for PC (DOS)

BootSector:Popular Vote:
Alternate Titles: Take the A-Train III
Company: Artdink / Maxis
Year: 1992
Genre: Strategy
Theme: Business / Logistics
Language: English
Licence: Commercial
Views: 18270
Review by BootSector (2017-01-23)

A-Train is a railroad-themed city-building game with a heavy business-management focus originally created by Japanese developer Artdink and published in English by Maxis in 1992. Although part of a popular series in Japan, enjoyment of this title is sometimes hampered by a smattering of design flaws, which notwithstanding, still make for an entertaining and engaging city-builder that is under-known and underappreciated. Despite this, A-Train failed to capture the imagination of the English-speaking world and was relegated to a legacy as one of Maxis' lesser known titles forever in the shadow of Sim City and its cohorts.


A-Train began its life as Take the A-Train III, the third installment of Artdink's already popular series and published in Japan in 1990. This popularity did not go unnoticed by the makers of Sim City who decided to dispatch then company president Jeff Braun to Japan in a bid to obtain the English-language publishing rights for the title. According to A-Train: The Official Strategy Guide, this resulted in an amusing misunderstanding where Artdink initially thought that Maxis wanted to confront them over the similarities between A-Train and Sim City.

Ultimately, Braun won the Artdink executives over and claimed the rights to publish A-Train in North America. Maxis then began the process of porting the program to platforms Maxis typically worked with at the time as well as localizing the game for an English-speaking audience. This heavy localization work apparently also included various improvements such as the ability to save a game to an arbitrarily named file and directory. Trademark concerns with Billy Strayhorn's jazz single, Take the 'A' Train, coupled with the fact that no previous A-Train game had been published in English under that name (A-Train II was published in the US as Railroad Empire by Seika Corporation) lead Maxis to publish the game simply as A-Train.

A-Train was Maxis' first (but not last) attempt at acting as a publisher for a game created independently by another developer. This probably indicates that Will Wright's company had hoped to position itself as major publisher in the longer term (much like EA which would later swallow Maxis up in an eventual acquisition and merger) rather than a mere developer of software. Unfortunately, A-Train did not garner much attention from the American audience and it proved to be the first major failure for the company that at this point had already developed and published Sim City, Sim Earth, and Sim Ant. I believe that I can personally attest to this in that I was able to obtain a mint-condition, factory-sealed copy of A-Train for DOS in 2008, some sixteen years after its original release, for a meager $10 US. Maxis also sub-licensed their English conversion to Ocean Software for distribution in Europe but I do not believe to much additional success.

Maxis would go on to publish an add-on known as the A-Train Construction Set which allowed the player to edit the terrain of a saved game and also place tracks and most anything else that can be done in-game without having to pay for it. The construction set also failed to dredge up much interest in itself or in A-Train proper.

Choose your platform

Circa 1992, Maxis had already published games for MS-DOS, the Amiga, and the Apple Macintosh in various combinations and opted to bring A-Train to all three. Japanese PCs of the time were apparently incompatible with their American counterparts and Maxis subcontracted the porting of the DOS version to RaxSoft and the Mac version to The Dreamers Guild (I assume both now defunct). I don't know about the Amiga version but imagine that it must have also been contracted out if the DOS version wasn't even ported in-house. The DOS version was released first around the middle of 1992 with the other platforms following towards the end of the year.

All three versions are identical in terms of gameplay but have somewhat different interfaces and other capabilities related to the individual host platforms. I personally own complete copies of and have played extensively both the DOS and Macintosh versions. I prefer the Macintosh version because I believe the interface to be superior as it complies with the Macintosh GUI sensibilities of the time vs. the DOS version which has a one-off GUI developed specifically for A-Train that just isn't as good.

The DOS version is also hardcoded to a specific low-resolution (one common for the time period) whereas the Mac version simply runs in a window in MacOS and so can leverage a variety of resolutions allowing the player to see more of the game board at once on modern higher resolution monitors when using emulators such as Mini vMac (use Mac II mode for color) that can emulate different screen sizes. I'm not terribly familiar with the Amiga version and have only seen screenshots of it in the manual (the Mac and Amiga ports share a single version of the manual; DOS has its own).

I also believe the sound is better on the Macintosh version than on the DOS version but I'm pretty sure this is where my own sense of nostalgia starts to trump my objectivity. Both versions basically play all the same tunes but the music seems to have been remixed somewhat on the Macintosh version (or maybe on the DOS version, or maybe both). Really, both versions of the music are catchy and have their own unique charms but I prefer the Mac version; probably because it was the first one that I heard and grew up with. I do think that, objectively, the “clickity-clack” sound effect when the trains move is better on the Mac, however.


The front of the box for A-Train reads “Trains,” “Cities,” “Money” and I think this pretty concisely sums up the different aspects of the game. The player assumes the role of the head of a railroad company tasked with the creation and operation of a city-wide transportation network. The city building portion of the game emphasizes placing tracks and stations and then buying and operating both passenger and freight trains which move passengers and building materials between stations.

The game uses simple white cubes to represent building materials which are used to construct all manners of buildings and facilities. As materials and people are moved to new stations the simulation starts to develop the land around the station and this is how the city grows. Laying out the track routes and stations where ever you want them to go is the enjoyable city-building action for the player and watching your city develop on its own really makes it satisfying.

In addition to the transportation business the player can also use building materials to construct structures for various types of secondary businesses which the game refers to simply as “subsidiaries.” The most functional of these is the factory which produces additional building materials at a regular rate so that they no longer have to be imported from off of the map. More mundane facilities like apartment buildings, office buildings, department stores, and hotels can be built as well and are also constructed automatically by the simulation given the right conditions (they are then owned by unnamed 'competing companies'). Finally, there are host of larger and player-only constructed subsidiaries that are more whimsical such as stadiums, ski resorts, theme parks, and golf courses.

All of these buildings produce income of some form and have associated operating expenses and so can generate profits or losses depending on various factors such as the surrounding population and the presence of competing facilities nearby. All subsidiaries can be bought and sold (the computer players can own types that they would never build on their own) and the value is affected by many of the same factors that affect profitability. In many cases buildings are much more profitable when built and sold rather than by just operating the associating business.

The subsidiaries aspect of the game is part of the city building appeal but really they speak more to the supposed ultimate goal which is not to develop the city but to make money. Even more to this end is the built-in stock market that the game includes where money can be invested in a whole host of companies and made and lost in the market ups-and-downs. After all, in A-Train you are operating a private company and so isn't making money really the goal?

In fact, A-Train has a win condition of amassing fifty million dollars after which you see a brief animation and are informed that the game is over (much the same as when the game ends in bankruptcy) and that you won. As a game, I think this is flimsy and yes you do need to build a successful business but that's just so that you have the resources to continue to expand the map (without that challenge it wouldn't be as fun). The win condition always seemed really arbitrary to me and it's something that you can easily reach or avoid as desired once you've built out most of the map.

Comparison with Sim City

A-Train is often compared to Sim City and to some degree this is apt since both are city-building games that simulate the economy and the population growth of an actual city in some capacity. However, there is a good deal of difference between the two and I think the comparison is made more often than it would have been otherwise merely because they both bare the Maxis label (in America and Europe, at least).

While both games simulate the population of a city, A-Train has a much narrower focus than Sim City which is a more well-rounded simulation. The Sim City city-model includes many things that A-Train ignores such as crime, pollution, an electric grid, cars and traffic, and municipal services such as the police and fire departments. The A-Train model includes none of this and simulates mainly just the growth of the population and the supply and demand for the labor market although I think both are still a more simplistic model than Sim City's. Also unlike Sim City, in A-Train the population of the city never really decreases.

A-Train's emphasis on railroads is obvious and it certainly explores it specifically in more depth than Sim City does but overall Sim City is still far more robust in terms of attention to city-planning. In A-Train the only real infrastructure that can be controlled (or even simulated) is the railroad where as Sim City includes rails, roads, airports, electric lines, and all manner of public building. The Sim City player even tells the simulation what it's allowed to build where and to directly manage the availability of each time of zone. A-Train mostly just does whatever it wants to as long as the player provides raw building materials.

Finally, the underlying mindset is different between the two games, at least in principle. In Sim City you assume the role of public servant charged with growing the city and ensuring it runs smoothly for the public benefit. In A-Train the player's chief concern is (supposedly) first and foremost to turn a profit and any public benefit of doing so is probably just incidental (although the A-Train manual speaks to this and tries to spin it from a much different perspective). However, I think these mindsets are more imagined than actual and I really do approach both games in the same way which is to build a cool city; that's what's fun about both and I think it is how most other people would also play them.

Design flaws: It's the economy, stupid

The biggest problem that I have with A-Train is that invariably, whenever I play, I reach a point where the growth of the city simply stalls out, seemingly inexplicably. I've seen it happen where the simulation no longer builds structures of any kind and I've also seen it where it only builds simple houses and no longer larger buildings like apartments or sky scrapers. Both the A-Train manual and the official strategy guide make vague references to this predicament and say that the player must keep the supply of and the demand for human labor in balance to ensure a healthy economy which continues to grow (in addition to external market forces that can fluctuate randomly).

Balancing supply and demand is supposedly accomplished by building the correct amount of labor consuming subsidiaries (basically anything that isn't an apartment building) to provide employment for the denizens living in houses and apartment buildings. Neither source is specific beyond this in a way that is helpful and my personal observations of the behavior of the game make me think that it's not true anyway. What I observe is more that the simulation always moves toward equilibrium on its own. As the population increases, the game will build more and more office buildings thus equaling this all out automatically until suddenly everything just stops for no discernable reason.

Now, it's possible that I'm just not very good at this game and if I played it differently I wouldn't run into this problem but this takes me to the other significant issue I have with A-Train. This issue is that the rules of the simulation are very opaque. They are not easily apparent to the player and they are not clearly spelled out in the manual either. For example, the way the game calculates fares for a ride between two stations is based on distance between stations but nothing about how this works beyond that is revealed by the game or the manual. The official strategy guide does explain this aspect in detail but the existence of the guide is something that I've only recently become aware of.

The bigger problem with transparency is the simulation of the growth of the city. If this is truly just a result of not building the right types of building then it's still not a code that I've been able to crack in all the years that I've played the game. If this is the case then the game doesn't provide enough insight into the state of the simulation in order to coupe with this (unlike Sim City which includes extensive graphs and reports about the state of the simulation). In the end, I really just think this is some kind of program bug that causes growth to stop (I wonder if maybe there's just an upper limit to the number of building the game will build?).

Smaller issues

Not surprising is probably that, as someone who has been playing this game for a long time, I've noticed my share of little things that could be improved or just don't seem right. Some of these things are legitimate enhancements while others are little oddities. For lack of a better place to put them, I am just going to spit them out all right here.

First, one of the cooler things about building your railroad network is that you get to schedule the trains (the time of day that passenger trains depart affect their passenger loads and thus profitability) which can be especially important if you are trying to time freight trains to not be at the station while a passenger train is going to be. Unfortunately, your ability to schedule arrivals and departures is entirely based upon when you have it leave the previous station. Worse still is that there are only 5 possible options for the times that a train can leave the station. The ability to pick any arbitrary departure time would be nice.

Freight trains are really dumb. If they are empty arriving at a station they try to pick up building materials or if they have any onboard they try to put them down at the station. This works fine in the simple case where you are just trying to move material between two stations in one direction. In more advanced cases where you have a loop or another multiple-stop line you probably have one or more production stations and the rest should just be receiving. What happens is that if there's space for a drop-off earlier in the line then the train will be empty after that and then pickup materials at the next stop which might just be another receiving station and thus steals materials that you don't want picked up. It would be really nice to be able to prevent a train from picking up or putting down materials at particular stops.

Also related to trouble managing transport of materials is the fact that the simulation always likes to build competitor subsidiaries right where you are planning to put future tracks (or your own buildings). The game manual says to buy the land ahead of time to prevent this but the problem with that is freight trains will drop material on any land you own that is in range. So, this land fills up with materials you don't want there which also blocks you from building on it until you haul the materials away. The best way to avoid this is just to build the future tracks around the station up front and not connect them to the existing line until you are ready. You can also place tracks temporarily to save land for future buildings. This costs more upfront but just makes things easier later on. It would be nice to be able to control which land can be used for materials beyond just owning it.

The freight business is a money pit. It is impossible to make money in the game by hauling freight. You have to do it to build the city but it's going to cost you; I guess it's just the cost of doing business. The reason is because the base fare (which is hard coded) for hauling materials is so low that no matter how short or long the freight route, the final computed total fare will be breakeven or less (except for exceptionally short trips) when the cost of running the train is taken into account. Factories can operate at a profit if you keep up with hauling away the materials they produce but the profit margin is small and any money made is eaten up by the losses produced by running the freight trains themselves (and then some).

Since freight trains always operate at a loss, the only way to make money with trains is to create profitable passenger lines. How to do this is affected by the choice of train that you purchase and operate but it's mostly just because each train has an associated and hard coded base fare. The trains with the highest base fare make the most money. There are some fifteen passenger trains to choose from so it seems like there is a lot of choice but really there is only a single choice because the AR-III (named for Artdink) has a much higher base fare then all the other trains and so produces gobs of money faster than the rest and easily pays for its own high price tag in mere months once a population base is established. The cheapest of the rest of the trains are basically not profitable at all and the mid-range ones are very moderately profitable and you'll probably just end up replacing them with AR-IIIs anyway. The manual fails to explain how fares are computed but it does tell you that the AR-III is the best train to use. Why'd they even both to design the other trains?

The distance and time scales for the game are kind of ridiculous. The game divides the game map into “blocks” which are the smallest units you can work with to build a building, lay one piece of track, etc. I think a block is maybe about an acre of real-life land-not a large distance. A station is three blocks long itself. What's crazy is that a “high-speed” train only moves three of these blocks per game hour and you generally want to put stations at least 13 blocks apart (to increase fares) which means it takes several game hours at least to move between stations. In a game where you are supposedly building commuter trains within a city it doesn't make much sense to think you'll leave the station for work at 8am and arrive at your job by noon.

The simulation determines how many passengers board a train at a station based solely on the number of buildings around the station and time of day. It does not take into account how many other trains are servicing that same station. The passengers also don't seem to care where they are going. This makes it somewhat easy to game the system and generate a lot of profits from a tiny section of the map if you want to. All that's necessary is to build up a population base around two stations and then you can run multiple lines and trains and even multiple stations between the two and generate a lot of ticket revenue since there seems to always be a limitless supply of passengers each day.

The game charges a 50% annual tax on all cash profits which I think is just punishing. Fortunately, if you convert those profits into assets like trains, building, or stocks they are taxed at a much lower rate. The best way to shelter large amounts of currency is probably just to buy stocks and the in-game stock market is kind of a neat feature. However, the market runs in really predictable and regular cycles. It makes it really easy to game the market and make huge profits for doing basically nothing especially when you consider you can just reload your last save from right before the market crashed.

Finally, the aforementioned arbitrary $50 million win-condition could use some work. As I said, I don't think this makes sense and is unusual for a Maxis title which are usually completely open-ended or have only optional goals (e.g. Sim City's “Scenarios”). Either way, the goal is too trivial to reach because it doesn't take taxes or debt owed into account. All you have to do is build your cash up a little by gaming the stock market then use your owned assets to get a big bank loan and finally sell off all your assets and you can easily inflate your cash to $50 million. However, much of that will be wiped out when tax time comes around and your loan comes due leaving you not much richer and your empire in tatters (doesn't actually matter though since the game is over by that point). If Artdink really wanted to make this a challenge they should have required the $50 million to be in net assets once all liabilities and taxes are deducted.


For all its faults, A-Train is game that I've played a lot and I keep coming back to it again and again because I like it. I like to play it because it's fun to play and ever since I first encountered in the computer lab of my school in third grade – our teacher owned a copy and loaded it on one of the Macintoshes and let some friends and I stay in during recess just to play it – I was hooked. It's fun to play for the same reason that all city builders are fun to play but it also has its own unique charms. You get the same enjoyment as from laying out any model railroad and watching your trains whip around is really endearing. As I write this, in early 2017, I am aware of the fact that the 25th anniversary of the game's release (in English) is coming up later this year and I have been playing it (on-and-off) for nearly just as long. I fully expect that I will still be playing A-Train in another twenty-five years from now.

Comments (6) [Post comment]

This game is a classic. I remember playing this when I was just 9 years old, on my Commodore Amiga, for long hours before going to bed. I just revisited it recently on my MS-DOS machine. It is still as good as in the 90's :D
I would love a detailed how-to-play and earn money in A-Train. I tried to play it again a few months back, but never could get used to the monetary dynamics and grow the map and make money. Seems like I was always in the red! It was fun creating areas, but the bank can only loan so much money. :P
Originally posted by MasterLee at 21:18 on January 23rd, 2017:

No. In Sim City your goal is
… to accomplish any goals you set for yourself.

Absolutely, I agree completely. This was part of the point that I was trying to make that both games are really open ended which is kind of Maxis' hallmark. This is the main reason that I think the $50 million win condition was kind of silly.

That said, there's a difference between goals and roles and both games are structured, at least roughly, around different real-world concepts. I think its fair to say the powers and responsibilities of the A-Train player look like that of someone running a private firm (maybe a COO) while those of the Sim City player look more like a Mayor, town council, or just a city-planner. This was the only other point I was trying to make and I still think its apt. It's still a game and you can always play how'd you like but it could be inline or not inline with how you might expect real-world individuals in such positions to behave.

In Sim City you assume the role of public servant charged with growing the city and ensuring it runs smoothly for the public benefit.

No. In Sim City your goal is
… to accomplish any goals you set for yourself.
Mr Creosote:
In Sim City you assume the role of public servant charged with growing the city and ensuring it runs smoothly for the public benefit. In A-Train the player's chief concern is (supposedly) first and foremost to turn a profit and any public benefit of doing so is probably just incidental (although the A-Train manual speaks to this and tries to spin it from a much different perspective).

Very interesting point, though philosophically speaking, would you really say this should make a difference? What is a corporation? I mean originally, before turbo-capitalism took over everybody's mindset? It's really nothing but a group of people coming together to ensure their personal social well-being, isn't it? Much like an organized community (i.e. a city). The perfect corporation, in my mind, would balance it so that it makes exactly zero profit, but manages to keep its participants well fed. Which makes the whole concept of shareholders etc. (i.e. "owners" of a corporation who have no active stake in it) so baffling to me.

The cheapest of the rest of the trains are basically not profitable at all and the mid-range ones are very moderately profitable and you'll probably just end up replacing them with AR-IIIs anyway. The manual fails to explain how fares are computed but it does tell you that the AR-III is the best train to use. Why'd they even both to design the other trains?

This is actually one of my pet-peeves within many games of this genre, too. I would love to build a more realistic transport network with smaller, slower trains feeding long-haul high-speed lines etc., but rarely is there ever a world simulation in which this makes sense.

P.S. I believe The Dreamers' Guild also developed the Amiga port.