How to make money in the early years?

[davidh]

Hi

I thought I'd read somewhere that one of the drivers for the experimental version was to make it easier to make money in the early years of a game, and more of a challenge later. However I'm finding it impossible to make money in the early years.  Is there a flaw in my game strategy or is there a flaw in Simutrans Experimental?

For example my current game played with the Pak128.Britain-Ex. I started in 1830. I've spent my initial money on a few passenger routes usingg horse drawn carriages, a few industrial routes using horse drawn carts and a few industrial train routes. Last year:

Revenue: 1460
Running costs: 1031
Infrastructure maintenance: 14,845

That's a HUGE maintenance cost for such a small revenue. What's contributing to this cost? Am I paying for all the public service infrastructure (roads) created when the game was started? It seems very high to me.

Regards

David

[jamespetts]

David,

thank you very much for your feedback. Pak128.Britain-Ex is still work in progress, and seems to be the case from your reports and that of several others that it needs rebalancing to make it easier to make profit. However, can you upload your saved game so that I can see whether there is anything in particular that you are doing wrong that might contribute to the unprofitability? Thank you.

[davidh]

Hi James

Thanks for your reply. It's the same save game as I uploaded for the 'out of credit' problem I reported.

Cheers

David

[neroden]

I also tried playing Simutrans-experimental with Pak128-britain 0.2, and it is completely impossible to make money in the very early years -- and I know why.

It's largely the maintenance cost for stations and depots.

It's quite possible to set up a route with revenues greater than running costs, and even large enough to cover the costs of roads and rails, and with some return on capital left over -- since the vehicles start out very cheap.

Unfortunately, you're earning maybe half a pound per run, and a run takes over a month.  (There's some funny stuff with time in Simutrans, you know.... a vehicle going at 1 km/hr takes 1 month to cross 1 tile, which should be 1/4 km.)  I can get, with fairly simply designed goods lines, to $50-$100 yearly difference between revenues and running costs; some easy expansion could double that.

This means that the station costs, starting at $4.50/month each, can *never* be covered.  The cost of two minimal bus stops is more per month than you can possibly make with the best possible road connection (you're limited by congestion in how many road vehicles you can run through a single station stop).  Rail stations are just as bad (number of station tiles limits train length).  My $50 with road vehicles requires 3 station tiles and would need more for any expansion, due to congestion; my $100 required 6 tiles due to train length, and would require additional stations tiles for any expansion. (Note lack of signals in 1750, otherwise it could have been 4 tiles.)

I tested this by making all the stops public, taking the maintenance costs off my hands.  This brought me *very close* to profitability -- but still not there yet.

Because I couldn't give away my depot.  That *$18* per month per depot is just too much back in 1750, or even 1830.  You have to set up a *large* network operated out of one depot -- I estimated at least 8 horse-train routes (one train per route, because no signals), more for road vehicles (due to congestion limiting the number of vehicles per route) -- before you can possibly produce enough money per month to pay for one depot.  In fact, I think the only way to make a profitable network in the early years is to design a very large network, get it exactly right, and build it in one go.  (Giving all your stops to the public service player.)

This obviously isn't intended.
There are three possible solutions:

(1) Increase profits, running costs, maintenance costs of roads and rails, etc. in the early years -- which all seem fairly appropriately in proportion to each other -- so that the running profit is high enough to cover the very expensive stations and depots.  This makes sense; the profit for moving a load of fruit was much higher in 1750 than in 1950.
(2) Make cheaper stations and depots.  Probably easier!
(3) Some of each.

It seems to make more sense to have cheaper, smaller stations *regardless*.  Most of the horse-drawn passenger and goods vehicles hold very few units, so a 32-person bus stop or rail platform is extreme overkill.  An 8-unit freight road stop would be a great addition to pak128.britain (perhaps make it a through stop, so the wagon is just pulling over at the side of the road -- a picture of a pile of sacks or boxes by the road would probably be sufficient); 8 and 16-unit passenger road stops are very worthwhile too (not sure what to do for the pictures).  Given that the longest horse-drawn train which fits in one station tile can only hold 14 units, 16-unit goods sidings and passenger stops for trains would be extremely suitable as well.  (The passenger stop in that case should simply have *no* platform, being a whistlestop or "dirt" stop.)

This would cut minimum station costs to half or a quarter what they are now, which would probably make profitable lines just about possible.  If something was also done about the $216/year depot.  Unfortunately there doesn't seem to be an easy way to require more expensive depots for more modern vehicles, though that would be *very* realistic (horse-drawn stuff needs a stable and a carriage house; diesel buses and lorries need a proper garage instead; etc.).  I like the idea of an old-fashioned wooden carriage house/stable combo which was unable to house motorized vehicles, but cost a lot less to maintain.

[neroden]

"An 8-unit freight road stop would be a great addition to pak128.britain (perhaps make it a through stop, so the wagon is just pulling over at the side of the road -- a picture of a pile of sacks or boxes by the road would probably be sufficient"

Incidentally this would also provide a close approximation to the sometimes-requested ability to just "deliver to the door" of industries, particularly terminal industries.  Delivery-only stops often don't need much storage capacity.  It solves the annoying problem of finding a terminating road in the middle of town for such industries.

[The Hood]

The smallest goods stop in pak128.Britain is the mailbox/loading bay - intended for exactly that.  The problem with capacities, IIRC, is that you can only specify a "level" for each station in the dat file.  Level 1 corresponds to 32 units IIRC, so you can't have smaller than that.  Also, maintenance cost is directly fixed to the level (so a level 2 costs twice as much to buy/maintain as a level 1), although the exact value can be set in simuconf.tab.  In my opinion, it would be best if these costs and capacities could be completely independent and set in the dat file for each station - something like you have for vehicles.  It would be easier to overcome the problems you describe.  (in fact I feel an extension request coming on...)

[jamespetts]

Neroden,

thank you very much indeed for your very helpful feedback. My initial reaction is that the revenue capacities of each type of goods/passengers need to be increased, as such small profits in the early years would make expansion difficult even if the depots and stations were cheaper. I think that there has been some discussion of Pak128.Britain generally having too low a profit margin.

I am currently working on the Simutrans-Experimental code (when I get any time at all), as there are some important changes needed for the next version (which will have a new major version number), but, when that is done, I shall revert (hopefully - unless I need to deal with bugs) to focussing on Pak128.Britain and see whether I can fix this balancing issue.

In the meantime, your feedback is very much welcomed - thank you again.

[sdog]

in my most recent game with 6.8 and pak128 i could create a profitable network, and also expand slowly. however i can see that as soon as my most profitable line's dakotas are outdated it will break down. i won't be able to make enough money by then to buy recent planes. (dc3 50k, new plane >600k, company net worth about 450k, monthly operating profit 3k)

pak128 is definitely easier than pak128.britain


i'll make the savegame available, it's on a really tiny map now, requires pak128 last weeks nightly (likely recent one works too) and zeno's plane set 1.1 http://forum.simutrans.com/index.php?topic=1470.0

http://home.tu-clausthal.de/~gsc/14.sve

[The Hood]

Would you mind expanding on the ways in which pak128 is easier than pak128.Britain?  Which bits of pak128.Britain are overpriced do you think?  Stations, general running costs of vehicles, construction costs?  Feedback like yours will help us get it right

[sdog]

first of all i used, as james suggested the pak britain speedbonus tab. with standard pak128 speedbonuses my network would operate at a slight loss.

it's a bit difficult to say with direct comparison between two similar maps, or city setups.

one difference could be the availability of planes and good buses in pak128 @1930. starting from the savegame posted above i tried to build a rail link instead of the airports. i couldn't make a profit.
i also excessively used the possiblity to make a station public. in pak britain this was not feasible, as the cost therefor was rather high.

in general infrastructure costs are very high, in comparison to possible revenues in pak britain exp. (PBX). Higher passenger numbers could compensate this, but this could cause some major difficulties in later games, with huge cities. (in particular with simutrans awkward way of signaling, compared to openTTDs powerfull path based signals.) A possible way could to increase revenue per passenger/unit of goods.

vehicle operation and maintenance costs were not a major factor however, with exception of some train engines.

slope tools seem to be overpriced in PBX.

high capacity, fast loading urban transport was also lacking in early years, this makes a passenger game before 1885 (int. date of first steam tram) almost impossible. without a tram backbone the horse carriages are to ineffective.

i suppose the size of a stations coverage can not be changed? this could reflect that in earlier times people would walk much further to get to the train. (or take taxis if they were rich)

introduction of a four horses set, to pull horse trams could help here. in 1.02 horse pairs for trams could not be stacked, making horse trams impossible.


the acceptable travel time also needs a major increase in the time of horse buses. collecting passengers from smaller towns was almost impossible, by the time they arrived at stations they soon timed out.
An alternative would be to reset the timer or increase the time left when they swap transport.

when operating a rail network in PBX @1885 i couldn't use a good locomotive on cheapest tracks due to weight limits. with more expensive tracks the infrastructure costs broke my neck.


ps.: please don't take inconsitencies, errors and mixups not to serious. i've got only a few more minutes online, and can't check thoroughly.

[The Hood]

Thanks for the feedback.  I think there are a few balancing issues here, some related to pak128.Britain, some to simutrans experimental, and some no doubt a combination of the two.

i also excessively used the possiblity to make a station public. in pak britain this was not feasible, as the cost therefor was rather high.

This shouldn't be pak dependent...

in general infrastructure costs are very high, in comparison to possible revenues in pak britain exp.

This is intended, to encourage you to build infrastructure in proportion to the service levels needed.  I.e. if you build a four-track high speed rail line to serve a tiny village, you will not make a profit.  However, this probably needs tweaking, so if you have more details on this, great.  Same goes for slope tools.  High cost so as to encourage you to build around hills rather than remove them!

i suppose the size of a stations coverage can not be changed? this could reflect that in earlier times people would walk much further to get to the train. (or take taxis if they were rich)

This can be set in simuconf.tab

introduction of a four horses set, to pull horse trams could help here. in 1.02 horse pairs for trams could not be stacked, making horse trams impossible.

I can change things to allow multiple horses, although in standard pak128.Britain two horses do the job fine.  I suspect this is because simutrans-experimental has a different physics engine, which means either (a) the physics model isn't balanced properly or (b) the horses need more power in pak128.Britain experimental than in pak128.Britain standard.  Jamespetts will be able to say more on this I think.

when operating a rail network in PBX @1885 i couldn't use a good locomotive on cheapest tracks due to weight limits. with more expensive tracks the infrastructure costs broke my neck.

Can you say which specific tracks and trains you were using?

[sanna]

This can be set in simuconf.tab

I understood sdog's question as if it was possible to let different stops have different coverage, so that f ex a early horse bus stop could have a wider coverage than a later stop. This is AFAIK not possible, the setting in simuconf.tab changes the coverage for all stations. Furthermore there would be no restriction on people using the older type of stop at later periods, therefore breaking all the realism that this would introduce. The level of realism that sdog wants (if I understood correctly) could only be achieved if station coverage could be made time dependant. Not sure if the increased complexity would be worth it.

[wlindley]

If station coverage were a parameter in the .dat file, this would permit year-based station coverage.  Older stations with larger coverage couldn't be built after awhile... although residents who had become accustomed to walking half a mile to their antiquated station could continue to do so, for auld lang syne.

This would also permit new Grand Central Stations in later years to have large coverage areas... is it possible to import OpenTTD's Toronto Union Station?  It could have a coverage radius of oh, say, 7...

[neroden]

The smallest goods stop in pak128.Britain is the mailbox/loading bay - intended for exactly that.  The problem with capacities, IIRC, is that you can only specify a "level" for each station in the dat file.  Level 1 corresponds to 32 units IIRC, so you can't have smaller than that.  Also, maintenance cost is directly fixed to the level (so a level 2 costs twice as much to buy/maintain as a level 1), although the exact value can be set in simuconf.tab.  In my opinion, it would be best if these costs and capacities could be completely independent and set in the dat file for each station - something like you have for vehicles.  It would be easier to overcome the problems you describe.  (in fact I feel an extension request coming on...)

Yeah.  The thing is, with the smallest stop holding 32 people, it's unreasonable to have any vehicle which carries fewer than, say, 16 people.  Until smaller stops are available, the small horsedrawn vehicles are simply not viable; you need so many of them to make proper use of a single stop that they end up congesting the stop.  Definitely a worthwhile change request for the main code; I think it's necessary in order to make the horsedrawn era practical with the current capacities.  (Even in pak.german under simutrans-standard where the horsedrawn era is just about manageable, the horsedrawn buggies carry 12 people, and the stops are really cheap, the road congestion due to horse vehicles is outrageous.)

There is one alternative, which is to simply enlarge the capacity of the horsedrawn vehicles and declare that each one represents, say, a fleet of four or eight such vehicles.  This would be in keeping with the way the cities don't really represent a full-sized city grid.  But it feels odd.

This volume issue is independent of the general need for higher revenues in pak128.britain-experimental, which I think is also true.

As an anecdote, the funnest thing I did in my 19th century pak.britain experimental game was to hook up eight pairs of tram horses in front of a huge line of 56 goods wagons to drag fruit from the orchard to the greengrocer.  (A *lot* of fruit gets generated!)  Note that I was playing prior to the invention of signals, so I could only have one train per railway.  Ten of these and I was making an operational profit -- though only because I gave all the expensive stations to the public service player first.
So simutrans-experimental definitely allows stacking of horses on trams or trains.

This leads to another point -- compared to those tiny, tiny wagons holding 1 or 2 units each, the amount of goods generated by the early-years industries is frankly enormous.  Something in the volumes has to be rebalanced as well as something in the prices.

[Dutchman on Rails]

I was reading this thread with some interest, as I prefer to start playing in the pre-industrial revolution.

With Simutrans Experimental and Pak.German, I finally came across setting the simuconf.tab file to have station coverage=5, passenger generation=1 or 2, and building maintenance costs=2/5 of the default.

It is good to realise in this that the changed passengers and cargo handling changes your business plan. In Simutrans original with Pak.German, I've never had much trouble making any passengers route turn a profit. Once you have demand for the capacity two passenger coaches, it takes off, even with the maintenance cost set to default.

With experimental, I found that the percentage of used capacity is much more important, basically you want every vehicle to be full. On the other hand, with the lower maintenance cost, I find that just one vehicle can almost cover the costs of the stops, a second one (if used up to capacity) will turn a small profit.

Another feature of experimental is that a vehicle has to be able to make one round trip in less than a month. I've never seen any route making a profit where the vehicles can't do this.

Another experience is that the 16 barrel fluid goods cart in Pak.German is one of my greatest profit-makers - once gasthauses and markets show up around 1835. The points made about capacity of vehicles relative to stops are shown clearly in that effect.

With Pak.Britain-Experimental, I only have some minor experience from yesterday, but I do get the same impression the other writers to this thread point out. Part of the problem appears to be the slow speed of the available carts, most can achieve speeds of between 1/3 and just over 1/2 of their Pak.German equivalents. The result is that the distance between industries has to be correspondingly closer.

I hope these experiences help in gameplay for those wanting to set up shop as early as I try to do.

[Pugget]

In my limited recent play time, I've been using Pak128 with the suggested experimental conf files.  In my last two hames I've been able to eek out a profitable system fairly quickly using buses, but have not been able to convert that to a rail network later.  Or even make enough to upgrade the buses in the 50s as the old buses become out-dated.

My general impression is that this is largely an effect of population density.  When I start with larger initial cities, it's pretty easy to get a profitable system up and running.  I'm planning to play with a larger starting population next time and see if that doesn't increase revenue enough to break through to profitable railroads.

[Dutchman on Rails]

Since my previous post, I have found out which settings make most cargo routes in Pak128.Britain-Experimental profitable in 1750.

First, a sidenote. Some mentioned the mailbox, however in the version of Pak128.Britain-Experimental I use, the mailbox only becomes available between 1910 and 1915 (contrary to the standard version, where it is already available from 1750).

That said, my 'recipe' for making cargo routes profitable consists of two changes in Simuconf.tab:

1) Set maintenance_building to 200 (default = 1800).

2) Set bits_per_month to 20 (default = 18).

If, as I said before, you guard against round trips exceeding 1 month, most routes should be reasonably profitable with 2 or more carts. You won't make much (in terms of single pounds to maybe a few dozen per month on busy routes), but to me that enhances the feeling of the 1750's (especially with starting money set back to 15000).

[jamespetts]

A new version of Simutrans-Experimental is out (see here), along with a new version of Pak128.Britain-Ex (see here). That combination is designed to address the problems that people have had making sufficient profits (goods revenues are increased, and the overall industry density is maintained, so that industries are not closed down without (eventually) being replaced).

It is recommended that players wishing to test the new versions start a new game rather than continuing a previous game.

[zengrath]

Hi guys, i recently came back to playing Simutrans again and started reading through all ongoing projects, one which caught my eye the most was Simutrans Experimental, read all features and the basics of it, and it sounds awesome. I am always looking for a more realistic, or life like, makes game much more rewarding for me.

However as some of you have said in this thread, starting in 1830 seems to be completely impossible at this  point.

It's disappointing for me since i always start with timeline on and default starting year. I don't like to jump ahead as i normally find this makes simulation games to easy and less interesting for me.

I read a lot about updates regarding passengers services and this has really excited me, especially since it is supposed to be easier in beginning, but I've been completely unable to make a real profit, i only barely break even at this point, and that is not enough to make back the investments i put in before everything goes obsolete!

Basically i just started a new game, with updated 7.1 and worked and tried working on 2 large cities that are side by side. I ran the cheapest horse and the 5 passenger carriege since the 10 person carriage costs 6x more in running cost. I ran about 12 of them in each city, which seemed to meet the demand of passengers just fine ..

However problem i am seeing is that the bus stops and the depots maintenance costs are really ridiculous compared to profit you make and operational costs of running the horses...

For instance, after a few years of running to see what happens, with all my horses making a profit and running at max capacity i have this:
Monthly running cost: 19
Monthly profit:   ~120   Looks good so far!
Monthly Maintenance: -88 Ouch!
Monthly profit fluctuates between -13 and +13

The initial investments of bus stops, horses, and 1 depot is:
12 bus stops at 125 each = 1500
1 depot = 7000
1 horse plus 5 pass carriage = 1200, have 25 of them = total 30,000
Not including roads (only built one small one). Total initial cost is 38,500

Let's say i do manage to squeeze 20+ a month, (which i can't see happening endless i'm doing something wrong, again remember everything running at maximum capacity and at a profit per vehicle, adding more cities would only create more maintenance from bus stops, etc. and maybe a slight amount of vehicles increased per city but not enough).

It would take me approximately:  1925 months to get my return on investment, = to 962.5 years !?!?

Now it does get slightly better as i expand to more cities, but this means far more maintenance (roads etc) eventually needing trains due to passengers not able to take horses long distances... which even huge maintenance costs increases, and far more investment costs, no way to ever make it back..


I tried 3 times now, 2 times i started of slowly making sure everything running at max capacity and efficiency, and once taking my entire initial 500,000 and trying to connect everything at once and still could not make any profit even though i had plenty of passengers to move around.



So unfortunately it looks like i'm going back to simutrans standard for now. However i hope that this is worked on more and continues to be improved, if done right this could be much more interesting then Standard for me, but i'll wait and see..


Perhaps forget about 1830? and maybe concentrate on working 1920-30 up first? or maybe work on a better way of increasing maintenance cost with time (increased wages, material costs, etc), rather then fixed which is seems to be now?



I also tried starting industry, I made a profit however the return on investment was so bad, it seemed like i could never make money back before i would have to replace everything or factories close... Although i didn't play more then 10 years yet. I try to play in baby steps and master each stage before moving on, since i can't seem to make any profit within first 5 years, i don't see a reason to go any further as i would simply run out of money without making any profit to spend on expansions later in game.


Maybe i should start with industry more in beginning.. build everything i can as soon as possible with all large industry chains? however my problem with this is that, say i am successful with this, once i start on passenger services i will not have any money left to start, and still i don't see how making profit in passenger service will get better, i think i would simply be wasting my time..


If anyone playing this has been successful, please tips would be greatly appreciated.


EDIT: I wanted to post my savegame in hopes that someone could see something i'm doing wrong... but it's 160kb and i can only upload 100kb, i tried compressing but doesn't seem to make it any smaller, i guess savegames are already compressed very well.

[jamespetts]

Zengrath,

thank you very much for your detailed feedback: it is much appreciated. Can I just check - are you using Pak128.Britain-Ex 0.4? That is the pakset specifically balanced for Simutrans-Experimental. I have increased the revenues from all goods by 1.5x compared with the Simutrans-Standard version of Pak128.Britain, and some others have reported that this helps.

As the name suggests, Simutrans-Experimental is still in development, and Pak128.Britain-Ex is not fully balanced yet. I am hoping to undertake a rebalancing exercise soon, to balance it specifically for Simutrans-Experimental, with very realistic relative costs, and your feedback in that respect is most invaluable, in particular your detailed mathematical analysis of the relationship between the profits made by early transport and the maintenance of fixed infrastructure.

Simutrans-Experimental 7.0 allowed for the first time the maintenance and capital costs of stations/stops to be set independently of the capacity of those stations/stops, which should help to make it possible to use your analysis to reduce the capital and maintenance cost of road based infrastructure in the pre-automobile days to put it in line with the likely return on investment, and to enable players to make a profit by running a network of horse-drawn transport.

Note, however, that the profits that you should expect from horse drawn transport are considerably less than the profits that you could expect from a railway, and, in the 1830s, I should advise starting with a railway connexion, and using the horse 'buses simply as a way of transporting rail passengers from the railway station to their destinations in towns. As in real life, you will find this much more profitable than just running horse drawn transport.

I shall, however, look at considerable reductions in the maintenance costs of road infrastructure in the horse power days to match the likely profits returnable for the next version of Pak128.Britain-Ex.

[zengrath]

Thank you james for your reply, i apologize if i haven't had a chance to get on and respond, week days are hectic for me.

I will defnietly try the suggestions you gave me.

As far as versions i've used. I first tried the pre-packaged version, which according to your sticky http://forum.simutrans.com/index.php?topic=1894.0 uses: Simutrans-Experimental version 6.6 and Pak128.Britain-Ex version 0.2

I also tried the latest versions, i may have not done it correctly, but i simply downloaded the 2 Windows Binaries and Configuruation files here http://forum.simutrans.com/index.php?topic=4010.0, and downloaded experimental .4 from here: http://forum.simutrans.com/index.php?topic=4013.0

I extracted those into the directly for the pre-packaged version i already downloaded and extracted.. and let it overwrite. I do not know however if that was the proper way to install the updates. I may wait until next Complete install version just to make sure i have everything correctly installed.. Oh and also i did start a new game, so i didn't use previous saves as well.

Thanks, i will continue checking for updates and test any new versions that are released


I can say that i'm not an expert Simutrans player, so i could be making many beginner mistakes to. I've played a lot of other transport games, but been fascinated by Simutrans for a while now, just haven't had time and motivation to play for any length of time.


EDIT: i'm reading this in more detail now http://forum.simutrans.com/index.php?topic=1894.0 I may not have installed latest versions properly, i thought i could simply extract everything in correct directories but i could be wrong . I using Windows 7 currently.

[zengrath]

I tried following all the instructions for installing latest version here: http://forum.simutrans.com/index.php?topic=1894.0

However part i don't understand is the Make-Obj Experimental section, do i need to do this? When i click to download Makeobj-experimental.zip it requires a username and password.. Maybe this is not something i need? I'm sorry if i'm confused about all this.


I assume this is what i do, download latest simutrans, 102-2-1, install that with whatever pak, then download extract .4 version of brit experimental pak and extract that in it's own directly. Download simutrans experimental .exe and extract that to simutrans main directory, download config files and copy those into main directory sub folders /cfg and /txt. Right? I just don't understand the make-obj part, i never ran that program.

[jamespetts]

Zengrath,

thank you for your reply. From what you have described, it does seem that you are upgrading using the correct method. Apologies that the Simutrans-Experimental experience is still a little rough around the edges! If you want to upload a saved game >100kb, you can use files.simutrans-germany.com.

As to Windows 7, I am glad that Simutrans appears to work in that platform, as I don't think that it's been properly tested for Windows 7 yet. Thank you very much for your feedback - I will take into account your useful analysis when I come to work on balancing.

Edit: You don't need makeobj unless you want to make your own paksets or addons; however, it should not be asking you for a username and password for downloading it.

[zengrath]

Thanks! By the way, for your info, the Makejob link on that site is an FTP server. I finally realized that, I used an FTP program to try to connect but it just says "This is a private system - No anonymous login" ( i tried using filezilla, i'm not good with FTP and my current browser doesn't have FTP supported).

So just to let you know.. again link i'm describing is under Make-Obj section and it's the windows version.

But yes works under windows 7 perfectly for me! i'm using 32-bit right now, not 64-bit.