particularly elevations, which about drove me blind. Just never could get the hang of it. And this is coming from someone that put a lot of time into painstakingly replicating 1:50k topo maps into 3km x 2km CMx1 maps - time consuming, but just barely doable in CMx1. And quite rewarding as the maps really looked great and just exactly like pictures of the real place. Many others have built incredibly good looking and realistic, yet gigantic, cmx1 maps as well. Notably of the Ardennes region.

The editor is definitely time-consuming, but it's really not that bad if you limit yourself to a CMx1 level of detail. Elevations are tricky though and take some time to wrap your head around, however it's easier if you treat them like contour lines. However, it sounds like you are into making gigantic maps. Not "significant", gigantic. No editor no matter how easy to use is going to save you from a ton of tedious work if you want to make a map that large. It's the trade-off for having higher resolution terrain.

[SlowMotion]

IMO getting elevations right is tricky because you see elevations in 3d preview mode, but edit them using the 2d map editor. No way to change things WYSIWYG.

There was earlier this thread about what kind of things people hoped to see added to Scenario Editor, I don't know if any of them have been added:

http://www.battlefront.com/community/showthread.php?t=90537

[Sergei]

Map data importing capabilities would be nice, although I don't know if eg. elevations data, either from CMx1 or other sources, would be easy to convert to a CMx2 like format. CMx2 elevations allow you to enter contour lines and then the intermediate elevations are automatically calculated, whereas on CMx1 maps each tile has its own hard defined elevation. It would be doable, but would probably require a lot of working on.

Map editor overlays would offer much better overall value, I think. Mapping Mission also had all sorts of import options, but in the end using a topographical map overlay was more efficient.

[Redwolf]

Reading in data requires a parser and various error handling routines to rectify problems and/or inconsistencies between the expected syntax and the available in-game options to cross link with. This isn't a quick or easy feature to add. In fact, it has the potential to be quite a hassle. In theory, however, we don't mind the ability to import map data. We just don't have time to implement it.

Steve

That's why you shouldn't just ignore XML. Don't do an interface that's just a custom binary blob like you tried for CMC, that'll never work.

But with XML you get parser and (more importantly) validation tools. The validation tools can be run outside the game before trying to import and would go a long way of getting things shaped up. Then you run them again inside the program so that you don't have dorks not running it outside crash the parser inside the game.

I know there's some kind of alliance in the wargaming world to just blindly assume XML is... I dunno even what wargaming people think. But they seem to think that all other industries moved to XML for no good reason whatsoever, which is a little stretchy. At least Panzer Command got the idea and it seems that things actually worked for them exactly the way that dorky nerd geeks like me told them it would, I haven't heard of any problems.

And the only editor that Panzer Command has is one that was done without the developer's help just because they had map import - from XML.

[Thomm]

Importing elevation data from an ASCII file sure would be nice.

I programmed a filter to convert SRTM data (Space Shuttle radar elevation data) from a NASA file to a triangle mesh for fluid dynamics analysis in 45 minutes lately. It is not an overwhelming task.

But the problem is that nobody(?) understands how the intermediate elevations are generated in CMx2, and this is crucial for using the editor effectively.

That, and the fact that the import function is not going to be implemented, anyway, or so it seems.

Best regards,

Thomm

[Migo441]

For those of us unfamiliar with the CMx2 map editor? What are we in for?

It makes sense that with smaller tile sizes, everything will be more time consuming. But treating elevation as contour lines makes (that part at least) sound easier. It's probably the fine tuning that is frustrating part, huh?

Map making seems to follow something like the 80/20 rule where you get 80% of the results with 20% of your effort. The great maps require someone obsessed enough to spend the 80% effort for that final 20% of the results which has less to do with effects on actual game play and more to do with the sense of verisimilitude.

That is true in CMx1, but is it more so in CMx2?

[Redwolf]

For those of us unfamiliar with the CMx2 map editor? What are we in for?

It makes sense that with smaller tile sizes, everything will be more time consuming. But treating elevation as contour lines makes (that part at least) sound easier. It's probably the fine tuning that is frustrating part, huh?

Map making seems to follow something like the 80/20 rule where you get 80% of the results with 20% of your effort. The great maps require someone obsessed enough to spend the 80% effort for that final 20% of the results which has less to do with effects on actual game play and more to do with the sense of verisimilitude.

That is true in CMx1, but is it more so in CMx2?

The problem that I see is that although the CMx1 editor was very nice and efficient, and the CMx2 editor is basically the same we still end up in a mess.

The problem is that the "CM editor" (it really doesn't matter whether you talk x1 or x2) lacks any kind of abstraction or composite feature. Just to name the obvious example, you cannot move that village over there 500 meters to the east. There is no cut and paste, there is no drag and drop, there is nothing.

This was kinda OK-ish in CMx1 where maps were less featureful and you had 20x20 tiles. But CMx2 has 8x8 tiles, which means your stupid village has 6.25 tiles the number of tiles. And each tile can be more complicated, so you are looking at essentially 10 times the effort to do something like moving the village.

That is why people like me say that although the CMx1 editor was cool, efficient and mostly adequate it is not adequate in CMx2, even though it's still the same editor.

[Paper Tiger]

For those of us unfamiliar with the CMx2 map editor? What are we in for?

It makes sense that with smaller tile sizes, everything will be more time consuming. But treating elevation as contour lines makes (that part at least) sound easier. It's probably the fine tuning that is frustrating part, huh?

Map making seems to follow something like the 80/20 rule where you get 80% of the results with 20% of your effort. The great maps require someone obsessed enough to spend the 80% effort for that final 20% of the results which has less to do with effects on actual game play and more to do with the sense of verisimilitude.

That is true in CMx1, but is it more so in CMx2?

Sounds like you are familiar with the CMx1 scenario editor? If you were able to produce good maps with it, then you'll be able to producegood maps with the CMx2 editor with only a little bit more work. Designing maps/missions has been the biggest part of my CM experience whether that be x1 or x2. Making a good map has always been a lot of work and the CMx2 editor doesn't have any new tools that make certain aspects of it easier to do.

Your 80/20 rule applies equally to CMx2. You can knock up a functional map very quickly in the editor. It's getting all the finer details in that takes the time. But when I do this, I listen to some Bach, or Mozart, or Pink Floyd and make the experience as pleasurable as possible. In fact I enjoy doing this almost as much as I enjoy playing the game itself, but that's me. The biggest new challenge that you'll face is flavour object placement.