Map size discussion

[Wooly]

Well, I have searched and read threads for discussion of the best map sizes to use, but I have not found any, so I figured I would start one.

The game supports up to 1024 x 1024 sized maps, but it does recommend 400? x 400? Is there an optimal size? I was about to make my map the biggest size (1024 x 1024), because I love big expansive maps, but then I thought it might be too much for some computers. The first version of the map will be a freeplay, and I expect it will have a lot of trains. What is the recommended size for busy maps?

I am sure there are many different sized maps that are best for different goals, so add what you know.

[WPandP]

I started a very detailed map of Virginia at 1024x1024 and had to more or less abandon that, as it was too unwieldy. My computer isn't the latest and greatest, but it does far exceed the minimum or even recommended system specs for RRT3.

I did my Paston Valley Lines scenario at 512x1024, mostly because I was working from square DEM's and the area I was working on spanned two; I could have cropped it down a bit, to perhaps 800 wide or so. I've got cities spaced pretty far, and the topography feels pretty detailed at this size.

One of my biggest gripes, though, is with maps that don't have enough cities in them... such as a map of the entire US, which includes Detroit but not Toledo. It's a matter of being comfortable with the map's scale, and what should be included. I like the thought of making up large metropolises by placing several cities close together, labeling them after suburbs, such as Chicago with Gary to the southeast, Aurora to the west, and Waukegan to the north, all close enough such that they blend together into a single city mass. Of course, that still really depends on the map's scale still, because you might end up with undeveloped land between the city and its suburbs, if you don't place them correctly.

A smaller map has the advantage of being something that you are more likely to invest a lot of detail into - you'll place the rivers where they really belong and make sure they flow in the right direction, you'll use the right kinds of trees and ambient forest sounds, you'll plan out your regions and industries a little more carefully. One tends to get sloppier with these things when the map is larger. I think 512x512 is plenty, most of the time, but of course a more oblong map needs a different aspect ratio.

[JayEff]

I favour map sizes that fit the story. If it is about a small island or a short line I think it should be smaller. I like 576 x 576 or smaller. I never play maps that are 1024 x 1024. Note that even Steve Lorenz' epic maps of Great Northern and the Trans-Siberian are not really too big.

Another consideration is how complex the map is in other ways. A lot of territories, cities and events, or high density, also make demands on computer resources. If you want an entertaining sequence of events, you should go with medium or small maps.

[nedfumpkin]

I personally like maps that aren't too big. I find that on the bigger maps I will get to a point where I will calculate whether I can achieve gold or not at the 10-15 year period and get bored laying extremely long lines. If the map is intense and makes it so you have to play to the end to know whether you have gold or not, I will generally play it out. But when I own my own company, and all the industries I service, and I am raking in 2 million a year with 5 million worth of construction to lay, and still 15 years, I will generally find something else to play. The story is what is really important to me.

Size only really matters in sex. :)

[Canadian Viking]

If you are mapping the real world, the size of the map must be related to how much land area the map will include. I am very happy with the scale of my Building to Buffalo map, which (east to west) stretches from New Haven to Erie in 896 (X704). That is a larger map than most that come with RT3, but I wanted to include more intermediate cities and not just the very big ones.

If you are creating an imaginary location, I'ld follow JayEff's advice and stick to 576 X 576 or smaller.

My next game's map is about the same size as B2B (I'm not at home so can't check) and I would not use anything bigger than 896 unless the 2nd dimension was smaller. WP&P, you might be interested in it when I finally get it finished. Chicago is on the east edge of the map and is surrounded by Waukegon, Elgin, West Chicago, Aurora, Joliet, Blue Island, and Calumet Junction (which is now Kensington). Gary is not there as it was not built until decades later. The game is Building to the Great West and I can't say that it is coming soon, as it still needs more play testing, which inevitably results in more modifications, which need to by play tested, etc., etc.

[wsherrick]

I like maps to have a feeling of space and distance. I made the Age Of Steam-Blue Streak Map to give a feeling of being out on the Prairie or building through the Ozarks. I made it from scratch by having an Atlas in my lap. The map works fine, but if you have a computer without a big memory or a super duper video card I can see how the size of the map can bog down the computer, especially after you lay a lot of track and get a bunch of trains moving. The scale is important to me as is realism. That's why I like bigger maps that represent smaller regions. I am working hard on a new scenario, the map is somewhat smaller than the Blue Steak map, but it represents half a State. I agree with Jay Eff that the more events you have in your game you need to have a smaller map to insure that it runs smoothly. I guess we all learn as we go.

[AZ Rail Rat]

And "they" say size doesn't matter!

[low_grade]

Hopefully this will get noticed. It kind of fits under the topic title. Just wondering how to get the Editor to let you make an oversized map. It seems like if I try to import anything bigger than 769x769 or 513x1025 the Editor shrinks it automatically. Yet I see plenty of 641x1025 and even 1025x1025 maps in the archives. So, what's the secret?

[Grandma Ruth]

Look guys, it's quality not quantity that's important.

[nedfumpkin]

If I remember correctly, someone calculate that a track segment is 11 miles. Anyone know how many pixels that is? Could allow for a bit of math and scale maps. :)

[Wolverine@MSU]

Hopefully this will get noticed. It kind of fits under the topic title. Just wondering how to get the Editor to let you make an oversized map. It seems like if I try to import anything bigger than 769x769 or 513x1025 the Editor shrinks it automatically. Yet I see plenty of 641x1025 and even 1025x1025 maps in the archives. So, what's the secret?

I've never had any trouble importing any size up to 1024 x 1024. If one dimension is larger than this, the Editor will scale it down to fit the longer dimension to 1024 and the shorter dimension gets reduced by the same scaling factor. If you post one of your heightmaps, I'll try importing it and see what I get.

[low_grade]

Huh. How about that. Looks like I was getting thrown off by that warning from the mapmaker. Imported a 1024x1024 just fine. Or maybe I tried to import one where a dimension was bigger than 1024. Sorry about the confusion...

[Gwizz]

I believe the biggest problem with large maps that bog down a computer, are events that test un-necessarily. On smaller maps I get lazy and don't fine tune the tests each event makes.

When I do fine tune the events, I seem to always find tests that the events did not need to make. Every added test sends another task to the CPU. I once counted almost 200 tests made by one event. That event was in a RT2 map.

I should, but seldom test a new map on an old computer to make sure it runs trouble free when computrer power is limited.

[Sugus]

RT3 was born some years ago. So, 400 x 400 wouldn't be a bad point to start ...

... as Grandma Ruth mentioned: quality - not quantity !!!

[nedfumpkin]

I just calculated the scale of a map in RT3 and 100 pixels equals 139 miles according to the game. 1.39 miles/pixel

This means that a map 1024x 1024 is covering an area of 1423.36 miles by 1423.36. Just over 2 million square miles.

[low_grade]

I don't believe there's a fixed relationship between pixels and their length in miles. Depending on where you grab your map in the mapmaker, a pixel is said to be anywhere from .4 miles to 1.5 miles, and that's in 1:1 scale. For higher scales, the relationship doubles, triples, etc. Then if you scale up or down or whatever with image editing software you get to change the relationship again. Mapmaker also gives you the pixel by pixel size you need to scale your map to in order to get something like 1.5 miles per pixel. If you're doing a map from scratch in essence the relationship can be whatever you want it to be. If it's a metro map, you could say a pixel is 0.05 miles. If it's some other giant planet you could say a pixel is 100 miles. Either way the game doesn't notice or care.

I do agree though that smaller maps make for more intense gameplay. Every map I've played that's really big I always get to a point where I've carefully developed my network in some starting locale, bought up and built a bunch of industries, and after 15 or so years have enough income that I don't have to be careful anymore with track laying and efficient routing and end up just building long straight track to meet the connection requirements that those maps normally have for medaling.

But even on smaller maps I'm usually done building after 15 years and the last 10 or 15 years is just doing my best to build PNW or industry profits or meeting hauling goals or whatever.

I guess I'm trying to think of a way to have developing the rail network something which keeps the player involved through the entire game. Probably by opening up territories and having requirements for profitability as well as connections, and perhaps limiting track sections available, so that it's both a race to build until the end, and you have to be careful how you do it. I love laying track carefully in a game, and to me a game looses much of its appeal once track laying is done. So it's either find some way to make a big map interesting, or try to make a scenario that only lasts 15 or 20 years for a normal map size.

One thing I'm thinking about it something like this. In the US, rail development was limited to the places that had been successfully pioneered. So up to 1850, the only extensive rail networks were on the east coast and to buffalo, with a few short lines scattered about the south. But by 1890, many plains states and mountain states, plus cities like Chicago, Detroit, Cleveland, etc., had seen a big boom in population, and rail networks became quite extensive in these places. So the first thought was wouldn't it be nice to be able to add cities as a game event, so that after 15 or 20 years, all of a sudden now you have a reason to build your network out some more. But since I don't think you can add cities, the idea I'm thinking of would be to have the cities start with like 25% density, so that they're not really attractive, and then have events trigger which add warehouses and industries and houses to simulate the rapid growth of a city. Have a newspaper announce "The West is Open! Rail service needed!" in like 1870 if your scenario starts in 1855 or 1850, and change the briefing to show new medal goals involving hauling to these new population centers. Just an idea. Proper play testing and goal setting with some penalties like taxes or depressions or having to pay lawyers or funding guerrillas to prevent your competition from building through the strategic pass, etc, all together can keep income modest and keep you biting your nails until the end.

Also, from my experience, large maps never seem to bog me down in any way which is really a problem, and I'm running a pretty minimal system here. Old Dell laptop with a 1Ghz P3 processor, like 64MB video ram on the motherboard, 512MB regular ram, running on XP. Yes, when I run the program on very fast with like 150 trains and lots of events, I get some lag to my commands and mouse movements, but after I pause I can do whatever I want just fine. Placing new industries does seem to create lag as well, even in pause. But these things don't really take away from my enjoyment of the game.

On the other hand, larger maps take longer to create and tweak, so that in the time it takes to make one 1024x1024 map you could make six 400x400 maps!

[nedfumpkin]

1.39 miles per pixels comes from a blank map where I ran a length of track 100 pixels long, then in the ledger, I put in company track miles and it said 139. Are you saying that the map size will affect this calculation?

[Wolverine@MSU]

On a 1024 x 1024 map, I could lay 1413 miles of track from cell 0,0 to 1023,0, yielding 1.38 miles per cell.

[low_grade]

Well you guys have been doing the experimenting, while I'm only going on theory. Perhaps what's happening here is that the game does have a fixed relationship between pixels and length in miles, but this doesn't correspond to the actual real world distances found in the map section you grab in mapmaker. That must be why mapmaker always gives the translation to 1.38 miles/pixel or whatever. So it sounds like if you want the performance of your game to be accurate to the scales on your map, you better follow mapmaker's advice and transform to 1.38 miles/pixel. Though gameplay can be great even without realistic correspondence. So what if your scale means it takes a train the same amount of time to get from Chicago to New York as it would take a train on another map to get from Detroit to Cleveland?

[nedfumpkin]

I think it all depends on the scenario. I'd like to start trying to make scale map on some scenarios, just to see what happend. 1400 miles is quite a distance so alot can fit on a large map.