Over the Reich: Single Player - Development Thread

[Civinator]

I have been meaning to build an Me109 for awhile now - when I do, I'll build it in your uncles colors

This great -but he was my great uncle (the brother of my grandfather)! There is a very good description about the coloring of those fighters in the linked article.

 

[JPetroski]

I was playing around with the scenario last night a little bit and if this is going to be the last real stab at it then I want to fix what I've always been annoyed me about this is just how distorted the map is. I took the OTR map, converted it to a bitmap, and squished it, losing the Atlantic which is unnecessary. While it's never going to be perfect with my skill set, what do you all think about the second one? I would need to clean it up quite a bit here and there from the distortion obviously but overall I think it's much better than the former.

Original: 29,784 surface tiles.

New (Proposed): 31,510 surface tiles, and in my mind, a much better approximation.

Option 3 - starting from scratch and not looking at the OTR map. 3,1640 tiles

What do you all think?

 

[tootall_2012]

As you said, when playing with the first map it felt somewhat squished. This one feels much more realistic and looks much better. The only caveat: will its new size significantly increase play time?

 

[JPetroski]

As you said, when playing with the first map it felt somewhat squished. This one feels much more realistic and looks much better. The only caveat: will its new size significantly increase play time?

I don't think so - the MP of units can be adjusted and getting rid of the Battle of the Atlantic sideshow can cut out as much as 20-30 minutes a turn right there. We're also likely going to automate trains and I wouldn't expect to have any more cities or targets, really, than the last version. Just a map that doesn't make me cringe.

I'm kind of leaning towards the 3rd option though @Dadais let me in on some map making secrets over on the discord and maybe I could try a hand at making exactly what I want with a grid instead of this somewhat rudimentary way I've been getting by with for years.

 

[Prof. Garfield]

I think the shape, size, and area of the third map looks the best. I might like a bit more space on the West edge in case the Germans want to fly around and attack Belfast, but that's a minor point.

 

[tootall_2012]

Yes, sorry for not being clear enough. Like the Prof. I think option 3 looks the best.

 

[JPetroski]

I'm going to see if I can build a map that looks like this. Dadais is helping to get me a grid on this so hopefully I can pull it off (it's just online diamond art, right? LOL) . Should pretty much cover everything we need and nothing we don't. I don't think there's much of a need to have all of France - this map pretty much captures everything 8th Air Force would hit and gets us all of Germany sans East Prussia. There should be enough space in the North Sea for bombers to get up to no good without being detected right away as well. Also, it kind of makes places like Berlin a good "end game" that you probably aren't going to be able to touch effectively until the Mustangs come in.

 

[Civinator]

In my eyes map option 3 is the best. In map option 2 some important targets in the east of Germany are missing, when the distances are not distorted too much. I prefer the Gary Grigsby map, that also offers to contain the operations of the 15th U.S.A.A.F.

Spoiler :

 

[techumseh]

Agree, map three. Less is better.

 

[Knighttime]

Hi @JPetroski, my two cents: only the minimap is accurate for representing landform shapes in a way that also honors actual movement points. I think the main map should always look squished because the diamond-shaped tile graphics are twice as wide as they are tall -- but remember it takes 1 MP to move horizontally (64 pixels) or vertically (32 pixels). If you build an England (for example) that looks "right" on the main map, your units will be able to move East-to-West twice as quickly as they can cover an equal number of real-world miles North-to-South (and England will look far too "narrow" on the minimap). When I want to preview a map I'm building, I always drag-to-stretch the minimap until it fills my whole monitor, because that gives me the best info on whether all landforms have the best shape.

If you'd like, I'd be willing to collaborate with you and build a true-to-life map for you, of whatever region you specify. (An example of what I've been able to do in the past is here; the preview image is the minimap.) You're welcome to PM me if you're interested -- but of course it's fine for you to approach this any way you'd like. Have fun!

 

[JPetroski]

Thanks for the kind offer @Knighttime. @Dadais was really helpful in creating a grid over a map for me which makes it mostly "diamond art" if you've ever done that before, so I should be good here.

If either of you wanted to be extremely helpful to the community sometime, an idiot-proof, step by step instruction thread on how to appropriately determine the dimensions of the grid, and create and layer the grid in a paint program would be outstandingly useful. Perhaps the next time one of you makes a map, you'd be willing to document the process?

I know I certainly have other scenarios (Germanicus and Caesar come to mind) where I would like to remake them with better maps than my old efforts! I'd love to get a better handle on how to appropriately size a grid to a picture and overlay it as a layer.

As to "squishiness," well, yeah that's unavoidable as you say, but my main concern with the original map wasn't that it was squishy so much as wildly inaccurate. The cities were nowhere near were they should have been which was frustrating to me. With this grid system I will be able to pick out the city locations, exact flow of rivers, etc. much much easier and accurately. I've just finished outlining the coast for OTR and now I intend to pick out cities with one color and rivers with a third before I then move in to doing hills and everything else. It should produce a much more immersive experience for the player.

This map, btw, is 32,760 tiles which makes it pretty much as large as a Civ2 map can get. I'm very excited about this

 

[JPetroski]

I've "placed" the city markers. We don't necessarily need all of them but they'll be useful guides. I'm going to see if I can't place the airfield markers too at this time.

A question for you @Prof. Garfield:

-Is it going to be possible to automate the trains (and maybe even trucks) by having them only follow paths that have a road or railway improvement on the terrain tile? I'm trying to see if I really need "railtrack" terrain or what can be saved now that Lua has come a long way since we started. Also, can you think of a solution for places like 15th Air Force to give them extra shields in the city for support purposes without needing a special terrain for it? I'd like to make this a nice map with both hills and mountains, and fields and pastures and such. Last go of things we had to make it somewhat sparse because we were using special terrain tiles for cities, airfields, 15th air force, installations, railtrack, etc... I'd rather just use "roads" and "railroads" so I could have the terrain tile for other purposes but wasn't sure if this will work or not.

 

[Prof. Garfield]

-Is it going to be possible to automate the trains (and maybe even trucks) by having them only follow paths that have a road or railway improvement on the terrain tile? I'm trying to see if I really need "railtrack" terrain or what can be saved now that Lua has come a long way since we started.

This is definitely possible if we don't let the player actually move the units or choose the destination. We can just pathfind only on road/rail and teleport the units as far as they can get. If we let the human player move the unit, we can do something similar to Battle of Italy, where an onEnterTile event moves the unit back onto the railroad. We probably don't want the AI to move freight independently, since it won't know what to do with it.

Use rivers and "railtrack" terrain for actual terrain purposes. I'm sure we'll be able to find an acceptable solution for logistics.

Also, can you think of a solution for places like 15th Air Force to give them extra shields in the city for support purposes without needing a special terrain for it?

Yes, this is easy with the onCalculateCityYield execution point.

 

[Knighttime]

... my main concern with the original map wasn't that it was squishy so much as wildly inaccurate. The cities were nowhere near were they should have been which was frustrating to me.

Ah, got it -- I misunderstood your concern a little bit. FWIW I think your new previews look very good and you're on the right track.

@Dadais was really helpful in creating a grid over a map for me which makes it mostly "diamond art" if you've ever done that before, so I should be good here.

If either of you wanted to be extremely helpful to the community sometime, an idiot-proof, step by step instruction thread on how to appropriately determine the dimensions of the grid, and create and layer the grid in a paint program would be outstandingly useful. Perhaps the next time one of you makes a map, you'd be willing to document the process?

I know I certainly have other scenarios (Germanicus and Caesar come to mind) where I would like to remake them with better maps than my old efforts! I'd love to get a better handle on how to appropriately size a grid to a picture and overlay it as a layer.

A grid layer and what you call "diamond art" used to be my process for making maps. (Believe it or not, I actually started out 20 years ago using clear transparencies with grids printed on them, overlaid on paper maps!) That can absolutely churn out great maps, but the process always seemed excessively labor-intensive to me, and very unforgiving: changing my mind just a little bit regarding the Civ map dimensions, or region to include, essentially meant starting all over.

So a few years ago I spent some time developing a new approach. My current process relies a lot more on latitude/longitude coordinates, Excel formulas, and some Lua code -- much less visual and more mathematical. I don't know if I can make my process idiot-proof because I still get tripped up myself, despite attempts to document my own steps; there's still a bit too much trial-and-error for me to be able to explain it coherently to someone else. Also, it works very well for coastlines, cities, and elevation-based terrain like hills and mountains; but rivers and precipitation-based terrain are still mostly a manual process. I'll keep refining and documenting as I go along, and maybe someday my process will be publication-ready.

 

[JPetroski]

I'm not going to keep doing map updates after this, but here are the airfields (white). I might add some more to the UK. I think there's about 27 of them in the UK for now.

I don't know that airfields should be buildable, at least by Germany. As you all can see there are a ton of them available, they're all historic, and they have pretty good coverage. There's a few sparse areas I'll double check but for the most part if you're paying attention to what is going on you should be able to get units in range to intercept the streams pretty much wherever they are.

 

[JPetroski]

Use rivers and "railtrack" terrain for actual terrain purposes. I'm sure we'll be able to find an acceptable solution for logistics.

Actually I'm now recalling that the reason we had a railtrack terrain was so that a go-to movement would work if the human player decides to drag their mouse. That's OK though. With the newer ToTPP terrain I don't have to use rivers for it as I have more "Terrain2" type slots available. Getting rid of "Installation" and "Other Theater" allows me to have proper mountains and swamps for a bit more interest.

I have to say I'm pretty energized about this. I'm finding the map making process entertaining for some reason. I know "Masters of the Air" is coming out (finally!) in late January so hopefully that will further inspire me.

 

[Prof. Garfield]

I understand wanting to hunt trains, but do you want to move them? Here's an example of an alternate mechanism: For each industrial city, the human player can choose an airfield to ship its production to (or no city). Production is automatically reduced in the industrial city, and increased in the airfield, and trains are randomly placed on the rail lines between the cities each turn. If the industrial city isn't connected to the airfield, the city doesn't have the option to send production to that city. For the AI, production is just automatically distributed based on calculated needs.

 

[JPetroski]

I'd 100% prefer that a player not have to move a single unit that doesn't fly in this scenario, but still have control over where supplies are sent and efforts are made, so something in line with your idea would probably work fine. I'd be perfectly fine just teleporting the trains where they need to be. Possibly on a regional level for airfields (so the "main" cities would have a choice of pushing resources to a group of airfields).

There will be **a lot** of airfields so I think simplifying is going to be the way to go here.

 

[JPetroski]

The map is completed, and uploaded to the downloads section for anyone who wants it. Desert = city center site. Tundra = airfield. Also note there are minor rivers on here as well you just can't see them zoomed out.

@Prof. Garfield: I haven't added any of the urban terrain because I think the best way to do that is via events when adding improvements to the cities, so I can see then and there with each placement if I've screwed up a coordinate. I'll also have to grab the code from the first version where placing an airfield also places the night airfield to make life much easier on myself.

It's a bit of an arduous process but then this map right here took ages. Definitely a fun and rewarding experience though and a great thing to hammer away at while you're listening to a good audio book. Many thanks to @Dadais for helping me out with the grid that made this map possible. It's certainly the most accurate map I have ever made (though that isn't saying much compared to its old friends).

I have a few days away from the main working computer and am probably going to take a little break and play the new WW2 in Italy scenario but hopefully I can find a few days soon to get the units and rules established, break ground on the cities, and place the radar installations where they belong. I wonder if a unit (in this case a radar facility) will disband automatically if it's home city is transferred to another civ via events?

 

[Prof. Garfield]

I wonder if a unit (in this case a radar facility) will disband automatically if it's home city is transferred to another civ via events?

If you change the owner of the city, any supported units will still be listed as being supported by that city, even though the owner is different. I recently wrote gen.transferTileContents which transfers units the same way that the template does when you set

simpleSettings.rehomeUnitsOnCityCapture = true

in simpleSettings.lua.

I'm pretty sure that civ.captureCity will cause the units to be disbanded (unless you've enabled rehomeUnitsOnCityCapture).