Fix exponential inflation

[Hyronymus]

As the result of my remark in the Dutch Trainset about the quick rising of vehicle costs I wonder if there is a way to fix the exponential inflation cost. Let's start with quoting a few posts from the Dutch Trainset topic:

I'm guessing you're playing with inflation? And started very early?
In that case prices will explode exponentially, making differences larger by the year. Given that the inflation in the game is a bit flawed, I recommend playing without it. With inflation it's nearly impossible to get the prices right.

I am but I don't get the inflation issue. How is that flawed and causes the engines to be so expensive?

EDIT:
Come to think about it, the issue probably is that vehicles are priced as the [price according to GRF] * [inflation] ^ [spent gameyears]. If you started in 1910 and but an engine in 1911 it'll be way cheaper then when you start in 1800 and buy it in 1911.

It's flawed because costs inflate more than income. So if you start early enough trains will end up costing more than they will bring.

That the difference in prices are so big is indeed because of what you explained

It's flawed because costs inflate more than income. So if you start early enough trains will end up costing more than they will bring.

That the difference in prices are so big is indeed because of what you explained

Sounds something that needs a fix. Food for thought.

It's flawed because costs inflate more than income. So if you start early enough trains will end up costing more than they will bring.

last time i crunched the numbers, costs relative to income were around factor 5 higher after 170 years of inflation

So the problem lies in an unequal application of inflation towards cost and income and the game years spent prior to purchasing vehicles. Can these problems be solved by changing the inflation code?

[andythenorth]

Just bin inflation, it's not a great feature, it doesn't do much for gameplay

[XeryusTC]

AFAIK this is how inflation is intended to work, it is meant to give you this (stupid) extra challenge of unequal increase. It is a s*** feature anyway

[PikkaBird]

http://bugs.openttd.org/task/4799

[michael blunck]

An inflation model with no divergence (income ./. costs) makes no sense.

regards
Michael

[PikkaBird]

An inflation model with no divergence (income ./. costs) makes no sense.

From a gameplay point of view, it would be purely cosmetic.
From a "realism" point of view, it's probably the only model that makes sense. After all, in a healthy economy, costs keeping parity with income is basically what inflation is.

Divergence between costs and incomes as time goes by can be managed by GRFs, in whatever way the creators of those GRFs see fit.

[Robbedem]

From a "realism" point of view, it's probably the only model that makes sense. After all, in a healthy economy, costs keeping parity with income is basically what inflation is.

This would only be true if there is no technological advancement and since in OpenTTD there is a lot of technological advancement, there would be a difference in inflation between cost and income.
like the introdutction of mass production: the income per product is lower but because there are so much more products the total income is higher

i see two options:
1) change the running costs of vehicles in the NewGRF taking into account there introduction date.
this would propably require two running costs per vehicle: one when inflation is active and one when its inactive. I think thats too much work.

2) change the inflation code so it doesn't increase too much (make the difference between cost and income inflation smaller and/or have deflation with recession)

[Hyronymus]

The real problem I have with inflation is that it makes vehicles too expensive given their introdcution date. Vehicle costs and running costs should not take into account the years already played. Yet this may result in another undesirable effect of vehicles becoming too cheap.

What if vehicle costs aren't affected by (inflation) ^ (amount of years passed) but by (inflation) ^ (amount of decades passed).

[michael blunck]

http://en.wikipedia.org/wiki/Quantity_theory_of_money

http://en.wikipedia.org/wiki/Austrian_School

regards
Michael

[andy518]

I believe that each cargo, except for passengers, valuables and mail which would have have 1% inflaton per year, should suffer inflation or deflation depending on much much cargo is transported. The avarage cargo inflation rate would be 1%.
To calulate inflation rates for all cargos, except passengers, valuables and mail, you would:
1)Add up the amount of cargo transported.
2)Divide the number caluclated in step 1 by the number of different cargos to get the average amount of cargo transported for each cargo type.
3)For each different cargo, you would divide the actual amount of cargo transported by the average amount of cargo transported and then multiply that number by 100. This gives you the percentage of cargo transported compared to the average amount of cargo transported.
4) For every 10 the value you get from step 3 is above or below 100, you add or take away 0.5% away from the average inflation.
5) Repeat steps 3 and 4 for each cargo type.
6) If there is a sudden fall or increase in transportation of cargo, defined as a 30% difference or greater in cargo transported within 3 months, there is a fall or rise in inlflation of 10%. The maximum rate of inflation or deflation is 15%

The rate of inflation for running costs would be 1% per year, matching the average rate of inflation.

[Robbedem]

Or what about putting 1950 (or 1900) as the standard and every year after 1950 has x% inflation and before x% deflation ->
(1,0x)^(current year - 1950)
This means that in the years before 1950 the game will be slightly easier.
There might be a problem of too small numbers though.

[andythenorth]

Why not just patch currency display?

So every time currency is displayed, do

Code: Select all

Length of game * 1.15 * amount of money

Inflation has zero usefulness for gameplay, so all that's needed is a patch to fiddle with the numbers, for those who care.

Or just remove it, and cut out another piece of cruft and source of non-interesting bug reports.

[michael blunck]

Inflation has zero usefulness for gameplay, so all that's needed is a patch to fiddle with the numbers, for those who care.

As written above:

An inflation model with no divergence (income ./. costs) makes no sense.

OTOH, the current TTD behaviour makes indeed sense with costs having a slightly higher inflation rate than income. This implements a penalty for the later game, trying to outweigh the money surplus from an established network. But as already written, same numbers for income and cost would be just silly arbitrary numbers.

Or just remove it, and cut out another piece of cruft and source of non-interesting bug reports.

It´s already possible to play with inflation set to "off".

regards
Michael

[PikkaBird]

For those who don't quite understand the problem:

I'm making a road vehicle GRF and I create a vehicle called the "Plozzowow Rootburger", a coal truck which is used between 1980 and 2000.

To test it, I create a quick game starting in 1980, the year the vehicle is introduced. My tests show that the Rootburger earns about twice as much as its running costs; perfect! So all is well and I release the GRF.

Simone Nooby downloads my GRF and starts a game with it in the game year 1920. By the time her game reaches 1980 and the Rootburger is introduced, TTD's inflation model has eroded its earning capacity to the point that it barely breaks even. Simone tries to use Rootburgers, but after a while decides they're stupid and useless. She removes my GRF and goes back to the much cheaper default road vehicles.

Basil Proplayer wants to play a 19th century game. He loads up his sailing ships, his horse wagons, and the Rootburger. By the time the game gets to 1980, TTD's inflation model means the Rootburger - along with every other vehicle in the game - is haemorrhaging cash at a tremendous rate. Basil concludes that the game is broken and stops playing it.

Or what about putting 1950 (or 1900) as the standard and every year after 1950 has x% inflation and before x% deflation

While this would fix the problem of wildly different economies depending on start date, it would not fix the problem of wildly different economies with inflation on or off, particularly at dates a long way from the epoch. Inflation being purely cosmetic is the only sensible solution to this problem.

[Robbedem]

Would it be possible to ignore inflation for running and purchase costs from NewGRF, while keeping the inflation for the rest? (building tracks, default vehicles, funding industry, landscaping,...)

[Supercheese]

Would it be possible to ignore inflation for running and purchase costs from NewGRF, while keeping the inflation for the rest? (building tracks, default vehicles, funding industry, landscaping,...)

Theoretically, given enough mucking around with the source code. Not entirely sure why you'd want that, however.

[jonty-comp]

They's some sexy graphs.

[Jacko]

my best solution, whilst pointless for the point of this thread..
is to play with inflation off.

Though I remember once on TTD a looooooong time ago when it got so much i ended up with a negative Company value...lol

[Robbedem]

Would it be possible to ignore inflation for running and purchase costs from NewGRF, while keeping the inflation for the rest? (building tracks, default vehicles, funding industry, landscaping,...)

To be clear: I mean that the NewGRF running and purchase costs should be inflated like income, while everything else remains as it is.

Why do this?
This would solve the problem of balancing NewGRF's for games with inflation off or on.

[Hyronymus]

I have a brain wave after reading Robbedem's suggestion about pre and post 1950 calculation. How about making inflation a proportional exponent of the passed game years divided by the amount of years between 2050 (default end year) and the start year rather than the amount of passed years. To prevent unwanted effects from raising to the power of a fracture it can be expressed as a percentage. In short, instead of doing:

(inflation) = 1,1 ^ (amount of passed years)

do:
(inflation) = 1,1 ^ (((current game year counted from the start year) / (amount of game years from start year to 2050)) * 100)