Feedback and questions from deMangler (merged)

[deMangler]

Hello community and thanks for a great game!
This, my second post, contains questions relating to my continued inability to make a profit in Simutrans Experimental.

First off, I have been playing Simutrans (and OTTD) for some time and while I am no expert I can pretty much have fun with stuff and make a profit. I started playing Simutrans Experimental because it seems like a much more satisfying underlying model, and most of the changes and the general direction of the thing seemed like where I would have gone if I had the time/ability. I would like to play this, I thought, and give feedback and maybe even be able to help in some way.
My first response to my approach with experimantal not working was to try lots of different optimisations,  both in network topology and in balancing maintainence cost/capacity, etc. All seem to end up with my finances slowly and consistantly bleeding dry.
My second response to having my efforts fail with financial ruin was to trawl through the forums, read feedback and check out some posted savegames to see if I could see what I was doing wrong by going over other people efforts.
I am pretty amazed by some of the things people have achieved, but I could see nothing that gave me a hint why my efforts in Experimental were failing.

So, now I ask:

Given my usual starting setup which is 1930, obey timeline, What would be a standard starting approach, maybe highlighting why Experimental might need some different methods from Standard?
I won't bore you with any more details about my failures unless requested, I can probably find savegames etc.

As an example, the type of game I have had most success with has been the one that starts out like the one I posted in the bug report thread.



Here is a save

This start up will be slightly worse than break-even, this is the best I have found out of several variations on this theme. I would then connect some power lines and set up some bus routes in the nearby town and a bus shuttle running workers to the factories.
The costs always seem to be just ahead of the proceeds, no matter how I scale. Sooner or later I will end up trying something out-of-the-box, which will finish me off.

Here is a later screenshot.



I ran it fast for a couple of years to introduce some optimsations, get a graph and an idea of where it was going. Line one is the coal train, 2 and 4 are the bus circular and it's reverse. being run by Leyland Clubs.
In previous games like this I will manage to scale up to many connecting routes but it will never become profitable.

One thing I have never tried is starting out by using the 500,000c to build a long range network. Hmmmm, maybe I'll try that next.....

In any case, I am doing something, or many somethings, wrong with Experimental.
Any advice appreciated.
dM

[Dutchman on Rails]

Hello deMangler,

Well, I'm perhaps not the best one to answer your questions, because I'm playing more in the 1700's - early 1800's. But I'll give it a shot.

First of all, you may want to check your maintenance_building settings. Bus stops can be very expensive at the normal settings, which are difficult to overcome.

Your train service has an obvious catch. You have a single train of 5 tile length. If you make a passing loop and run two trains with cheaper running costs, less powerful locomotives pulling 3 tile length trains, you save the maintenance costs of 4 tiles (which are quite expensive) for a cheaper extra section of track and maintenance-free signals, and your two trains should be able to make enough profit to overcome the maintenance costs.

The same should go for the other routes. Your vehicles should maximise profit by having the greatest performance for the lowest running costs (if in doubt, especially the latter), and your traffic density should be high enough to make the combined profit of the vehicles overcome the maintenance costs.

If this doesn't help, you can also try trucks, as roads are cheaper.

I hope this helps.

[deMangler]

Thanks for the help Dutchman,

I tried that rail link with a passing loop and 2 trains of length 3.



Actually, how about a less wasteful way of doing it...



It is definitely an improvement. I can see that I will need to be very careful of my maintanence costs, and be happy with very small margins.
I don't plan on changing any settings because one of the things I wanted to do was check out the balance of the default settings. I will continue to play with 1930 obey era defaults and post feedback.

<update>
I took the above beginning as far as 1945 with some success before it went titsup.

Here is a save in case anyone is interested
<update>

Any other suggestions welcome, I feel like I need all the help I can get!

dM

[Dutchman on Rails]

Hi deMangler,

If you feel it helps, I could start a game on the same map (if you have a savegame at startup for me), or a game with the same (default) settings, and report how it's going and how and why I make my decisions in my screenshots threat on tt-forums.

I think your passengers service in the first post needs more passengers. You need to have several buses full. I suppose making multiple lines to give more coverage to the rest of the city might help, but I'm not sure. Also, you might want to try a point to point (A-B-C-D-E-D-C-B-A) connection, or have the same bus run the circles in both directions (clockwise then counterclockwise or vice versa).

I hope this helps some more.

[deMangler]

Hi deMangler,

If you feel it helps, I could start a game on the same map (if you have a savegame at startup for me), or a game with the same (default) settings, and report how it's going and how and why I make my decisions in my screenshots threat on tt-forums.

I think your passengers service in the first post needs more passengers. You need to have several buses full. I suppose making multiple lines to give more coverage to the rest of the city might help, but I'm not sure. Also, you might want to try a point to point (A-B-C-D-E-D-C-B-A) connection, or have the same bus run the circles in both directions (clockwise then counterclockwise or vice versa).

I hope this helps some more.

That is a very kind suggestion, good idea Dutchman.
That would probably help a lot.

Here is a save

It is Simutrans 102.3 Experimental 8.2. Release date: 27th of June 2010.
Pak128.Britain-Ex downloaded from http://experimental.[ simutrans [dot] us (site down, do not visit) ]/Pak128.Britain-Ex.zip on 29/06/2010 (I can find no version number)

I may start another game with that map afterwards to see how it goes.
This looks like it might be fun....

dM

[Dutchman on Rails]

Well, Pak128.Britain-Ex is probably version 0.6.

I use tt-forums for my screenshots thread. A direct link is:

http://www.tt-forums.net/viewtopic.php?f=22&t=48765.

I need a few days to set up things and start posting.

[deMangler]

Hi,
I will use this thread for posting my observations on my Simutrans Experimental 1930 Obey Timeline games. I am a noob to Experimental, experimenting. I am not expecting feedback, just offering up my experience not knowing in advance what will be useful to the devs or other players.
Of course I will appreciate feedback or try to provide any other information about my experiences if asked.

Anyway,

Here is the tail end of a game I ran to investigate the differences between Standard and Experimental with bus routes.
The network was started as a few interconnected circulars that was then tweaked and evolved by random trying things out and survival of the profitable tweaks, rather than planning.
There are a few industry chains going, but more to supply realism and passengers than profit. I have avoided power lines because they don't seem to work in a way I can understand yet.
My observations are:
Setting up line graphs with operational costs, profit, maintainance and proceeds for lines and just see what happens, troubleshoot problems found here by setting up bus stop graphs with happy waiting unhappy, and vehicle graphs with spare capacity.
This is all pretty standard stuff perhaps, but I had to learn it all over again for Experimental, the new passanger modelling is satisfying, and I am finding that watching the behavoiur of interconnecting bus routes is a good way of beginning to appreciate it before moving on to more complex and mixed networks.
Things that are making it difficult for me are lack of mouseover info in the depot, not knowing costs before buying. Not a biggie as you can just buy a vehicle and run it to see, unless I am missing something.
A bigger one is due to me playing 1930 obey era. I just don't know what is coming up so I sometimes upgrade a line in a way that costs me money. Also, finding a vehicle that pays for itself before it goes obselete is part of the obey era challenge. Again, not unique to Experimental.

Really like the upgrade / replace system by the way.



Here is a save


It is Simutrans 102.3 Experimental 8.2. Release date: 27th of June 2010.
Pak128.Britain-Ex downloaded from http://experimental.[ simutrans [dot] us (site down, do not visit) ]/Pak128.Britain-Ex.zip on 29/06/2010

dM

[jamespetts]

dM,

thank you very much for your feedback - that is most appreciated. One or two preliminary points before I have time to respond in depth and look at your saved game: firstly, you seem not to be using the Simutrans-Experimental configuration files. This might account for your not being able to see the running costs, etc., in the depot window. See here for how to get the Simutrans-Experimental configuration files.

Secondly, starting in 1930 is not ideal (that is the default with the non-Experimental configuration files; with the correct configuration files, you would start by default in 1830). Not only do you miss going through the very interesting Victorian and Edwardian eras, which have many of their own unique vehicles, but the 1930s themselves are currently very lacking in higher density buildings in Pak128.Britain, so if you start during that time, towns become very low density and spread out, making it difficult to make a decent profit with passenger services.

Thirdly, you seem to be using the Standard version of Pak128.Britain. There is a Simutrans-Experimental version of that pakset, Pak128.Britain-Ex, which contains a number of additional Simutrans-Experimental features (such as specific comfort and loading time ratings for the individual vehicles) that makes playing in Simutrans-Experimental more satisfying.

As for playing with the timeline on and not knowing what is coming up - that is just as it would be in real life!

[jamespetts]

deMangler,

thank you very much for your post and feedback, and welcome to the community! I am glad that you find Simutrans-Experimental interesting.

The first thing to note, however, is that the cost/profit balancing of Pak128.Britain-Ex is not complete. Proper balancing, to work well with Experimental as opposed to Standard, is a very significant undertaking, and the Standard version is not even fully balanced yet. It may well be that it is not possible to make a profit in all cases because the settings have not been properly balanced.

The second thing to note, as I have pointed out in another thread, is that you seem to be using Simutrans-Standard configuration files, which will not have the optimum settings for Simutrans-Experimental.

The third thing to note is that you are starting in 1930, which is not ideal, as the town density is very low when starting in that era, as people have not got around to building higher density buildings from that era yet. It is very hard to make a profit with passengers and mail with low town densities.

I hope that this is helpful; do let me know if you have any other feedback :-)

[deMangler]

Thanks for the reply, I posted this reply in the other thread as well.
In games like this balance is a relative thing, a compromise between realism and playability that can only be achieved with a lot of testing and feedback. I like your plan with experimental, but I am no C programmer so I thought I might be able to help with feedback on a consistent set of games at least.
I was using the latest Experimental Pak 128 Britain, and I *thought* I was using the experimental config files, but it appears they unzipped to the wrong location, so I was using the standard config by mistake.
The default Start date for my games *was* 1830, but I just decided to start all my games in 1930 to give feedback to compare with Standard. I may take your advice and start at 1830, it sounds interesting,  I live in Swindon, and you can't move for GWR stuff (round here, a lump of ancient locomotive embedded in concrete and stuck on a plinth is art).
Anyway...
I hope to be able to give more meaningful feedback now that I am not using incompatible (or incomplete) configs.
Thanks for the heads-up.
And thanks for the welcome.
dM

[jamespetts]

Feedback is always helpful. I have merged your two topics, however, as there seems to be some repetition between the two. I should be grateful if you could try to avoid making two identical posts, too.

In any event, I should be very interested to know how you get on starting in other eras (pre-1920; if you start then, and progress into the 1930s, you do not get the same problem with city densities), and with the correct configuration files.

Happy playing!