News:

Simutrans Forum Archive
A complete record of the old Simutrans Forum.

Stop and station balance issues

Started by freddyhayward, July 31, 2020, 06:52:26 AM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.

freddyhayward

I don't have the relevant data to support changes, so I can only point out apparent inconsistencies. I'm sure that there have been planned or proposed changes to them in the past, so perhaps those could be revived.

Figures are from the in-game display, not the .dat files:

* A staging post (including the railway staging post), represented by a wooden pole in the ground, has:
- a passenger capacity of 10
- a maintenance cost of 0.96¢

* A staging inn, represented by a medium-sized building with accommodation for people and horses has:
- a passenger capacity of 40 (4x the above)
- a maintenance cost of 1.12¢ (hardly more than the above)

* A large staging inn, represented by a larger version of the above has:
- a passenger capacity of 120 (3x the above)
- a maintenance cost of 8.00¢ (over 7x the above)

* A basic bus stop, represented by a wooden pole and bench has:
- a passenger capacity of 12 (hardly more than the staging post, over 3x less than the staging inn)
- a maintenance cost of 2.40¢ (2.5x the staging post, over 2x the staging inn)

* A bus stop with a shelter, represented by the above with a basic wooden shelter, has:
- a passenger capacity of 25 (over 2x the above, 2.5x the staging post, over half the staging inn)
- a maintenance cost of 13.44¢ (5.6x the above, 14x the staging post, 12x the staging inn)

* A wooden platform, represented by a raised platform with signage and two benches has:
- a passenger capacity of 0 (10 less than the railway staging post, 12 less than the basic bus stop)
- a maintenance cost of 9.92¢ (10x the railway staging post, over 4x the basic bus stop)

* A wooden station building, represented by a small building with modest facilities has:
- a passenger capacity of 50 (1.25x the staging inn)
- a maintenance cost of 27.36¢ (nearly 25x the staging inn)

Sirius

I agree, from a gameplay standpoint this is very strange and inconsistent, causing the anomaly that players usually won't want to upgrade their staging posts.

About capacities, I am quite sure it is bases on some ingame considerations rather that real-world data.
About costs, I have no idea about their relation, as I do not have any data about staging posts, but I remember some data about modern railway stations and bust stops. I'd have to search for this in the web again to say anything definite about this, but the cost was quite a bit.

I can imagine simply putting a wooden pole does not come with much expenses in both cases.

However, the capacity of a platform with a bench should definitely be more than 0.
The implication that comes with it is "nobody can wait at that platform, so those benches are not allowed to use and passengers cannot stand there", whilst they can stand at the staging post.

jamespetts

This is a byproduct of the costs not being balanced yet - ultimately, we will need proper, data-based balancing of all stop types, but all the balancing enabling works and balance critical features will need to be completed first.
Download Simutrans-Extended.

Want to help with development? See here for things to do for coding, and here for information on how to make graphics/objects.

Follow Simutrans-Extended on Facebook.

freddyhayward

Quote from: jamespetts on August 01, 2020, 11:31:27 AMThis is a byproduct of the costs not being balanced yet - ultimately, we will need proper, data-based balancing of all stop types, but all the balancing enabling works and balance critical features will need to be completed first.
Given that we don't know how long these will take, would it not be far better make interim changes to make it somewhat more reasonable and consistent?

Sirius

#4
I agree with freddy.
As stations are not balanced yet and a real-world balancing does not make much sense without the related, yet unimplemented features, we should go for an interims balancing that encourages players to upgrade their stations and bus stops.

jamespetts

Does anyone have any suggestions for how to set out an interim balancing for the stops for internal consistency?
Download Simutrans-Extended.

Want to help with development? See here for things to do for coding, and here for information on how to make graphics/objects.

Follow Simutrans-Extended on Facebook.

PJMack

I think for now, the best way to balance the station costs would be to have the maintenance and construction costs fixed per unit capacity.  A surcharge could be added for rail platforms, as platform length does effect game-play.  I would also adjust some of the capacities for extension buildings to prevent older buildings from having higher capacities as the newer ones.

jamespetts

One thing that I noticed from some of the balance calculations carried out over Christmas 2021 was that the overall recurring cost for 19th century railway companies for stations was much, much higher than the equivalent amount being paid by players on the Bridgewater-Brunel server game, which at that time was in the 19th century. To have meaningful balance, we need to address this issue. A real complexity of this is 'bus stops, which in reality have very, very little ongoing cost; but players need not to have an incentive to make all their airports a runway and 100 'bus stops for capacity, for instance, to circumvent realistic costs for air terminals.

Also, there needs to be some balancing that takes account of different types of stop: earlier, wooden structures need to have a lower construction cost but a higher maintenance cost, whereas later brick/stone/metal structures need to have a higher construction cost but lower maintenance cost.
Download Simutrans-Extended.

Want to help with development? See here for things to do for coding, and here for information on how to make graphics/objects.

Follow Simutrans-Extended on Facebook.

DrSuperGood

Airports likely need additional constraints rather than worrying about bus stops adding capacity. Specifically a functional airport, one which the parking bays are valid targets for aircraft to go to, could be made to require terminal buildings and other expensive airport structures. To support bigger aircraft, bigger and more expensive terminal buildings could be required. These buildings could automatically add sufficient passenger capacity to the airport, meaning there is little incentive to spam cheap bus stops to boost it further.

jamespetts

Quote from: DrSuperGood on May 21, 2022, 05:25:15 PMAirports likely need additional constraints rather than worrying about bus stops adding capacity. Specifically a functional airport, one which the parking bays are valid targets for aircraft to go to, could be made to require terminal buildings and other expensive airport structures. To support bigger aircraft, bigger and more expensive terminal buildings could be required. These buildings could automatically add sufficient passenger capacity to the airport, meaning there is little incentive to spam cheap bus stops to boost it further.
We already do this with control towers - but for railways, this is more complex, since railway station costs show as a very high proportion of 19th century railway accounts, and it is more difficult to do this with railway stations, as unstaffed halts were very much a thing.
Download Simutrans-Extended.

Want to help with development? See here for things to do for coding, and here for information on how to make graphics/objects.

Follow Simutrans-Extended on Facebook.

KneeOn

I agree with PJMack. 

Per unit capacity would work to balance stops in the interim across all transit types. 

Eventually this can be scaled to fit a per-unit-km for global balancing strategy - this may depart from the data driven, historically accurate goals.

wlindley

In 1950 we have the Bulk goods bunker, Car yard, Fluid storage tanks, livestock pen, and Goods warehouse all with the same costs and capacities — except the Goods warehouse also stores as much Mail (i.e., it has twice the total capacity and thus may represent twice the value).  So these represent merely eye-candy instead of giving us choices.

One imagines a car yard should account for one automobile being perhaps 100 times as large as one crate of goods; and that a coal bunker would cost less than a warehouse building.

At least there should be four different capacities each priced accordingly. 

Even better, perhaps there could be a way for station extensions to specify which kind of goods they handle, just as passengers and mail are handled separately.

Octavius

Quote from: jamespetts on May 21, 2022, 07:12:55 PMWe already do this with control towers - but for railways, this is more complex, since railway station costs show as a very high proportion of 19th century railway accounts, and it is more difficult to do this with railway stations, as unstaffed halts were very much a thing.
The reason is that stations were far larger than would be required just to let people board and alight trains and sell tickets. Some of the additional space could be used for a restaurant, which generated additional profit and increased tolerable journey time; this also applied to staging inns. But more important than that was that stations served as living space for railway employees. As these people didn't have to rent their own living space, wages could be kept lower.

Sirius

Quote from: wlindley on July 02, 2022, 10:44:50 PMEven better, perhaps there could be a way for station extensions to specify which kind of goods they handle
I like the idea.
I suspect the issue is about much more than just splitting capacities.
For example, consequences of station overloading practically doesn't matter pretty much on freight networks, so I'd rather overload my freight hubs than paying much for capacity.

In any case, splitting capacities should be a rather simple improvement that can quite much help with balancing the costs.
Storig a car might be much more expensive than storing a cow and storing a unit of cooled goods will most likely be much more expensive than storing a unit of non-cooled piecegoods.

Quote from: jamespetts on August 02, 2020, 10:52:47 AMDoes anyone have any suggestions for how to set out an interim balancing for the stops for internal consistency?
A fair interims solution could be to calculate costs from two components:

1. Cost per unit of capacity
Each unit of capacity has a cost, which is consistent between all types of transport.
That cost may (and should) consider the type of good for sure(pax/mail(freight for now), but not the type of transport.

2. Cost per-tile of transport type
This cost is about "enabling" a specific type of transport to stop at a tile.
For example a tile of bus stop, railway platform, dock or airplane apron.
Extension buildings don't enable any type of transport to stop there, so so this doesn't apply to them.

That way, we get consistent costs between transport modes whilst we can still adjust costs per transport mode, so railway infrastructure will be more expensive than bus infrastructure, without encouraging people to build cheap bus stops to increase their airports capacity.

There is one last step missing: Mapping that consistent number to construction and maintenance costs.
To get this done, we use a percentage of that result as monthly cost, whilst the remaining part multiplied by a consistent number is the construction cost.

So the balancing parameters will be:
unit_cost_pax, unit_cost_mail, unit_cost_goods
base_cost_road, base_cost_rail, base_cost_ship, base_cost_air
material_percentage
construction_cost_factor

Then the cost is calculated as follows:
maintenance_cost := (base_cost + unit_cost*capacity) * material_percentage
construction_cost := (base_cost + unit_cost*capacity) * (1 - material_percentage) * construction_cost_factor

Might play a bit with the numbers to get a result that is not too far away from the current balancing.

Matthew

Quote from: wlindley on July 02, 2022, 10:44:50 PMIn 1950 we have the Bulk goods bunker, Car yard, Fluid storage tanks, livestock pen, and Goods warehouse all with the same costs and capacities — except the Goods warehouse also stores as much Mail (i.e., it has twice the total capacity and thus may represent twice the value).  So these represent merely eye-candy instead of giving us choices.

One imagines a car yard should account for one automobile being perhaps 100 times as large as one crate of goods; and that a coal bunker would cost less than a warehouse building.

At least there should be four different capacities each priced accordingly. 

Even better, perhaps there could be a way for station extensions to specify which kind of goods they handle, just as passengers and mail are handled separately.

The arguments in this post are all good. However, I do not think the first/pakset proposal (give different capacities to different extension buildings) should be implemented until the second/game proposal (restrict extension buildings to particular goods types) has been implemented.

If the coal bunker is the smallest type of freight extension building, then it will usually economically sensible to build coal bunkers next to cattle and sheep farms (because their monthly output is relatively low). But visually, that is all wrong. Cattle and sheep farms should have livestock pens. There would be many such visual inconsistencies. I think that would be worse than the current set-up.
(Signature being tested) If you enjoy playing Simutrans, then you might also enjoy watching Japan Railway Journal
Available in English and simplified Chinese
如果您喜欢玩Simutrans的话,那么说不定就想看《日本铁路之旅》(英语也有简体中文字幕)。

jamespetts

I agree with Matthew - the choice of having lots of different types of goods extension buildings with identical capacities and price is deliberate in the absence of being able to have extension buildings with a specific type. Having different capacities of extension buildings associated with different visual appearances but not the ability to handle different kinds of freight is a broken gameplay mechanic.
Download Simutrans-Extended.

Want to help with development? See here for things to do for coding, and here for information on how to make graphics/objects.

Follow Simutrans-Extended on Facebook.

KneeOn

Sirius describes a solution - I struggled to even describe the problem! Moving from freight, airports are much, much larger buildings than railway stations 

Gatwick handled circa 46m passengers in 2019.
During the 2018-2019 reporting period, Waterloo handled 94m passengers.

Gatwicks two terminals cover a combined area of 258,000m2
London Waterloo covers 99,000m2 - about the size of the North Terminal of Gatwick Airport.

The running costs of each are not publicly available however I'd imagine at almost 150,000m2 more space, the cost to run the airport is much larger. I think Sirius deals with this well. 

One area I do disagree on with Matthew and James is the issue of having economic vs visual reasons for building specific extensions. At some point the developer/pak-set maintainers need to trust the player to play the game within the spirit of the rules. 

Indeed, flexibility in the game engine is important to allow newer players some forgiveness. I give an example of Football Manager. There are various underhand tactics which could be written out of the game with relative ease
1. Save scumming - saving before a big game and re-loading if your team loses until you get the desired result
2. Perpetually adding yourself back in to your team - by retiring a sacked manager and creating a brand new one with the same name to take back over your team.
3. 'Dodgy lasagne' - taking over a team you're about to play with a new manager and playing goalkeepers up front and strikers in goal from the youth team
4. Official tools - turning off attribute masking (so all players states are open without the need to scout), using the ingame editor to see hidden attributes (such as injury proneness, big game readiness etc) or to even create a team of wonderkids, give yourself an unsackable contract etc. 

If you are new to Football Manager, a very large and complicated game, you may resort to one or more of these tactics to ensure you make it to the end of your first season without getting sacked and then starting again, never seeing it past February of year 1. 

Extended is a very complicated beast and adds complexity on a game that is somewhat known for having a higher learning curve (albeit made easier recently with tutorials). 

My own take is that while the feature discussed is good, I do not think it would be an inherent problem to have coal bunkers/livestock pens/warehouses all having their own values in the interim. In fact it may help highlight where specific goods values (either the income for transporting or the expenditure for the infrastructure) is not balanced more quickly. If it is never profitable to have the right type of extension building for a given type of freight, that would be a problem.