News:

Simutrans Wiki Manual
The official on-line manual for Simutrans. Read and contribute.

Pak128-ilu

Started by Iluvalar, November 09, 2015, 03:58:45 AM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.

Iluvalar

Latest version (V11) : https://docs.google.com/uc?id=0B9FqMMo3L6W_Vk5FcGp2X1J3ZFk&export=download

Pak128-ilu is a complete rebalance of your favorite pak128 1.5.3 with a competitive, challenging and complex gameplay in mind using most of the mechanics in the game at their fullest extent. Sadly it came at the cost of some realism, but if you are here for a game this pak is for you !


Oh ! I want to make it clear here. I recently added assets from the pak128.Britain into the mix. I didn't really bothered up until now with copyright, but obviously I don't own any of the assets here.


Pak64 is now also in the party, but it make a graphical bug. I don't plan to fix it in near futur, it's up to you to chose if you want to use the vehicle.


Features

       
  • Full control of the speed bonus and maintenance cost to give players much more complex choices to make.
  • Players encouraged from clever gameplay to build one massive and extremely complex network instead of smaller isolated one.
  • Complex "optimal solution" dependent on your available money and very fragile to outside event. You might discover that one single new monument made you eventually rebuild half your entire network.
  • All vehicles and road unlocked at 1930. The are all viable for some situations. No obvious strong choice you need to pick all the time.
  • Auto-balance managed by a script with a very specific balance in mind. This again mean, all vehicles balanced correctly, no obvious outliers and a ton of choices !
How to start
It's crucial that you play with timeline around 1930 ! This only control the speed bonus which will make the game later much harder then at the start. If you feel like you're struggling, you can try up down to 1830 instead of beginner bonus.


My favorite setting right now is : 512x512 16 cities, 48 factories, 32 attractions with the starting 250'000$. I feel like it gives a large array of projects of different size to tackle, enough money to keep you busy the first year and yet a small enough area so in term, you'll start to maximize it and not just expand small projects.


To be successful, you need to understand the concept of a "line". Each factories produce 100 units of 100% of deliverable a month or the equivalent $ value of another product. It's controlled this way. If you build a straight 30km/h road between the producer and the consumer, you should make a profit of 5% of your investment (6$ per km). If you try to build a 50km/h road the same way, you might find out you do even less profit. This is because I designed the maintenance cost of the roads that way. To make good use of a 50km/h road, you need to merge 2 lines together. You should then retrieve roughly your 5%* per month at now 10$ per km. So you get more benefit from the same amount of product. The mechanics are reiterative and for every slice of 20km/h you need 175% more trips to make the optimal profit for the optimal initial cost.The railway is balanced using the same code, but expect 2 rails on any given speed. So you should get a decent result even at start with 1 single rail at 50km/h instead of the missing 2 at 30.

This will slowly push you into merging more and more traffic into the same enormous network and keep you busy upgrading again and again. In spiraling complexity. This might sounds much like your standard game, but thrust me, this is balanced by a script. It does that to a whole new level. Once you grasp the basics, making profit shouldn't be hard at all. The challenge reside into reinvesting your profits and making the most out of it.


Be careful about the assets of some vehicle which could be high enough to reduce the operation cost. Don't be fooled by this, it's also taken into account. Also, be very careful about the cost of some late game items. You might bankrupt on the very first km of rail if you dont look at the price !


*There is a small reduction of the cost which make faster roads just a bit more rewarding.


In numbers :
30 km/h,  1 supplier, 6$ per km per supplier, 5% investement per month
50 km/h,  1.75 suppliers, 10$ per km per supplier, 6% investement per month
70 km/h,  3.06 suppliers, 14$ per km per supplier, 7.2% investement per month
90 km/h,  5.35 suppliers, 18$ per km per supplier,8.8% investement per month
110 km/h,  9.4 supplier, 22$ per km per supplier, 10% investement per month
130 km/h,  16.4 supplier, 26$ per km per supplier, 12% investement per month
380km/h 18k suppliers (good luck with that)


Alternatively 200 passengers count for 1 line so theoricaly  with 3.5m passengers a month, you could run the fastest train in this pak.


Beginers tips
An experimented player would immediately merge as much road as he could with his starting money, but it's still totally profitable to start small with straight 30km/h lines between everything. Slower roads are both easier and need less money. IMPORTANT: you need to build the roads first, and choose your vehicles after. There will always be a vehicle for the speed you need.


The "special" vehicles are more expensive. The all rounder vehicles of a given speed (medium capacity) will also happen to be the cheaper. In doubt, just seek for the cheaper vehicle.

Conclusion:
The major break to the pak progress is play testing. It involve playing a game just long enough to see a problem and then scrap it. It's quite annoying and last time it sent me into a 18 months break from simutrans. So please if any play tester is volunteer ? I'll take any comment gladly.


Did you managed to break the figures I just mentioned ? please report it. How did you enjoyed the pak ? Any suggestions ? comments ? What is the fastest road you managed to build and run correctly ?



Below, an image of a 70 road meeting a 30 road. I chose this one because the same material is transported in both using 2 different vehicles and road in a 90 degree curve. In pak128, you would have no reason to chose the kroytor gaz-4 over the rvg-white 1064 if you had both avalaible.

Iluvalar

Pak128-iluV5 is finaly out !


This exactly show what I meant yesterday in my conclusion : I start a game, get annoyed by an aspect of the pak and then code another 4 hours to get a better result.


This time I ran a new pass before the actual balance, built a statistical reference of the density of the vehicles given a specific freight and waytype and rebalanced the capacity of the vehicles to meet my requested density (1 vehicle per 20km). You should always find a vehicle with a decent pricetag now.


To eventual people who downloaded my V4, you can still play your game and have lots of fun, but you might notice that the price for the trains are a bit high compared to road. You're playtesting and experience is still valid to me, hopefully the pace of the changes should slow down now. Sorry about that.

Iluvalar

V6
Today I changed the price of the locomotives and tractors, they were too expensive.
I changed the prices and maintenance of tunnels and bridges. It is still possible to play solely underground if you so choose, but you'll be punished more now.


if you are playing the V5, you can still enjoy your game. It's safe to wait until the next version, just remember the locomotive eat the profit of the first wagon pretty much entirely. With trains of 3+ wagons, you'll still get enough profits to have a good game.

IgorEliezer

Hey please, add some screenshots in the 1st post so people can see the looks of the pakset before downloading. :)

gauthier

#4
If I understood well, it will look the same as pak128 2.5.3. Anyway, as soon as possible (not now and probably not soon) I will give it a try. Pak128 needs a complete rebalancing so maybe you are the one to fix that.

EDIT: There's a problem with your pak. It behaves like pre-half_height pak128 but graphics are from pak128 2.5.3 which features half heights :/

TurfIt

The config file included appears to be from an old version. Attached is a corrected one. Rivers had the wrong names, and tile height wasn't set resulting in graphic glitches.

I've not had a chance to really test gameplay yet; It appears some of my comments from before are still valid, but it'll take more time to see.

It's nice to see you continuing, there's definitely room for a pak that's balanced completely different to the norm - yielding a very different game.

Iluvalar

Omg you are right, I fixed the height and the river by myself while Turfit was posting, it will be in V7. I tought it was a graphical bug on my client, now I feel stupid.


@gauthier

From your perspective, the main 2 main changes I made was to

First make the base speed at 25km for all vehicle and boost the speed bonus to 100% for all goods. This is how I can motivate the player well enough to merge their roads together. While keeping the slower roads viable. So all speed and all vehicle get to be used.

Second, control aggressively the value of goods and the production of the factories so I know for sure how much money to expect from 1 straight line. Otherwise you balance on thin air...

Here is a preview of my variables I use if someone is curious.

$_REQUEST['goal']['air']['min']=80;
$_REQUEST['goal']['air']['max']=200;
$_REQUEST['goal']['water']['min']=20;
$_REQUEST['goal']['water']['max']=70;
$_REQUEST['goal']['default']['density']=20;
$_REQUEST['goal']['road']['density']=20;
$_REQUEST['goal']['air']['density']=120;
$setup['baseprod']=100; //units of deliverable per line per month
$setup['fullroadspeed']=30; //base speed for 1 single straight line
$setup['speedstep']=20; //amount of speed require to incease the amouunt of line by 1.75 fold
$setup['benef']=.1; //percent of benefit to be expected per month based on expense
$setup['profiteperkm']=5;
$setup['tractorfrac']=.2; //part of the cost induced on the locomotive and removed from trailers
$setup['maintroadfrac']=1; //fraction of total earning to pay for road per month
$setup['operationfrac']=1; //fraction of total earning to pay for running cost and assets devaluation



Iluvalar

Ok V7 is out. On top of fixing the half height, it may contain all road vehicle, trains and factories from pak128.Britain  8)


however, I'm affraid there is a large bug where some factories produce much more then what is supposed. I saw one that produce 9 lines on itself the balance script don't seem to find them to balance them. It's not game breaking, but those factories will make much more money then expected. I chose to release this now because it's getting late here and I wont have time to post the half height fix for the next 72 hours or so... I didn't had any time to play test this...



DrSuperGood

Quoteit may contain all road vehicle, trains and factories from pak128.Britain
I did not think pak128 and pak128.britain were compatible. Not physically but stylistically.

gauthier

Indeed pak128 and pak128 britain may not look well together, and pak128 alone is already quite big. I will test V7 but not soon, I'm busy with my homeworks unfortunately :(

Isaac Eiland-Hall

I suppose the stated reason for the pak is the gameplay, so perhaps style is one of the compromises :)

Iluvalar

I might revert this in the futur, I still didn't played my test game for V7. I will see if the style hurt my eyes so much. But I had a few good reasons to give it a shot.

1- I can. It's a great test for the strength of my formula to mash 2 paks or more together and see if they still play fair all together. I bet the only reason why no one really tried this is because of the balance problems it would create. It need a complete revision of the balance, and that's what I happen to have.
2- I failed this time, but the pak128 lack of cargo airplane and ships. I need 3 for each goods and each speed. I'd like to import them but the image format crash the compile for now.
3- From game play perspective, more varied factory chains is fun. right ?
4- Even for vehicles, now I seem to have several options for the 30km/h roads and that's great for gameplay.

The end result for now is not as bad as it sound stylistically. pak britain is more dark, colorful and small. I the garage it shows, but side by side on the road, it look already better. For now it's an experiment because I CAN and i'm open to opinions.

DrSuperGood

Quote
2- I failed this time, but the pak128 lack of cargo airplane and ships. I need 3 for each goods and each speed. I'd like to import them but the image format crash the compile for now.
There are not many airplanes for cargo in real life and generally the ones that do exist are for very specific cargo. There certainly is no "Bulk" airplane. Also airplanes are bad for Simutrans because it encourages people to place 2 airports and make money for nothing compared with someone who spent hours laying a double track.

Iluvalar

#13
V8 . Oh sweet ! just saw the 2 30km/h oil truck form britain \o/ not more stuck with the pesky 35km/h magirus ^^

V8 is out. I lied, I worked on it lol. Contain the factory fix. All britain factories seem good. Also, it may contain all the planes and ships from britain ^^ . this pak look fun :D


EDIT : There was a post here about the balance of the vehicles in the pak, in answer to DrSuperGood. But the moderation forced me to post only once. So I had to chose between the announcement of the release of this version which is mandatory to this topic, and the chat with DrSuperGood. Sorry. Also, obviously this is my last version unless someone answer here.


EDIT 2 : Today Nov 2015, I have a great screen shot of vehicles between 1 vehicles from pak128 and 3 from pak128.britain. Curiously it's one of the 3 from britain that look the most stylistically out of place. Too bad I can't...

Iluvalar

V10

After talking to jamespetts about diffculty I chose to reward the players who manage to merge the most line. The ROI now start at 5% and grow with higher speed instead of reducing.

I corrected the electricity productions. The producers which consume a good will produce much more and all the little solar and wind turbine produce less and appear much less often.

I precalculated the amount of good produced and consumed by the factories and ajusted the odd of appearance of the factories in a way every good should now show at the same rate.

The relation between the engine power and the price as been reviewed.

Finaly I made a good dictionnary that merged some goods together.

Known bug : the pak64 vehicles look small, I use the code name of the imported things from pak 64 and britain instead of the translated name. I need to fix that. Careful between the freshfish (special freight) and the fresh fish (perishable) goods.

DrSuperGood

#15
You should be able to fix the pak64 convoys by bulk scaling all their images by 2*2 times. If you use nearest pixel resampling they will not blur which may or may not not look better, but in any case they will look less sharp than the 128 assets.

gauthier

I downloaded V10, all vehicles appear in 1829 and retire in 2300.

I think that a good balancing must be function of the introduction date, so old vehicles become obsolete (not only because they are better ones, but also because they get less and less profitable, through the increasing of speed bonus). Anyway, can you at least switch dates back to normal ?

Iluvalar

Quote from: gauthier on December 27, 2015, 02:52:12 PM
I downloaded V10, all vehicles appear in 1829 and retire in 2300.

I think that a good balancing must be function of the introduction date, so old vehicles become obsolete (not only because they are better ones, but also because they get less and less profitable, through the increasing of speed bonus). Anyway, can you at least switch dates back to normal ?

I let you start with much more money early game for the fun factor, but if you try at lower money. You'll find yourself completely unable to build anything but 30km/h roads for a while. Maybe one well planned 50km/h. With the actual money, I managed a 100km/h little rail of about 30km long. And it was quite a stretch I needed 4 years to make it work fine. The pak still simulate the technological and speed progression, but on it own. You can't build an highway right away in 1930. Unless you add 1 or 2 zero to your account number...

Faster vehicles DO gives better profits and DO allow to race the speed bonus reduction throughout the game. If you go up to 1970, you'll find out that dirt roads and 50km/h roads are pretty much obsolete. Because of the game mechanic itself. If you play a game you WILL replace your fleet of vehicles periodicaly.

In other words, the game provide you with an array of speed choices all valid with increasing complexity. This is exactly the goal of the balance. Doesn't mean you can afford the highest level of complexity with your starting money.

With the current money, you seem to have the technology level of 1965-1975. It's not a priority for me to fix that doe. You could, start with less money or make the clock fit 1965 (but that would make the game harder) or just pretend that the calendar of the game is 35 years off ?

----
I do have the design space to allow for introduction dates, and makes some vehicles a tad better then the competitors as you propose, but I have several reasons not to. First as I said, the player already upgrade his fleet from time to time, forcing him to do it at a given date doesnt seem to fullfill any real purpose in term of game experience. Would be just an extra chore and the game is already stimulating enough now. More importantly, I felt the need to throw the pak64 in already, because there were still holes in the speed/freight/capacity matrix. It's awesome to swim in different choice of passenger wagons... trying to find THE gem which will suit your needs knowing they are all perfectly balanced and viable. Seeing that, from game to game, your choice don't land on the same asset.

If someone have something like 10k more vehicles to offer to me, I'll gladly balance them all and reintroduce the timeline. But as it is, it's much more fun to let the player simulate his own timeline and showcase the fact that even with all the vehicles avalaible, the choice is still hard to make.


gauthier

I understand your point of view but it's hard to find something when your depot window is completely saturated with vehicles from the whole timeline. I agree with you about the need for more choice in vehicles and strategy, which current pak128 has not (that's why I always play with addons). I personnally can't spend a hour everytime I need a convoy to find out what vehicles are okay to use or not.

As I said, I'm interested in your work in order to improve balancing of pak128, could you explain some of the maths in your balancing process ?

Iluvalar

#19
Quote from: gauthier on January 02, 2016, 12:59:54 PM
I personnally can't spend a hour everytime I need a convoy to find out what vehicles are okay to use or not.
First, you build your network. You try to make the fastest road possible by funneling the most traffic on the same road.

So, when you come to chose a vehicle, you know the speed of the road it will cross. There is no vehicles with the right speed that are not "okay" at this point, you know that if you make that traffic happens, you'll make 10% of the price of the road per month in profits. If you don't want to bother in your first play, just pick the vehicle with the middle capacity and middle power. That one is more or less guaranteed to make you 5% of his cost per month in profits.

You don't need to bother and make math to find the vehicle that is "okay".

The only time this is kinda bothering, it's the first time you go through the passenger cars. They are ordered in capacity instead of speed. But that's a gui issue more then a problem I can solve. Once you browsed a little and found the right passengers wagons for your speed, it shouldn't be that hard to find the one with the middle capacity if you don't have any extra constraints. You know that wagons will worth it as long as if full enough.

Maybe you are having a problem because you try to find the vehicle before having the roads ?

------------------------
For the balance, there is 2 key ingredients you need to change and I can't find any other way around :

1- You need an high speed bonus. This allows for different speed to be available for different traffic. If you don't do that, and all vehicles are perfectly balanced, the optimal speed in a straight line will dominate any other network design and the game will be boring. In real life, the traffic increase much faster on faster lines so you don't need that high speed bonus. But that would imply a factory on both side of the road in pretty much the most crowded way possible (like in real life). I don't really think i want a game that make me micromanage 3k factories in a real life sized city. Actually, yes ! it would be fun, but it would require a rewrite of the engine to make the construction of the factories along the already existing road and pretty much a whole new game...

2- You need to control the amount of profits that the factories output and input. It's not technically mandatory, but you will confuse the player to no end if you force him to calculate the profits of every lines manually. They'll have to be familiar with every single factory chain you create. Also if you don't the factories won't be balanced themself, some factory chain will be much more profitable and players will have gravitate around them and forgive about the smaller chains. Which is sad in my eyes in term of game diversity. Here in this mod, I went big it's 5$/km/month for every line of goods transported also I enforced the 100 boxes per month rule which make the traffic calculation much much easier for the players. I know it's not very realistic this way, I could compromise a bit between realism and ease of use for the player, but the main key here is to control the profit per line via the cost of the goods and the production of the factories.


Once there 2 things are controlled, you know how much profit and how much traffic each speed of road can produce. You know the player will ugrade his road for the extra speed bonus if the traffic go higher, so you can balance the price and maintenance of the road correctly over this information. You have all the information, I bet the actual maths are not your problem. The last thing you need to chose is this :

$setup['profiteperkm']=5;
$setup['benef']=.1; //percent of benefit to be expected per month based on expense
$setup['maintroadfrac']=1; //fraction of total earning to pay for road per month
$setup['operationfrac']=1; //fraction of total earning to pay for running cost and assets devaluation

the fraction of maintenance will affect how much a road can be curvy and still do benefits. At first, an high fraction make it look like the game would be more challenging, but it also force the player to build straighter way which at the end will reduce the possibility he have to merge the road together into an higher traffic road. You end up reducing the amount of options he have to chose from, so it make the game easier, oddly.

the operationfrac vs maintroadfrac will affect how much a vehicles would prefer reusing an existing road even if it's a bit of a detour instead of building a new straight one. Again, it favor complexity if you let them reuse the roads, but if you reduce the operation cost too much, they can do anything and it become too easy.
-----
Finaly... balancing the capacity, the power and the weight. If you blindly balance the capacity vs profit vs cost, all vehicles for the same speed would worth the exact same for the player and it would be boring. Instead I made the price go in a V shape, were the cheapest vehicle is the one with the middle capacity. BUT I also calculated precisely the devaluation of the asset of the vehicles, I intentionally over appreciate it by 30%, and reduce the operation cost of those vehicles accordingly. Right of the bat, those vehicles are nerfed, because you then do less then the optimal 10% benefit per month of the cheapest vehicle. But those DO more money at the end per trip because I over appreciate the asset depreciation. Also, as they age, very expensive vehicles tend to make more and more profit since their assets drop. So there is no real "best" option regarding the capacity of the vehicles.

gauthier

In fact the problem I'm experiencing is more an inconvenience than a problem. I know how to choose what vehicle to use, but they are so many available at the same time that, even if they are sorted out by speed or capacity, I have to "hide" from my sight (mental process requiring already a lot of brain computing power because of the amount of objects to choose from) vehicles that are too old (obsolete and not profitable) or too modern (too expensive at that time). Then, being still "hiding" these vehicles, I have to compare those that are not hidden to choose the best one.

What I want to explain is that, unless you know nearly perfectly your pakset (it's your case since you developed it, but clearly not mine or that of any new player of your pakset) it's a hassle choose something. I trust you when you say that it's a great thing to have a wide array of possible strategies, but it makes it too uneasy for players until they get used to it, and that can take long in this case.

----------------------

1- I disagree with the need for high speedbonus. It's precisely how current pak128 is done and the main consequence is that the only way to make a network profitable is having main lines at the highest speed possible so as to make money for all the local lines that are hardly (or even not at all, especially subways and buses in modern eras) profitable.
If you keep high speed bonus but you balance so that slow vehicles are still profitable, you end up with high speed vehicles being completely cheated.

Regarding this, I can't understand how you came to the conclusion that high speed bonus could allow for different available speed. Higher the speed bonus, higher the advantage of using high speeds rather than slow speeds. It's as simple as that.

2-
Quotethe 100 boxes per month rule
What is this rule ?
QuoteOnce there 2 things are controlled, you know how much profit and how much traffic each speed of road can produce. You know the player will ugrade his road for the extra speed bonus if the traffic go higher, so you can balance the price and maintenance of the road correctly over this information.
I understand and fully agree with this.

3- I have a question: how speed bonus increase over the time ? This is what describes how vehicles become obsolete.

QuoteIf you blindly balance the capacity vs profit vs cost, all vehicles for the same speed would worth the exact same for the player and it would be boring.
Interesting but how do you define your middle capacity ? If you define it by the average (or any function depending on) of the vehicles you have, it won't be the same if you add or remove vehicles from the pakset. End it's even more irrelevent regarding possible addons. Nevertheless, in a regular pakset (not in yours obviously :p ) vehicles are already nicely filtered when you took into account waytype, introduction and retirement dates and speed so you don't need to make a difference between capacities cause you only get one or two vehicles in the end.
However, doing this, you advantage one vehicles above the others, isn't that a lack of choice leading to a boring game ?

Quotethe devaluation of the asset of the vehicles
Could you define it please ?

QuoteI bet the actual maths are not your problem.
Yes they are ;) . I suppose you got some formulas or at least basic arithmetic operations to balance everything. In fact, what I'm curious about is knowing the maths (what function, what operations, ...) you decided to use according to the choices you just explained. Don't worry about exposing eventually complicated things, I'm familiar with maths ;)

Iluvalar

Quote from: gauthier on January 03, 2016, 10:59:54 AM
What I want to explain is that, unless you know nearly perfectly your pakset (it's your case since you developed it, but clearly not mine or that of any new player of your pakset) it's a hassle choose something.
No, even I have no clue. I'm double checking my maths on 3 bulk trucks, the pegasus and a boat. Since I change high level variable, like profite per km or mean density of vehicles, I have absolutely no idea how all the other vehicles really are. All I know to help me is this :
Quote
30 km/h,  1 supplier, 6$ per km per supplier, 5% investement per month
50 km/h,  1.75 suppliers, 10$ per km per supplier, 6% investement per month
70 km/h,  3.06 suppliers, 14$ per km per supplier, 7.2% investement per month
...
Alternatively 200 passengers count for 1 line
Once you accept that as a fact, and wrap your head around the consequences of it (and this might take 2-3 games indeed), choosing the right vehicles is not a shore.


Quote from: gauthier on January 03, 2016, 10:59:54 AM
1- I disagree with the need for high speedbonus. It's precisely how current pak128 is done and the main consequence is that the only way to make a network profitable is having main lines at the highest speed possible so as to make money for all the local lines that are hardly (or even not at all, especially subways and buses in modern eras) profitable.
If you keep high speed bonus but you balance so that slow vehicles are still profitable, you end up with high speed vehicles being completely cheated.


Regarding this, I can't understand how you came to the conclusion that high speed bonus could allow for different available speed. Higher the speed bonus, higher the advantage of using high speeds rather than slow speeds. It's as simple as that.
No because I can match the maintenance of the road with the speedbonus and that maintenance also expect more traffic. If your network don't have the complexity to generate enough traffic, you can't build the higher speed. If the speedbonus was 0, the high speed bus could drive on a dirt road with no consequences. You'd expect the same profit from both and therefore, the operation cost per unit of freight would be the same. There would be also very few reasons to try to merge several road together instead of just drawing the straight line all the time between 2 points.


Quote
2- What is this rule ?
I fixed the production of deliverables type good (paper,ink,electronics) to 100% of 100 units per month on all factories. This allows the player to access the amount of traffic he generate as easily as it can be.


Quote
3- I have a question: how speed bonus increase over the time ? This is what describes how vehicles become obsolete.
It's the other way around, the base speed for bonus calculation is 20 at 1830 and 30 at 2030. My script make the balance assuming the base speed is 25 at 1930. This makes the dirt roads (straight lines) slowly become less and less profitable. It hurt everything else of course, but the dirt roads suffer the most. So you want to race the clock by complexifying you network more and more to stay profitable.


Quote
Interesting but how do you define your middle capacity ? If you define it by the average (or any function depending on) of the vehicles you have, it won't be the same if you add or remove vehicles from the pakset.
$_REQUEST['goal']['default']['density']=20;


Sorry i missed that part for you ;) . That's 1 vehicle every 20 avalaible spot (accounting for the traffic). It seem low, but in practice it's halfway to a huge traffic jam. To make sure I calculated the mean capacity of a given speed and freight and enforced it to that value. So when I say "the middle capacity" I really mean it ;) . You don't need to go to that extreme if you control your stats manually of course.


Quote
However, doing this, you advantage one vehicles above the others, isn't that a lack of choice leading to a boring game ?
Yes and no. It's true that I favor middle ground vehicles to extreme vehicles. But it's kind of an intention in the design, when someone create a vehicle with 5x the normal capacity, he don't expect it to be used as a default choice. On the other end, vehicles which were designed to have close to the middle capacity have their price tags close together and make an interesting choice. Also, at the end of the day, the more expensive vehicles are cheaper to operate and WILL produce a tad more benefits. I'd generally recommend to expand instead of investing into expensive vehicles, but you see that the solution is not that simple


Quote
Could you define it please ?
I think it's hardcoded at .3%, you'll see it in the formula below for the max price possible.



Here is the core of the script for vehicles. Hope it make sens.



         $speedboost=1+($obj['speed']-$basespeed)/$basespeed;
         $traffic=($obj['speed']-$fullroadspeed)/$speedstep;
         $dist=$obj['speed']*$hourpermonth;
         $nline=pow(1.75,$traffic);

         $gainPKm=$payload*$good['value']/$goodprop*$profitperkm/$goodfrac*2.5/10;


         $gainPKm*=$speedboost;

         $tg=$gainPKm*$dist; //total gain
         
         $pmin=$tg/100*(2-(1-$noroad)*$interestway)*pow($powerFrac,1)/$vprice;
         $pmax=$tg/100*($operationfrac+$noroad*$maintroadfrac)/$goodfrac/1.3/.003/(1-$vprice);

         $pf=$pmax-$pmin;
         $capfrac=$densityperkm*$nline*$gden;
         if($capfrac>1){
            $capfrac=1/$capfrac;
         }
         $ffrac=(1-$capfrac);
         $fprice=$pf*$ffrac*$powerFrac;   
         $fprice=min($pf,$fprice);     
         $price=$pmin+$fprice;

         $assetcost=$price*(1-$vprice)*.003; //depreciation on your net worth each month
         $assetPKm=$assetcost/$dist;
           $rc=($gainPKm*($operationfrac+$noroad*$maintroadfrac)/$goodfrac-$assetPKm*1.3*100)/$buffspeed; //final running cost
         

gauthier

Hi, sorry for the delay since my last answer, I have been busy.

Quote[...] Once you accept that as a fact, and wrap your head around the consequences of it (and this might take 2-3 games indeed), choosing the right vehicles is not a shore.
I'm not sure we really understood each other here.
You are the average player, you don't know by heart each vehicle in the pakset. You open your depot window and get you face slapped by a completely saturated window. I have no doubt that after some games, one gets used to some vehicles, but it's still a big effort. A compromise should be found between realism and choice on one side and playability and delight on the other side.

QuoteNo because I can match the maintenance of the road with the speedbonus and that maintenance also expect more traffic. If your network don't have the complexity to generate enough traffic, you can't build the higher speed. If the speedbonus was 0, the high speed bus could drive on a dirt road with no consequences. You'd expect the same profit from both and therefore, the operation cost per unit of freight would be the same. There would be also very few reasons to try to merge several road together instead of just drawing the straight line all the time between 2 points.
I have to admit that I did not think much about changing ways'maintenance. I don't speak about shuting speedbonus down to zero. Of course the game would be completely silly. But I though about making it far less important than it is now in regular pak128 for the reasons I explained earlier.

QuoteI fixed the production of deliverables type good (paper,ink,electronics) to 100% of 100 units per month on all factories. This allows the player to access the amount of traffic he generate as easily as it can be.
Makes sense.

QuoteIt's the other way around, the base speed for bonus calculation is 20 at 1830 and 30 at 2030. My script make the balance assuming the base speed is 25 at 1930. This makes the dirt roads (straight lines) slowly become less and less profitable. It hurt everything else of course, but the dirt roads suffer the most. So you want to race the clock by complexifying you network more and more to stay profitable.
Did you come to this results after several retries or is there a bit of maths (or even approximative mathematical thinking) behind it ?

If I ask you how long a vehicle can make profit after its introduction date, can you answer ?

QuoteSorry i missed that part for you ;) . That's 1 vehicle every 20 avalaible spot (accounting for the traffic). It seem low, but in practice it's halfway to a huge traffic jam. To make sure I calculated the mean capacity of a given speed and freight and enforced it to that value. So when I say "the middle capacity" I really mean it ;) . You don't need to go to that extreme if you control your stats manually of course.
I'm not sure to have fully understood that. You took a given vehicle, put it once in every 20 tiles, made them run (taking speed, power and weight into account) and just took as a result how much they are able to transport in a given period of time, didn't you ?

I have not read your script yet, I will do it later.

Anyway, thanks for these explanations.

Iluvalar

Quote from: gauthier on January 17, 2016, 03:16:45 PM
Hi, sorry for the delay since my last answer, I have been busy.
It's fine i'm not in a hurry :) . As long as you dont forget me.

Quote from: gauthier on January 17, 2016, 03:16:45 PM
I'm not sure we really understood each other here.
You are the average player, you don't know by heart each vehicle in the pakset. You open your depot window and get you face slapped by a completely saturated window. I have no doubt that after some games, one gets used to some vehicles, but it's still a big effort. A compromise should be found between realism and choice on one side and playability and delight on the other side.
Yes he will face slap. But hopefully, he will go back to my original post and read the "how to start" and he will understand that the speed he need depends on the traffics on the road and that he's better to start at low speed straight lines if he's in doubt. Once you know the desired speed, searching for a cheap vehicles is not that much of a shore.

Quote from: gauthier on January 17, 2016, 03:16:45 PM
I have to admit that I did not think much about changing ways'maintenance. I don't speak about shuting speedbonus down to zero. Of course the game would be completely silly. But I though about making it far less important than it is now in regular pak128 for the reasons I explained earlier.
Well there would be other ways around to balance the speed of the vehicles and the speed of the road into something related to the network. But not without changing the game engine. At this point the best is to give it a shot for yourself. Challenge : try to get a profitable 70km/h road before 1931 ? You will see the dynamic caused by the speedbonus and how you need both 30km and 50km roads just to achieve it. In the same network at the same time.


Quote from: gauthier on January 17, 2016, 03:16:45 PM
Did you come to this results after several retries or is there a bit of maths (or even approximative mathematical thinking) behind it ?

If I ask you how long a vehicle can make profit after its introduction date, can you answer ?
I calculate the profit made in the script assuming the base speed is at 25km/h and then give the player an extra 5% of his investment in profits. Obviously when the speed bonus goes down from an increasing base speed, the profit of the player plumet. It's possible to race it by increasing the traffic on the roads, by complexifying the network, etc. but by definition, it gets harder. I could have used an introduction date and calculate for that vehicle at the right base speed, now that I think of it.

A vehicles can always make profits. However, it's already outclassed by faster vehicles. The player is always seeking to replace his vehicles by faster/better one. From game mechanics, not fixed date. But for that, he need a bigger, more complex and more expensive network to support the higher traffic.



Quote from: gauthier on January 17, 2016, 03:16:45 PM
I'm not sure to have fully understood that. You took a given vehicle, put it once in every 20 tiles, made them run (taking speed, power and weight into account) and just took as a result how much they are able to transport in a given period of time, didn't you ?
No, It's calculated directly by the script, but the end result is there.

gauthier

QuoteA vehicles can always make profits. However, it's already outclassed by faster vehicles. The player is always seeking to replace his vehicles by faster/better one. From game mechanics, not fixed date. But for that, he need a bigger, more complex and more expensive network to support the higher traffic.
Don't you think it would be better to have old vehicles unprofitable in modern ages than having just them less profitable than modern vehicles ? The point is to kinda force the player to replace too old trains. Of course not every ten or fifteen years like what current pak128 was balanced for but more realistic periods. For example a train usually stay in service for about 50 years in real life, that could be, in the game, the limit after which the train goes unprofitable.

QuoteNo, It's calculated directly by the script, but the end result is there.
Yes, obviously it's calculated by the script. By saying "you did this, you did that" I actually meant that you coded it that way in the script or not ?

QuoteYes he will face slap. But hopefully, he will go back to my original post and read the "how to start" and he will understand that the speed he need depends on the traffics on the road and that he's better to start at low speed straight lines if he's in doubt. Once you know the desired speed, searching for a cheap vehicles is not that much of a shore.
Yes, I got that :) What I mean is that the average player would still see this as a hassle, and 95% of the players won't even try getting to your message. Well, it's fine in your pakset, but if your script is to be used on pak128, introduction dates for vehicles will stay in place.

Anyway, I think you gave me enough explanations for now. I have to put time into testing your pak in depth and understanding your script. I can't tell you when I will be able to do this as I already strive to work on the next release of pak128.

Vladki

Hello Iluvalar. How much do you care about realism in your pak? If you do, then timeline is a must. If you ignore inflation, you'll find out that modern vehicles are much better choice than old models - better technology means more efficiency, higher speed,reduced costs and maintenance.

If you take a small railway startup company, if it had the choice of any vehicles built in last 200 years, all well supported by spare parts etc, they would start with some  modern diesel engines or dmu. There's no place for steam engines, if you have diesel and electric ones. (unless you just love steam and are willing to spend extra for that).

But if you balance without regard to real life vehicle specifications, then it is your choice.

Iluvalar

I'll answer both of you at once.

Gauthier, the beginners can always resort on straight lines using 30km/h roads and setup their basic operations that way. It will not be the most thrilling game ever, but it will be a start. You also really underestimate the hassle for a new player to jump in a hand made balance pak, some vehicles are better then other you need to run the maths for every single vehicles yourself to figure where they perform the best, they have different speed so the combination of the best combo at a given speed is very very hard to figure out. And worst, that optimal speed is ever changing due to the timeline. It takes years of gameplay in a pak like that to really master it. These problems doesn't exist in my pak.

Obsolescence... There is no definitive "best" vehicles in the pak, the faster vehicles are making more profits, but they also require a much more complex network to be effective. So the player really have to change his network often, and change the speed of the sections. He's constantly upgrading the network. Not only the vehicles but also the road. This part of the game is definitively covered by the gameplay I created.

I could balance some vehicles to be worst then other and give them in a timeline, but for the purpose of this proof of concept, I'd rather have 3 vehicles for any way type at any speed possible. And demonstrate this way that they are all balanced and offer the player good choices.

@Vladiki, I'm aiming for gameplay over realism. At the start of the game, faster engines are too expensive. I don't make any distinction between a 1930 or 2015 vehicle at 70km/h however, but that's for the reason just above.

Iluvalar


So many things happening in the same image, I had to share :)

3 different road speed : 30,50,70 km/h
3 of them transfering bulk at once.
yet 5 different bulk truck the shorty volvo also having configurations with and without trailers... so 6 vehicles on 3 lines at once...
90km/h trucks on 70km/h
70km/h trucks on 50 and even 30 km/h
50km/h buses on 30km/h roads

You'd think I'm mad if I was playing any other pak, but no.. I'm successful. All of this is viable.

I believe this image is an awesome showcase of the complexity you can achieve in my balance. Don't you guys think ?

---
Side note, I've been browsing for a good 70km/h bus. Oh god so many choices. So many bus balanced yet deliciously unique. Which one shall I pick ? Gauthier will hate me but I love that so much.