The International Simutrans Forum

 

Author Topic: [Experimental release] Pak128.Britain-Ex 0.4  (Read 16222 times)

0 Members and 1 Guest are viewing this topic.

Offline jamespetts gb

  • Simutrans-Extended project coordinator
  • Moderator
  • *
  • Posts: 18745
  • Cake baker
    • Bridgewater-Brunel
  • Languages: EN
[Experimental release] Pak128.Britain-Ex 0.4
« on: December 19, 2009, 11:35:47 PM »
A new version of the preview and testing version of the Simutrans-Experimental version of Pak128.Britain is now available: download it here. This is a complete pakset and should be installed in its own directory. It is only compatible with Simutrans-Experimental (version 7.0 and higher) - the normal Pak128.Britain should be used with Simutrans-Standard.

This is a preliminary version meant for testing and evaluation, although it should balance better with Simutrans-Experimental than the standard Pak128.Britain. Not all of the Simutrans-Experimental features are implemented yet (although I am planning to implement them all in due course).

Here is a list of the features supported by this release of Pak128.Britain-Ex:

  • Comfort
  • Loading times
  • Catering
  • Upgrading (partial)
  • Way constraints
  • Reversing
  • Tilting trains
  • Overcrowding (only on a few vehicles so far - work in progress)
  • Industry obsolescence
  • Industry upgrading
  • Weight limits (NEW)
  • Tractive effort (NEW)

See here for a more detailed description of the work outstanding and so far completed, and also for information on how to get involved in contributing to the project and obtaining the sources.

Changes from 0.3 include:

  • ADD: New ways and bridges, with weight limits (no graphics yet)
  • FIX: Powered vehicles with a zero weight travelled at infinite speed. All powered vehicles now have a weight of at least 1t.
  • FIX: Tram depots were unavailable between 1950 and 1957
  • FIX: Problems with the productivity of the sawmill
  • FIX: Problems with the output capacity of the orchard
  • FIX: The APT buffet car had the introduction date set incorrectly
  • FIX: The tender used for the 8P and 7MT had a speed limit too low for the 8P
  • FIX: Numerous steam locomotives had incorrect weights.
  • CODE: Synchronised the filenames for Mk. 1 buffets with the standard version
  • ADD: Mk3a vehicles (from the standard version - with adaptations to suit)
  • FIX: The LMS 4F had the introduction date set incorrectly.
  • FIX: Some wagons had an invalid weight of 0.5t (weights must be integers).
  • FIX: Cooled goods wagon now retires in 1903, not 1931 (from Standard).
  • CHANGE: Signals now available from 1750 (from Standard)
  • FIX: Incorrect introduction year for food carts and wagon
  • FIX: Sheep farm had zero capacity
  • CHANGE: Reduce the running cost for the BR 8P
  • CHANGE: The early GWR dining car now has a higher purchase and maintenance cost than ordinary carriages of the same set.
  • FIX: Various unsynchronised introduction dates of rail vehicles.
  • ADD: BR Class 423, "4-VEP" EMU
  • ADD: Bulleid/Raworth CC (later BR Class 70) 3rd rail electric locomotive
  • ADD: Rebuilt Merchant Navy class (can be upgraded from unrebuilt Merchant Navy)
  • ADD: Unrebuilt and rebuilt West Country class
  • CHANGE: LNER A1 and A3 are now different locomotives (using the same graphics) with different characteristics, and the A1 can be upgraded to the A3.
  • ADD: New road and rail bridges and ways, with various weight limits configured. Note that there are no new graphics for these as yet and they all re-use existing graphics.
  • CHANGE: Existing ways and bridges now have weight limits configured.
  • CHANGE: Increased passenger factor
  • CHANGE: Reduced congestion density factor (towns become congested more easily)
  • CHANGE: Increased city car lifespan (city cars remain in world for longer)
  • CHANGE: Corrected retirement dates for some suburban units
  • FIX: (From Standard) Corrected introduction years of some harbours
  • CHANGE: New graphics for some bridges
  • CHANGE: Can now upgrade from 2HAP to the GLV
  • CHANGE: Removed retirement date of higher density residential, commercial and industrial buildings that do not yet have a replacement
  • CHANGE: Adjusted retirement dates for some railway locomotives and carriages
  • ADD: Tractive effort values for most locomotives (it was not possible to find tractive effort figures for some locomotives and all multiple units)
  • CHANGE: BR Class 104 DMU now has the correct power rating
  • ADD: Trolley service versions of the BR Class 156 and 153, upgradeable and downgradeable
  • CHANGE: Class 170 Turbostar now has a trolley service in the middle vehicle
  • ADD: Compat.tab
  • CHANGE: Recalibrated distribution weights of industries to give a sharper industrial decline in later times.
  • CHANGE: Increased range of production of early industries to keep up with canals' transport capacity.

Any feedback on this pakset variant (especially relating to the gameplay balance) would be very much welcome.

Offline wlindley us

  • Devotee
  • *
  • Posts: 978
    • Hacking for fun and profit since 1977
  • Languages: EN, DE
Re: [Experimental release] Pak128.Britain-Ex 0.4
« Reply #1 on: December 20, 2009, 03:16:30 PM »
Postbox ("LoadingBay" is its name in the .dat file) still has a very late introduction, which makes early years gathering mail in cities difficult.  The standard version in bus-stops.dat says:

   Obj=building
   Name=LoadingBay
   # must be 1*1
   type=stop
   waytype=road
   intro_year=1750

Offline ӔO

  • Devotees (Inactive)
  • *
  • Posts: 2345
  • Hopefully helpful
  • Languages: en, jp
Re: [Experimental release] Pak128.Britain-Ex 0.4
« Reply #2 on: December 20, 2009, 05:36:56 PM »
2 bugs

- all tunnels, including road, rail and water, now have 0t max weight. (I have a few other paks loaded, so can someone confirm?)
- GWR 3100 2-6-2T prairie tank has 1903kW of power, which is the introduction date. It should be a lot lower.

Offline wlindley us

  • Devotee
  • *
  • Posts: 978
    • Hacking for fun and profit since 1977
  • Languages: EN, DE
Re: [Experimental release] Pak128.Britain-Ex 0.4
« Reply #3 on: December 20, 2009, 07:42:01 PM »
Wooden Road trestle is useless as it has 1t max weight, but 1 horse + 1 cart/carriage = 2t.

Offline ӔO

  • Devotees (Inactive)
  • *
  • Posts: 2345
  • Hopefully helpful
  • Languages: en, jp
Re: [Experimental release] Pak128.Britain-Ex 0.4
« Reply #4 on: December 21, 2009, 04:04:44 AM »
couple more bugs.
- There is something wrong with the "place signal", it acts like you are holding down the mouse button when selected, it will place down a signal on the first applicable way tile highlighted by the cursor and then places all signals down when you click down on a tile. It happens with basic, pre, long and choose signals for rail and traffic signals for road.
this is with 7.1exp, using 102.2.1 as the base.

- There is some strange graphical thing happening with underground. the grid lines are half a tile off, or the rendering for the grid is one level above.


Offline rsdworker

  • *
  • Posts: 281
Re: [Experimental release] Pak128.Britain-Ex 0.4
« Reply #5 on: December 25, 2009, 09:44:08 AM »
i discovered the bug - the tram tracks - one is slower - 20kph other one is 85kph - i tested them
the first one (slower tram track allows any trains go on that track its works on all years) but the faster tram track which probits normal trains using it but i found out thj in early years the trains was allowed go on tram tracks then later years the tram tracks are probited to allow trains on it but to my disovery - trams which go off the street - leaving the road so they won't go there - they display No route found
same happens on normal train track to tram track
i ran the tests and i found out the fast tram tracks are bugged because sometimes trams won't go on that fast tram track
now feedback
trains stock - the london underground rolling stock is missing - only one in list from 1907

Offline jamespetts gb

  • Simutrans-Extended project coordinator
  • Moderator
  • *
  • Posts: 18745
  • Cake baker
    • Bridgewater-Brunel
  • Languages: EN
Re: [Experimental release] Pak128.Britain-Ex 0.4
« Reply #6 on: December 25, 2009, 11:30:59 AM »
Thank you everyone for your feedback on this - it is much appreciated. Some replies:

Post boxes

Ideally, these should be introduced from 1842, which is when the penny post started. However, there is no pre-1842 type post office yet in the pakset, so I have, for the time being, set the postboxes to be introduced in 1750 as suggested. The change will take effect from the next release.

Tunnels

This, I have found, is due to a bug in makeobj-Experimental. This bug will be fixed in the next release of Simutrans-Experimental, and the tunnels will have the correct weight limit in the next release of Pak128.Britain-Ex.

Wooden road bridges

I have changed these to 2t rather than 1t from the next release. Please note, however, that it is the weight of the individual vehicles, not of the convoys, that counts for these purposes. However, the loaded weight of a 1t vehicle may exceed 1t, hence increasing the limit to 2t.

GWR large prairie

This appears to be a bug from Pak128.Britain-Standard; I have corrected it for Pak128.Britain-Ex, however, which fix will take effect from the next release. The power will be 1,300kW.

Signal placement

This is a bug in Simutrans-Experimental 7.1. I will look into fixing this by the next release. In the meantime, just place signals one by one in the same way as in Simutrans-Standard, and ignore the signal spacing dialogue box.

Grid lines

This is an issue from Pak128.Britain-Standard. If you use non-sliced underground view in Pak128 or PAk64, you will notice that, in recent versions of Simutrans, you will see grid lines representing the surface shape of the land in outline. In Pak128.Britain, however, those lines are black, which means that they do not show in underground mode, except over objects such as track. If you turn on the actual gridlines (key
  • ), you will see that the real gridlines are properly aligned.
Tram tracks

This is not a bug: it is intended. Normal tram tracks cannot accomodate normal railway trains: trains require signals and remotely controlled points, whereas trams drive on sight and control the direction that they take at points from the cab. Trams are also able to take far sharper corners than ordinary railway trains. There are, however, some places that have tramways designed to take ordinary trains, such as the Weymouth harbour line (which operated until 1987). However, trains could not run at more than a very slow speed on that section. See here for a discussion on the subject.

London Underground

This is incomplete because work has not yet finished on London Underground rolling stock. We hope eventually to have a full set. If anyone would like to contribute graphics in the style of Pak128.Britain for London Underground stock, we should all be extremely grateful.

Offline rsdworker

  • *
  • Posts: 281
Re: [Experimental release] Pak128.Britain-Ex 0.4
« Reply #7 on: December 26, 2009, 10:01:10 AM »
Quote
Tram tracks

This is not a bug: it is intended. Normal tram tracks cannot accomodate normal railway trains: trains require signals and remotely controlled points, whereas trams drive on sight and control the direction that they take at points from the cab. Trams are also able to take far sharper corners than ordinary railway trains. There are, however, some places that have tramways designed to take ordinary trains, such as the Weymouth harbour line (which operated until 1987). However, trains could not run at more than a very slow speed on that section. See here for a discussion on the subject.
yes i agree but there more on wiki which its possible to allow it - which its very common now - some has full trains running through city on road but UK don't have yet but there alot in USA - so maybe you should add train/tram track - its different from other tram track but it will have speeds set - 15mph or 30 mph

Offline Dutchman on Rails

  • *
  • Posts: 239
Re: [Experimental release] Pak128.Britain-Ex 0.4
« Reply #8 on: December 26, 2009, 11:27:50 AM »
Hi James,

I have two more, probably Pak.Britain-Ex specific, finding for you. Both of which have been produced with Simutrans 102.2.1 as base, Simutrans-Ex 7.1, and Pak128.Britain-Ex 0.4, but not with Pak.Britain 1.05.

1) The Livestock drover doesn't appear to function, when put on a valid route for livestock, no livestock is dropped onto the loading dock.

2) When being overdrawn, the interest costs are displayed in the finance window, but not deducted from the account balance.

A savegame (with some availailable routes that just the livestock drovers need to be added) has been uploaded:

http://simutrans-germany.com/files/upload/DoR%20Aberdeen%201781-02.sve

As usual, please don't hesitate to ask for further tests or information.

Offline jamespetts gb

  • Simutrans-Extended project coordinator
  • Moderator
  • *
  • Posts: 18745
  • Cake baker
    • Bridgewater-Brunel
  • Languages: EN
Re: [Experimental release] Pak128.Britain-Ex 0.4
« Reply #9 on: December 26, 2009, 11:36:11 AM »
RSD Worker,

I am having some trouble following what you mean; do you mean that the tramways that are suitable for full-sized trains have their speed limit set too low? I am not aware of any instance in the UK of a tramway for main line trains with a higher speed limit; do you know of any?

Dutchman on Rails,

thank you very much for the investigations and reports. I am currently staying away from home and cannot look into them now, but will look into them when I get back. Thank you again.

Offline Dutchman on Rails

  • *
  • Posts: 239
Re: [Experimental release] Pak128.Britain-Ex 0.4
« Reply #10 on: January 01, 2010, 02:14:59 PM »
First of all, a happy new year to all.

Following that, I have another three findings to report.

1) One additional industry chains appears in the first month after loading a savegame, and one more in the second, regardless of wether the threshold for a new chain has not yet been reached. This does not reproduce with Pak.Britain 1.06.

2) When loading a saved game which is in an overdrawn financial status, the credit limit rises spectacularly.

3) The credit limit also rises spectacularly in the following pattern (this takes some explaining): After recovering from being overdrawn, the credit limit starts to recover slowly. When, after a few months, investments cause me to be overdrawn again, this recovery stops, but the first month that the message: "Your credit rating is being affected" pops up, it suddenly rises spectacularly before slowly dropping in the following months.

Unfortunately, I do not yet have a stable enough test game for Pak.Britain 1.06 to test if this behaviour reproduces there. I'm killing some time with Pak.Britain 1.06 though while waiting for 0.5 of the experimental pak, and when I get to a stable testgame, I will give it a few tries.

The same savegame dropped earlier shows these findings too. As usual, please don't hesitate to ask any further questions.
« Last Edit: January 01, 2010, 02:31:30 PM by Dutchman on Rails »

Offline jamespetts gb

  • Simutrans-Extended project coordinator
  • Moderator
  • *
  • Posts: 18745
  • Cake baker
    • Bridgewater-Brunel
  • Languages: EN
Re: [Experimental release] Pak128.Britain-Ex 0.4
« Reply #11 on: January 01, 2010, 03:25:22 PM »
Dutchman,

thank you very much for your help. Issue (1) is caused by a change in industry weightings from the previous version of Pak128.Britain-Ex: it will not happen on games created with the latest version of Pak128.Britain-Ex.

Issues (2) and (3) I suspect are related, and would be a bug in the code, not an issue with the pakset (a pakset cannot cause this issue). The credit rating should indeed recover slowly after one gets out of debt, but should not rise spectacularly. I shall have to look into that when I get back home and am able to access my Simutrans-Experimental development setup. Thank you very much for the reports!

Offline Dutchman on Rails

  • *
  • Posts: 239
Re: [Experimental release] Pak128.Britain-Ex 0.4
« Reply #12 on: January 01, 2010, 03:49:26 PM »
Hi James,

Some additions and replies.

Regarding the issue of the additional industry chains, you said:

"Issue (1) is caused by a change in industry weightings from the previous version of Pak128.Britain-Ex: it will not happen on games created with the latest version of Pak128.Britain-Ex."

I'm sorry to say this, but the games I've reproduced this with were in fact created with the latest version of Pak.Britain-Ex known to me (version 0.4)...

Regarding my findings with being overdrawn, I could now reproduce my finding 2 in my post of 01/01/2010 03:14:59 in this thread, but also my finding #2 in my post of 26/12/2009 12:27:50 in this thread (interest not deducted), both with Simutrans 102.2.1, Experimental 7.1, and Pak.Britain 1.06.

Offline jamespetts gb

  • Simutrans-Extended project coordinator
  • Moderator
  • *
  • Posts: 18745
  • Cake baker
    • Bridgewater-Brunel
  • Languages: EN
Re: [Experimental release] Pak128.Britain-Ex 0.4
« Reply #13 on: January 01, 2010, 05:00:49 PM »
Dutchman,

thank you for your comments. In respect of (1); how did you check whether the threshold had not been passed? In Simutrans-Experimental 7.0 or later, it is based on industry density.

As to the other findings, I can't quite follow that last sentence - you could reproduce the issues or not? And how would you reproduce an issue relating to credit rating in Simutrans-Standard that has no concept of credit rating?

Offline Dutchman on Rails

  • *
  • Posts: 239
Re: [Experimental release] Pak128.Britain-Ex 0.4
« Reply #14 on: January 02, 2010, 11:38:25 AM »
Hi James,

Re: Industry density. This I must have missed. I assumed it was still governed by the cityrules.tab, which is set to 5000. I have to look into that more closely. Sorry for that.

Re: Credit: I do comparative tests between Simutrans 102.2.1 with Simutrans Experimental 7.1 and either Pak.Britain 1.06 or Pak.Britain-Ex 0.4. The credit issues appear to be happening in both situations. Sorry for the confusion.

Offline ӔO

  • Devotees (Inactive)
  • *
  • Posts: 2345
  • Hopefully helpful
  • Languages: en, jp
Re: [Experimental release] Pak128.Britain-Ex 0.4
« Reply #15 on: January 02, 2010, 03:22:37 PM »
not a bug, but thought I might mention it.

City list in experimental is much shorter than the one in standard.

Offline jamespetts gb

  • Simutrans-Extended project coordinator
  • Moderator
  • *
  • Posts: 18745
  • Cake baker
    • Bridgewater-Brunel
  • Languages: EN
Re: [Experimental release] Pak128.Britain-Ex 0.4
« Reply #16 on: January 02, 2010, 05:04:43 PM »
Dutchman,

from 7.0, the system for determining when new industries are built completely changed: it is governed by an industry density factor ratio against the overall population, not population growth thresholds in individual cities.

As to credit, I still don't understand what you mean - presumably, you can't test credit ratings in Simutrans-Standard? It doesn't have such a thing.

Offline Dutchman on Rails

  • *
  • Posts: 239
Re: [Experimental release] Pak128.Britain-Ex 0.4
« Reply #17 on: January 03, 2010, 12:09:34 PM »
Hi James,

Simutrans version: The installation I have is an installation of Simutrans standard 102.2.1 on empty hard disk space, with the executables and config files of Simutrans Experimental 7.1 installed over it, overwriting applicable files. Over that, I installed Pak.Britain 1.06 and Pak.Britain-Ex 0.4 in seperate directories, so I can do comparative tests.

Regarding industries, I am doing comparative tests to find out the behaviour of the industries using the Pak.Britain-Ex 0.4. I can now confirm an anomaly in saving and reloading a game, as I started two comparative games with just over 5000 inhabitants and one industry chain in January 1750 and let them run for some time.

The 'control', I did not touch, not save, not reload until 1775, by which times the population had grown from 5,348 to 6,673, but (as your explanation confirms), the industry chains had remained identical at one chain and 19 industries.

The test only ran until January 1st, 1754, by which time the population rose from 5,369 to 5,698. However, it was saved and reloaded on 1-4-1750, 1-7-1750, 1-1-1751, 1-10-1751, and 1-10-1752. In this game, industries had grown by January 1st, 1754 to 11 chains (with 101 industries)... I did not change anything to the version, configuration or settings.

If you'd like to have the savegames of the control and the actual test, please just give a small reply or PM, and I'll upload them.

I hope this helps. Sorry to keep bothering you over this.

Offline jamespetts gb

  • Simutrans-Extended project coordinator
  • Moderator
  • *
  • Posts: 18745
  • Cake baker
    • Bridgewater-Brunel
  • Languages: EN
Re: [Experimental release] Pak128.Britain-Ex 0.4
« Reply #18 on: January 03, 2010, 12:42:43 PM »
Ahh, I understand now! Thank you for the explanation, and apologies for the confusion. Yes, please do upload your saved games: that would be useful. I should be able to start looking into bugs in the code again from to-morrow or Tuesday. Thank you for your report!

Offline Dutchman on Rails

  • *
  • Posts: 239
Re: [Experimental release] Pak128.Britain-Ex 0.4
« Reply #19 on: January 03, 2010, 07:25:56 PM »
Hi James,

You're welcome. The two savegames have been uploaded.

http://simutrans-germany.com/files/upload/DoR for James Petts Control game 1775-01.sve

http://simutrans-germany.com/files/upload/Dor for James Petts Testgame 1754-01.sve

Incidentally, about the industry density, are there any plans for override switches in the simuconf.tab?

Offline jamespetts gb

  • Simutrans-Extended project coordinator
  • Moderator
  • *
  • Posts: 18745
  • Cake baker
    • Bridgewater-Brunel
  • Languages: EN
Re: [Experimental release] Pak128.Britain-Ex 0.4
« Reply #20 on: January 03, 2010, 08:29:50 PM »
Dutchman,

what do you mean by over-ride switches? What, precisely, would be over-ridden by what?

Offline Dutchman on Rails

  • *
  • Posts: 239
Re: [Experimental release] Pak128.Britain-Ex 0.4
« Reply #21 on: January 04, 2010, 07:47:26 AM »
Hi James,

Before answering, I have to report my finding of December 26th #1 (Livestock drover) as an incorrect report. There's nothing wrong there, I made a mistake. I'm very sorry for any inconvenience.

Regarding the industry settings, you explained to me that the industry density is determined by the number of chains existing at start and the starting population. The industry density and the weight factor in turn determine the number of industries.

What I mean by overriding is the following: Being able to set a different industry density in Simuconf.tab, which overrides the Starting Population/Industry Chains calculation. I'm just curious if you plan anything like that.

Offline jamespetts gb

  • Simutrans-Extended project coordinator
  • Moderator
  • *
  • Posts: 18745
  • Cake baker
    • Bridgewater-Brunel
  • Languages: EN
Re: [Experimental release] Pak128.Britain-Ex 0.4
« Reply #22 on: January 04, 2010, 10:25:29 AM »
Dutchman,

I'm not sure what the point of that would be, since both the starting population and initial number of industry chains are already customisable.

Offline Dutchman on Rails

  • *
  • Posts: 239
Re: [Experimental release] Pak128.Britain-Ex 0.4
« Reply #23 on: January 04, 2010, 04:55:10 PM »
Hi James,

What being able to set the density in simuconf.tab does is uncouple the industry growth from the starting position. So you can have a very weak industry starting position, but rapid growth. Or, more interesting for me (I like hard settings) a relatively favourable starting position, but slow growth. Or, better still, a weak starting position and even slower growth. For me, the latter might be especially important, as I generally don't have much population to play with (normally some 5000).

I hope this helps explaining where I'm going.

Offline jamespetts gb

  • Simutrans-Extended project coordinator
  • Moderator
  • *
  • Posts: 18745
  • Cake baker
    • Bridgewater-Brunel
  • Languages: EN
Re: [Experimental release] Pak128.Britain-Ex 0.4
« Reply #24 on: January 04, 2010, 07:25:14 PM »
I think that I see what you mean - but that strikes me as of somewhat marginal utility; bear in mind that pakset authors can, by specifying different industry weightings for industries of different eras, change the number of industries that appear as time passes in any event.

Offline Dutchman on Rails

  • *
  • Posts: 239
Re: [Experimental release] Pak128.Britain-Ex 0.4
« Reply #25 on: January 06, 2010, 08:13:19 AM »
Hi James,

True, authors of the PakSets can alter the weight settings of the industries, even relative to each other. However, am I correct in assuming that players downloading and playing a PakSet can't just edit the weight settings anymore without getting the source (if available), editing them, and using MakeObj to create their own variant PakSet?

However, I do agree that it may be of somewhat marginal utility, depending on how many players change the settings of their .tab files anyway.

Offline jamespetts gb

  • Simutrans-Extended project coordinator
  • Moderator
  • *
  • Posts: 18745
  • Cake baker
    • Bridgewater-Brunel
  • Languages: EN
Re: [Experimental release] Pak128.Britain-Ex 0.4
« Reply #26 on: January 06, 2010, 10:27:39 AM »
Another difficulty is this: the actual numerical value used for the target industry weight is an incomprehensible floating point number that would be extremely hard for players to specify manually. It is calculated by dividing the starting population by the sum of one divided by the distributionweight of each industry present at startup; imagine trying to work out what would be a meaningful value for that!

Offline Dutchman on Rails

  • *
  • Posts: 239
Re: [Experimental release] Pak128.Britain-Ex 0.4
« Reply #27 on: January 06, 2010, 01:48:31 PM »
So if I get you correctly, basically the formula is: G = P / (1/da + 1/db +1/dc + ...)

Where G = industry growth, P = population, and da, db and dc are each distribution weights of the industries. I agree that's difficult and a bit meaningless to analyse. But more importantly, P is the only number that can effectively be controlled, though some control is possible over the number of industries at the start (but to a far less extent the type).

What might be possible though is the following modified formula:

G = (P*(W/100)) / (1/da + 1/db +1/dc + ...)

Where W = a customisable number in the range of 1-9999 in simuconf.tab.

Now if W = 100, industry will grow the same as if it doesn't exist. However, if W = 50, industry will grow twice as fast as normal. If on the other hand W = 200, industry will grow twice as slowly as normal. This affects all industries equally.

Did I get this right? How do you think?

Offline jamespetts gb

  • Simutrans-Extended project coordinator
  • Moderator
  • *
  • Posts: 18745
  • Cake baker
    • Bridgewater-Brunel
  • Languages: EN
Re: [Experimental release] Pak128.Britain-Ex 0.4
« Reply #28 on: January 06, 2010, 02:00:44 PM »
That formula is close; however it is subtly but importantly wrong. What you call "G" does not represent industry growth but the density ratio. The game will try to maintain a consistent industry density ratio over time. Every month, it checks to see whether the current actual industry density is less than the reference (initial) industry density. If it is, it there is a chance that, in that new month, a new industry chain will be built. The chance is higher the higher the difference between the current and reference industry density.

Offline Dutchman on Rails

  • *
  • Posts: 239
Re: [Experimental release] Pak128.Britain-Ex 0.4
« Reply #29 on: January 08, 2010, 05:02:01 PM »
Hi James,

So if the Reference industry density number becomes artificially higher than the actual number, this should slow down the industry growth, right?

That means for my purposes that (W/100) should only be calculated when G is the reference density. If G is the actual density, it should not be calculated.

Did I get it right this time?

Offline jamespetts gb

  • Simutrans-Extended project coordinator
  • Moderator
  • *
  • Posts: 18745
  • Cake baker
    • Bridgewater-Brunel
  • Languages: EN
Re: [Experimental release] Pak128.Britain-Ex 0.4
« Reply #30 on: January 09, 2010, 11:16:02 AM »
Dutchman,

there will be no industry growth at all when the actual industry density is higher than the reference density. Only when the reference density is higher than the actual density is there any possibility of growth, and, the greater the difference, the greater the chance of a new industry chain in any given month. But do not confuse "G" (the industry density ratio) with the industry density itself. "G" will never be the actual density, since, in your formula, it is the density ratio. You have to multiply G by P to get the actual density (which makes me think that you probably have the sum in your calculation backwards, too).

Offline Dutchman on Rails

  • *
  • Posts: 239
Re: [Experimental release] Pak128.Britain-Ex 0.4
« Reply #31 on: January 10, 2010, 10:53:00 AM »
Hi James,

I see.

Well, in any case, I'm getting a bit worried that I might bore you and everyone else by continuing to press my case. So I'll leave it in peace from here. But thanks for your many patient explanations.

Offline ӔO

  • Devotees (Inactive)
  • *
  • Posts: 2345
  • Hopefully helpful
  • Languages: en, jp
Re: [Experimental release] Pak128.Britain-Ex 0.4
« Reply #32 on: January 20, 2010, 04:18:02 AM »
while playing, I've noticed something odd about the cross-connecting. Pretty sure I started the game with no cross connecting enabled, but they connect, and oddly at that.

somehow the game linked a cattle farm to a textile mill.

Offline jamespetts gb

  • Simutrans-Extended project coordinator
  • Moderator
  • *
  • Posts: 18745
  • Cake baker
    • Bridgewater-Brunel
  • Languages: EN
Re: [Experimental release] Pak128.Britain-Ex 0.4
« Reply #33 on: January 20, 2010, 11:02:58 AM »
Hmm, that's odd. What happens when you supply it with cattle?

Offline VS

  • Senior Plumber (Devotee)
  • Devotee
  • *
  • Posts: 4855
  • Vladimír Slávik
    • VS's Simutrans site
  • Languages: CS,EN
Re: [Experimental release] Pak128.Britain-Ex 0.4
« Reply #34 on: January 20, 2010, 01:25:08 PM »
There was a similar report for pak128 + standard a few days ago:
http://forum.simutrans.com/index.php?topic=4232.0

(Personally I don't see how a corrupted file would cause only mismatching goods types, so... it might be a real bug.)