breakdown system, how it work?

[sobrano]

hello, can someone teach me how breakdown system work?
i really dont understand it

i have about 25 buses with condition between 80%-100% (likeilhood of breakdown=50%)
i have 8 tram with condition between 90%-100% (likeilhood of breakdown=10%)
i set manitence for tram 95% and manitence for bus 30%

i tested it for 8 month: and i got 5 tram's breakdowns vs only 3 buses brakdowns!!!
why??

buses have 4 negative points vs trams:
1.likeilhood of breakdown 50%;
2.manitence only 30%;
3.high numbers of veichles 25 vs only 8;
4.general conditions worse!

why trams have more breakdowns?
it seems a bug, someone can explain me please?
ty

 

[unmerged(273151)]

I believe the maintenance % only affects the speed at which condition decreases.
I "think" the wage of the technicians may affect the rate of breakdowns but haven't tested it yet.

Other than that, I don't think anyone actually has any idea how the breakdown system is currently working. May aswell just be a random number generator.

 

[immi]

probability doesn't work well with human perception
a 50% chance of breakdown wont give you a guaranteed number of breakdowns. It is just the probability of a breakdown.
tough if you played 1 million games and keep track of the statistics your numbers would be close to those 50%.
But just looking at one game/a couple of (game)years those numbers could be quite different.

for comparison: roll a dice 12 times. Statistics says that each number has the same amount of probability. But you won't have each number twice as a result in your experiment.(at least it is rather unlikely).
think back to all those hours of probability theory at school

 

[unmerged(273319)]

All I know is, happens way too often, especially with rail vehicles. It's super annoying with metro trains because they hold a lot of people and as a result, a lot of people get pissed off.

 

[sobrano]

probability doesn't work well with human perception
a 50% chance of breakdown wont give you a guaranteed number of breakdowns. It is just the probability of a breakdown.
tough if you played 1 million games and keep track of the statistics your numbers would be close to those 50%.
But just looking at one game/a couple of (game)years those numbers could be quite different.

for comparison: roll a dice 12 times. Statistics says that each number has the same amount of probability. But you won't have each number twice as a result in your experiment.(at least it is rather unlikely).
think back to all those hours of probability theory at school

lol nice, i have a degree in physic, so i know very well the theory of probability
naturally the events can have fluctuations.
but my analysis is not based only on last 6 month test!

i am playing cim by about 30 hours and still dont see likelihood percentual respected,
I highly doubt that my observations are part of a rare case of deviation from normal probability.

instead beginning to think that the length of the path of the vehicle between two stops in some way influence the breakdown
I observed something like that with waterbus too

It is as if the repairs take place at bus stops, and fewer stops = more breakdowns

what do you think guys?

 

[Illanair]

The chance of breakdown is a chance percentage - That means that every X days/weeks/months, the game will roll a die. If it rolls a number above the breakdown percentage, the vehicle will break down - below, and it will remain unharmed. Rinse and repeat.

 

[sobrano]

exact! this post is open because we are thinking that the game has some problem and dont respect your theory

 

[sobrano]

I believe the maintenance % only affects the speed at which condition decreases.
I "think" the wage of the technicians may affect the rate of breakdowns but haven't tested it yet.

Other than that, I don't think anyone actually has any idea how the breakdown system is currently working. May aswell just be a random number generator.

i think it too, but what a low condition affect? attractiveness and nothing else
i think low condition should influence breakdown rate

 

[Illanair]

exact! this post is open because we are thinking that the game has some problem and dont respect your theory

You must realize how many rolls it takes for a percentage system to make sense to a human. Sid Meier did a talk on this a while back, concerning percentage chances and players feeling cheated. Something along the lines of players expecting a 25-50% attack chance to almost always be successful, whereas a 75% and above to ALWAYS work, making them feel cheated and unfairly treated if the computer manages to hit those 0-25% twice in a row (or generally too often). Humans are geared for perfection, failure isn't always entertaining - especially not in Singleplayer Sandbox games.

 

[unmerged(272700)]

One missing factor from the equation so far is "how often does a vehicle do its check (or its roll) to see if it is going to have a malfunction or not". Is it every stop? Perhaps they do it after a set number of KM traveled? I haven't figured it out yet.

But I do sometimes get the feeling that things breakdown on me too often. Buy a new metro train, have everything set to 95 and then watch it fail a week later. Ok, maybe that is going to happen 5% of the time, hehe. I think Illanair's comments do bring an interesting light to the topic though.

But I think I will keep an eye on this a little more closely and see if I can try to develop a "feel" for what factors might be involved. How often is the check being made, etc. I'll never come up with a concrete number, but with time I may develop a general sense and that may subtly adjust the way I play.

 

[immi]

Sid Meier did a talk on this a while back, concerning percentage chances and players feeling cheated.

funny, that talk came to my mind as well
if i remember this right(or might from somewhere else) one of his conclusions was that at some point you have to screw the math to give the player the experience he expects.
After all this is not about correctness but about having fun

But i agree in so far with the OP that you didn't get enough feedback to feel the difference your settings may or may not make.
That might be just some fancy statistics to show how the breakdown rate of the different vehicles changes with time depending on your maintenance.
Or maybe the formula can also be tweaked that the randomness factor is toned down and condition/vehicle characteristics are more important.

 

[sobrano]

ok guys i have news:
i tested breakdown system for 5 game years.
1.with manitence 95% i spent -755 dollars/month and i needed to spend for repairs about 80/140 dollars (total about -865 dollars)

2.with manitence at 10% i spent -86 dollars/month and for repairs always about 80/140 dollars (total about -196 dollars)

so i can say:certain money spent for manitence dont reduce breakdowns!!

so now we need to know what can reduce breakdowns?
and why someone should spend about +600dollars/month to set high manitence?

 

[unmerged(272700)]

Was your test over the same exact 5 years? With the same lines running, the same vehicles and their starting condition? Because that would be required for an equal comparison.
But regardless, your findings are interesting.

I will say that in general, when I have my maintenance set very high... I get fewer breakdowns. This is, of course, as long as I sell the old vehicles when they have been used too long.

 

[sobrano]

i have setted very high speed for 4 years exactly (not 5 i am sorry) and I was watching the events without doing anything
anyway this gap is really too high!!!
any veichles was young because i did the test in berlin scenario in the 1924 so the oldest veichle was 1 year old

 

[woz]

Some one at the German forum have been modding this.
If you go to tips and tricks on the German forum you will see work around breakdown. I am not sure what figures you put in though.
http://translate.google.com/transla...l=en&safe=off&sa=G&biw=1024&bih=639&prmd=ivns

 

[sobrano]

so it is confirmed: it seems that there is something strange in breakdown system but i didnt understand very well what are saying in that forum...

i still didnt understand how the breakdown system work...

 

[woz]

so it is confirmed: it seems that there is something strange in breakdown system but i didnt understand very well what are saying in that forum...

i still didnt understand how the breakdown system work...

I did what they said. And it doesn't seem to be breaking down for me. So i put in 1.0 and 00

 

[sobrano]

breakdowns are heavy affected by distance and a little by condition

manitence costs are TOTALLY not balanced
it is much more usefull set manitence at 0%-10% and rebuy the veichle when its condition is low

 

[eagleFMJ]

i have a degree in physic,

Yet you can't figure out how to register your game on this forum?

Pretty interesting, you have been a member at Paradox forums since 2006 yet you have no registered games.

 

[douglasrac]

ok guys i have news:
i tested breakdown system for 5 game years.
1.with manitence 95% i spent -755 dollars/month and i needed to spend for repairs about 80/140 dollars (total about -865 dollars)

2.with manitence at 10% i spent -86 dollars/month and for repairs always about 80/140 dollars (total about -196 dollars)

so i can say:certain money spent for manitence dont reduce breakdowns!!

so now we need to know what can reduce breakdowns?
and why someone should spend about +600dollars/month to set high manitence?

Vehicle maintenance affects condition, and condition affects attractiveness not breakdown rate. Breakdown rate is probably not affected at all. If you bus have 20% of breakdown rate, will be 20% for the rest of his life.
Technicians wage affects how fast a vehicle recover from a breakdown.
So, breakdown just happens.