News:

Use the "Forum Search"
It may help you to find anything in the forum ;).

Help with revenues calculation in Simultrans

Started by frank2016, November 25, 2016, 04:58:52 PM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.

frank2016

Hello all,
I am new to Simutrans I first downloaded the Standard version from Steam and I really liked it once understood the base mechanics I started to look for more advanced ones and I realised that there was Simultrans Experimental which was even more  technical than the standard version and which I like a lot. What I am playing is the Britain pakset and I started the game in 1750. So there are few things about revenues that are unclear:
1) I have this short route for furniture which is 16 tiles. My 10 crates cargo has operating cost of 0.75c/km that matches the total cost of 3c for roundtrip. In the goods list each piece of furniture should pay 0.41c for 2 km which means that 10 crates should be make a revenue of 4.1c; however they only make 2.3c resulting in a loss of 0.7c for each trip. There's no speed bonus so I wonder if there is something I am missing here in the calculation.
2) What does it mean when the detail window of each cargo is opened there's a line that says base profit per km (0.88c in the above case) which I don't understand what is referred to

Finally a bit off-topic: I cannot make the barge pulled by the horse to load livestock while Norfolk wherry sailboat has no problem doing it

Thank you very much for the game it is fun and addictive! :)

Ters

Simutrans can be configured to use one of three formulas for calculating revenue, but the key difference in any case is that operating costs scale by the distance driven along the road/tracks, while profits are calculated from distances as the crow flies. So for optimal profits, the route should be perfectly straight.

frank2016

Thank you very much! All the numbers make sense now. I don't know how real it is to get paid for the euclidean distance rather than actual distance travelled probably there should be a compromise betwen the two but at least now I can plan the network better!
Thanks again

Ters

Unfortunately, it is difficult to code the game to understand that the player took a big detour because of a serious obstacle and not because the player was silly. Especially since what is considered a serious obstacle changes with time.

DrSuperGood

Quote
I am new to Simutrans I first downloaded the Standard version from Steam and I really liked it once understood the base mechanics I started to look for more advanced ones and I realised that there was Simultrans Experimental which was even more  technical than the standard version and which I like a lot. What I am playing is the Britain pakset and I started the game in 1750. So there are few things about revenues that are unclear:
1) I have this short route for furniture which is 16 tiles. My 10 crates cargo has operating cost of 0.75c/km that matches the total cost of 3c for roundtrip. In the goods list each piece of furniture should pay 0.41c for 2 km which means that 10 crates should be make a revenue of 4.1c; however they only make 2.3c resulting in a loss of 0.7c for each trip. There's no speed bonus so I wonder if there is something I am missing here in the calculation.
2) What does it mean when the detail window of each cargo is opened there's a line that says base profit per km (0.88c in the above case) which I don't understand what is referred to
Is this experimental or standard?

In standard each tile is 1 km. And payment depends on the payment model used. Default payment models for pak64 and pak128 are the difference in distance between stops. Some people like to use the distance from source to destination.

Quote
Simutrans can be configured to use one of three formulas for calculating revenue, but the key difference in any case is that operating costs scale by the distance driven along the road/tracks, while profits are calculated from distances as the crow flies. So for optimal profits, the route should be perfectly straight.
Between payment milestones only. One could bounce cargo back and forward as much as one wants if payment model allows.

Quote
I don't know how real it is to get paid for the euclidean distance rather than actual distance travelled probably there should be a compromise betwen the two but at least now I can plan the network better!
Thanks again
Both running costs and payment use Manhattan distance.

Quote
Unfortunately, it is difficult to code the game to understand that the player took a big detour because of a serious obstacle and not because the player was silly. Especially since what is considered a serious obstacle changes with time.
That is not what needs to be coded. People have no problem with this in real life, as long as it is economical to do so. What governs it in real life is how much one is willing to pay for transport.

Ters

Quote from: DrSuperGood on November 25, 2016, 09:00:19 PM
That is not what needs to be coded. People have no problem with this in real life, as long as it is economical to do so. What governs it in real life is how much one is willing to pay for transport.

I don't think anyone is willing to pay someone for shipping the goods twice around the world to get to a destination just ten kilometers away across safe ground, even if they can afford it. Once around the world could be for novelty reasons, but even that is rare. In real life, time is also a factor.

DrSuperGood

Quote
I don't think anyone is willing to pay someone for shipping the goods twice around the world to get to a destination just ten kilometers away across safe ground, even if they can afford it. Once around the world could be for novelty reasons, but even that is rare. In real life, time is also a factor.
People do not care how many times something gets shipped around the world as long as it makes economic sense. Real life has exactly the same problem as Simutrans except instead of companies shipping a product physically around the world multiple times they just add an inflated markup to shipping costs to produce an inflated shopping cost which in some cases might as well be around the world multiple times.

What generally stops this happening in real life is someone else will then offer shipping cheaper, with a lower markup, and so win the contract. Hence markup values are kept low due to competition. To make more profit companies seek out more efficient ways to deliver so they can under cut competitors or have larger room for margins.

What does not happen in Simutrans is this competition approach. I cannot adjust my margin and hence I have to ship around the world to do so. If a competitor offers a more efficient/shorter route the company will still use me even though it makes no sense to do so. If I am the only player running a route the company has no choice but to use me however it will still use me no matter what I ask, even if it is unreasonable like 100,000 per km unit which might as well be no route at all as no one can afford that.

Ters

Cheaper and/or faster. And in my example, the sender or recipient might chose to just do it themselves.

The Øresund bridge is competitive with the ferries between Denmark and Norway, despite being a detour when looking at the maps (except if going to northern Norway). Given the right circumstances, the bridge route is faster, however the ferries have an advantage in that it lets a single driver keep moving and take the mandatory rest at the same time. I don't know exactly how the cost of using either compares. It is also possible to send only the trailer with the ferry, eliminating the driver and associated costs altogether.

frank2016

The problem that I see in Simultrans is for those goods with low margins for which the only way to make profits is really a straight line trip. If you do an L-shaped trip without strange fancy traveling around the world you still lose money!!

Ters

Quote from: frank2016 on November 26, 2016, 05:41:45 PM
The problem that I see in Simultrans is for those goods with low margins for which the only way to make profits is really a straight line trip. If you do an L-shaped trip without strange fancy traveling around the world you still lose money!!

Yes, because as I wrote, Simutrans does not know the definition of "strange fancy traveling" and thus can not tell if your detour is justifiable.

Vladki

I thought that you can switch the revenue calculation mode between shortest path and really travelled path, somewhere in simuconf.tab

Ters

Quote from: Vladki on November 27, 2016, 11:38:07 AM
I thought that you can switch the revenue calculation mode between shortest path and really travelled path, somewhere in simuconf.tab

No, all options take only shortest path into consideration, the difference is what the shortest path is between.

frank2016

What about having a weighted average between manhattan distance and euclidean distance? I think that the code for the manhattan distance is already in simultrans standard so it shouldn't be difficult to implement and then you might give the possibility to set the weights in simulconf .
I would like to have this possibility however i'm totally fine with the game as it is now :)

Ters

When we write "straight line" in Simutrans, we're talking about Manhattan distance. The non-straight alternative is along whatever bends, twists and loops the vehicle is moving. So dragging euclidean distance into it is only going to make it worse.

River

a simple solution would be to use the manhattan distance and add a small percentage to take in to account the "strange fancy traveling".

frank2016

Ok I really got it now! Manhattan distance is fine to me; the route I was talking about was losing because the little detour was necessary otherwise I needed to bulldoze an house to have an L-shaped trip which of course didn't seem fair to me. So basically for short trips it's very difficult to make up for very little detours but I do see the necessity to avoid to pay the "strange fancy travelling" that you could create with waypoints

DrSuperGood

The reason the payment model in Simutrans is so exploitable is because it is so simple. There are no ways to adjust profits if you are a monopoly. There are no limits to how much one can charge to move goods between two points.

In your example you would want to charge a premium to cover your detour expenses. The factory should only ship if what you ask is still a reasonable cost from source to destination. It should only ship via you if you are the most reasonable, or only reasonable, option available to it.

frank2016

So I am trying to do the math but to me it seems that I am getting paid less than what I should get. Can you please help me with this?
I have a route from the slaughter house to the market (which is not the route I was talking about earlier).  The slaughter-house stop has coordinates 104,16 the market stop has coordinates 119,43. So now If I understood correctly what I should get is (119-104)+(43-16) + 1=43 (I added a plus one if you want to count also the starting tile)  then (43/8)*0.18*6=5.805 . Where 0.18=price of the meat per km and 6 is number of crates carried. What I actually get in revenues is only 4.41. Can you see where my calculations went wrong?
Thanks

DrSuperGood

The distance is measured from the convoy start point to the convoy unload point, using the convoy head as reference.

Speed bonus is in percent per 10% speed bonus speed difference with average maximum speed. Speed bonus allows goods to be worth values that include fractional currency units.

Ters

Quote from: DrSuperGood on November 28, 2016, 03:51:51 PM
The distance is measured from the convoy start point to the convoy unload point, using the convoy head as reference.

To me, it looks like it calculates distance for each individual vehicle, so for a train that flips around at either end (and doesn't do an odd number u-turns along the way), the rearmost cars will pay less.