The International Simutrans Forum

 

Author Topic: Wainwright  (Read 1803 times)

0 Members and 1 Guest are viewing this topic.

Offline wlindley us

  • Devotee
  • *
  • Posts: 960
    • Hacking for fun and profit since 1977
  • Languages: EN, DE
Wainwright
« on: October 21, 2017, 03:38:44 PM »
Enclosed, proposed Wainwright industry, in three versions (1750-1789, 1790-1849, 1850-1909), with snow images.  Also an updated auto-industry Gimp bitmap layer file from which this is created.  Zipfile with sources and playable pak here.


Now you know where all those carriages (citycars and stagecoaches) are built!  In 1910, the proprietors naturally switch to those newfangled horseless carriages.








Offline jamespetts gb

  • Simutrans-Extended project coordinator
  • Moderator
  • *
  • Posts: 18502
  • Cake baker
    • Bridgewater-Brunel
  • Languages: EN
Re: Wainwright
« Reply #1 on: October 21, 2017, 09:09:37 PM »
Splendid, thank you for that - it is excellent to see you back in production of lovely new things for the pakset!

One thing, however: I have been working lately on implementing the passenger and mail classes feature, and also the logistics feature. This involves adding a number of new parameters to industries (some of them have been available since the passenger generation work a few years ago but had not hitherto been used in the pakset).

These parameters are:

population_and_visitor_demand_capacity
employment_capacity
mail_demand

class_proportion[0]
class_proportion[1]
class_proportion[2]
class_proportion[3]
class_proportion[4]
class_proportion_jobs[0]
class_proportion_jobs[1]
class_proportion_jobs[2]
class_proportion_jobs[3]
class_proportion_jobs[4]

Adding these to all the industries retrospectively is a large amount of work, so it would be exceedingly helpful if you could add these parameters to new industries that you create.

Many of the industries on the passenger-and-mail-classes branch of Github have these parameters set (and those that do not currently have them set should have them set shortly). Here is an example of an industry that does have the parameters set, as well as some notes to explain how to calibrate a number of them (especially employment_capacity).

The upper group above are all things that will work and be recognised in the current master branch, whereas the lower group will only work in the passenger-and-mail-classes branch (but the .dat files with these things set will still compile in the master branch).

The population_and_visitor_demand_capacity sets either the population if the building is residential or the visitor demand if it is non-residential. Visitors are any arriving passengers who do not work there: those who work there are commuting passengers. This parameter works in the master branch. However, in the passenger-and-mail-classes branch, this has added significance for consumer industries. Consumer industries will now only deplete their stock in proportion to the number of visiting passengers that actually arrive. For example, a consumer industry that now depletes its stock at a fixed rate of 100/game month (i.e. 6 hours and 24 minutes on the short time scale) in the master branch and that has a visitor demand (after adjustment - the figure that you input will not be the same as the actual value in the game because of automatic adjustments) of 10 would now deplete its stock by 10 for every arriving visiting passenger (so that the stock depletion rate would be the same as in the master branch if the actual visitors match the visitor demand).

The employment_capacity sets the number of people employed by the industry (again, this figure is subject to adjustment: see the notes in the comments in the file here for a worked example). This has effect in the master branch, but, as with visitor demand, is more important in the passenger-and-mail-classes branch. In that branch, an industry will only attract commuting passengers if it is producing. An industry that is not producing will lay off its workforce, and commuting passengers will not start their journey to that industry: if they have reachable alternative destinations, they will travel to one of them, or else they will register as a failed journey.

Similarly, an industry with less than a certain complement of staff (I think 80%) will produce at a reduced rate owing to a staff shortage, that rate reduction being proportionate to the degree of shortage. A consumer industry with a serious staff shortage will close, and customers (i.e., visiting passengers) will no longer be able to start their journey to the consumer industry, which will prevent it from depleting its stock.

The mail_demand parameter sets the building's mail demand (again, this is adjusted) and works in the current master branch. There is no current direct link between an industry's productivity and its mail fulfilment, but this parameter is likely to be more important when the town growth algorithm is changed. It is in any event important properly to calibrate mail production.

The class_proportion and class_proportion_jobs figures (one for each of the five classes) work only in the passenger-and-mail-classes branch (but are simply ignored by the master branch makeobj, so can safely be added to industries intended for use in the master branch for the present). They set, respectively, the proportion of (1) population (for residential buildings) or visitor demand (for non-residential buildings); and (2) workers that belong to each class. For example, an office building might have quite a high proportion of higher classes, whereas a factory might have a higher proportion of lower classes. The classes are not intended to be quintiles, so there will not be an even number of people in each of the five classes (and do bear this in mind if producing residential buildings, to which these parameters also apply).

All of these parameters apply equally to town buildings that are not industries as they do to industries.

As should be apparent, these new parameters will be very important for the game to balance correctly once the new features are fully implemented (and this should be fairly close - the pakset work is the main thing delaying merging this with the master branch). You should try to use historically accurate figures in so far as possible, and extrapolate from known data for similar industries (including using the data that I have added for other industries and town buildings) where data cannot be found.

Thank you again for your splendid work - it is much appreciated.

Offline jamespetts gb

  • Simutrans-Extended project coordinator
  • Moderator
  • *
  • Posts: 18502
  • Cake baker
    • Bridgewater-Brunel
  • Languages: EN
Re: Wainwright
« Reply #2 on: November 12, 2017, 01:11:36 AM »
I should note that I have now added the additional parameters and incorporated this into the passenger-and-mail-classes branch. Thank you for your work on this.