The International Simutrans Forum

Simutrans Extended => Simutrans-Extended development => Topic started by: ӔO on February 13, 2013, 10:35:36 PM

Title: [10.26] Runway length not calculated correctly
Post by: ӔO on February 13, 2013, 10:35:36 PM
It seems that runway length is not calculated correctly in the game on both takeoff and landing.

For an airplane with 2500m minimum runway length and a game tile length of 250m, you would need 10 tiles minimum, but it seems that even if the runway is made to 3000m (12 tiles), 4000m (16 tiles) the airplane will always display a message that the runway is too short.

Opening and closing its schedule seems to allow it to take off, but the same cannot be said for landing.

same issue with RC 10.9019
Title: Re: [10.26] Runway length not calculated correctly
Post by: jamespetts on February 13, 2013, 10:53:42 PM
The issue is that the runway length is calculated at the same time that the schedule is calculated, so you need to trigger a recalculation of the schedule in order to get the new runway length to have effect for convoys which have already calculated their schedules.

What exactly is the issue on landing, however?
Title: Re: [10.26] Runway length not calculated correctly
Post by: ӔO on February 14, 2013, 12:08:12 AM
This should reproduce the problem.

http://dl.dropbox.com/u/17111233/runway_test.sve


The airplane is stuck on take off, but after opening and closing its schedule window, it should take off.
However, despite having the same runway length, it will refuse to land on the other runway.
Title: Re: [10.26] Runway length not calculated correctly
Post by: jamespetts on March 29, 2013, 02:47:13 PM
Hmm - I seem to have forgotten about this one: apologies. Can you check to see whether you can reproduce this on the latest RC?
Title: Re: [10.26] Runway length not calculated correctly
Post by: ӔO on March 30, 2013, 03:54:35 AM
yes, it still persists with 11.9002

Some of the airplanes are bugged with 65535m, but even using the one with 24000m runways seems to not work correctly.
Title: Re: [10.26] Runway length not calculated correctly
Post by: jamespetts on April 04, 2013, 09:24:36 PM
Hmm - this is an odd one. Have been looking into this. This cannot be reproduced with the latest development version of the pakset, but can be reproduced with 0.8.4. A more detailed enquiry shows that the runway length data appear to be corrupted in that earlier version, but it is not entirely clear why. As this problem now only occurs with a somewhat obsolete version, it does not seem on the face of it economical to spend a long time trying to fix this legacy issue. Thank you for reporting it, however.
Title: Re: [10.26] Runway length not calculated correctly
Post by: killwater on May 05, 2013, 03:16:32 PM
Hello
Is it possible to get this not legacy pakset with working planes?
Title: Re: [10.26] Runway length not calculated correctly
Post by: jamespetts on May 05, 2013, 03:31:20 PM
You can either compile it from the Github repository (https://github.com/jamespetts/simutrans-pak128.britain), or wait for its official release, which should not be too long now.
Title: Re: [10.26] Runway length not calculated correctly
Post by: killwater on May 05, 2013, 05:09:19 PM
Tried to compile it, but no airplane images available and pak file has 36kB. If I load the game I cant see any airplanes at all. Any hints?
Title: Re: [10.26] Runway length not calculated correctly
Post by: jamespetts on May 05, 2013, 06:20:06 PM
Hmm - what was the method that you used to compile it?
Title: Re: [10.26] Runway length not calculated correctly
Post by: killwater on May 05, 2013, 06:23:22 PM
I put the makeobject.exe to air folder. Then I accessed it in commandline by Makeobj.exe PAK ./air then it gave me resulting file air.pack
Title: Re: [10.26] Runway length not calculated correctly
Post by: jamespetts on May 05, 2013, 06:34:04 PM
Ahh, you need to run the automated scripts to make it work properly, not least because there are larger aircraft which are not actually 128 pak size, and because you need to add command line parameters to makeobj...
Title: Re: [10.26] Runway length not calculated correctly
Post by: Milko on May 06, 2013, 08:12:23 AM
Hello

Quote from: killwater on May 05, 2013, 06:23:22 PM
I put the makeobject.exe to air folder. Then I accessed it in commandline by Makeobj.exe PAK ./air then it gave me resulting file air.pack

Windows: I use this file to build pak128brit-ex http://forum.simutrans.com/index.php?topic=6482.msg111753#msg111753. Open a windows command line to launch.

Giuseppe
Title: Re: [10.26] Runway length not calculated correctly
Post by: killwater on May 06, 2013, 11:57:36 AM
Yeah tried that too. It gives me empty folders.  Where should I put makeobj and script file?
Title: Re: [10.26] Runway length not calculated correctly
Post by: Junna on May 06, 2013, 02:21:23 PM
Quote from: killwater on May 06, 2013, 11:57:36 AM
Yeah tried that too. It gives me empty folders.  Where should I put makeobj and script file?

In the air folder.

Just write

makeobj pak128 ./ ./

in a text document and save, change its extension to .bat, and run. If you have extracted all, it should work.

For the 192 and 224 folders, you need to  change the "pak128" to "pak192" or "pak224" respectively. Just writing pak will give 64 and thus won't work.
Title: Re: [10.26] Runway length not calculated correctly
Post by: Milko on May 06, 2013, 04:17:54 PM
Hello

Quote from: killwater on May 06, 2013, 11:57:36 AM
Yeah tried that too. It gives me empty folders.  Where should I put makeobj and script file?

On my PC:

Base dir c:\pak (could be different)

files and dirs on c:\pak :

26/10/2012  11.29           236.032 Makeobj.exe
19/12/2010  12.22           421.200 msvcp100.dll
18/03/2010  09.15           770.384 msvcr100.dll
23/01/2013  17.59             2.265 pak128Britain.bat
06/05/2013  17.51    <DIR>          pakdest
06/05/2013  16.27    <DIR>          paksource                          <- Here files extracted from github (without first directory).

I open a windows command shell, then I command line:
c:
cd \
cd pak
pak128britain .\pakdest .\paksource

In folder pakdest you will find the pak.

Giuseppe
Title: Re: [10.26] Runway length not calculated correctly
Post by: killwater on May 06, 2013, 09:16:21 PM
Thank you for help.it worked. However they are of no use due to giving loss even if flying full (ford trimotor). Should I add more then just air folder?
Title: Re: [10.26] Runway length not calculated correctly
Post by: jamespetts on May 06, 2013, 09:25:35 PM
How far are they flying full? Very short distance trips should not be profitable by air.
Title: Re: [10.26] Runway length not calculated correctly
Post by: killwater on May 06, 2013, 09:46:09 PM
Not so far - about 100 km. I loose 1.5k each month on every plane.
Title: Re: [10.26] Runway length not calculated correctly
Post by: jamespetts on May 06, 2013, 10:18:55 PM
It's hard to see how a real life airline could be profitable with such short runs.
Title: Re: [10.26] Runway length not calculated correctly
Post by: Milko on May 07, 2013, 08:38:45 AM
Hello

Quote from: killwater on May 06, 2013, 09:16:21 PM
Thank you for help.it worked. However they are of no use due to giving loss even if flying full (ford trimotor). Should I add more then just air folder?

All aircraft that I drew are not balanced (with regard to the variable costs, all other values ​​are aligned to the data sheets). I have tried to provide a relative assessment each other, but in absolute terms may be too expensive or inexpensive. I also design but lately I do not play, so I find it hard to understand if you need to lower or raise the variable costs. If losses are high probably should see lower costs.

Giuseppe
Title: Re: [10.26] Runway length not calculated correctly
Post by: killwater on May 07, 2013, 06:08:31 PM
Well it's an island to island route so passengers have no other choice. Lack of ticket prices is a problem in circumstances like this one. What is aimed as a minimum profitable air travel length?
Title: Re: [10.26] Runway length not calculated correctly
Post by: jamespetts on May 07, 2013, 06:14:02 PM
What about sea transport? Not all islands have air services for precisely this reason.
Title: Re: [10.26] Runway length not calculated correctly
Post by: killwater on May 07, 2013, 06:32:34 PM
Its so slow and I don't like slow, but I think I have no choice.
By the way those newest changes - walking passengers combined with time tolerance are in my opinion one of the biggest steps in Experimental. Everything is self-regulating now and looks much more realistic. Really great job.
It is a little bit frustrating though that player lost most of his influence on city growth but I think that it is a good direction to follow.
Title: Re: [10.26] Runway length not calculated correctly
Post by: jamespetts on May 07, 2013, 09:41:22 PM
I should note that the pakset has not fully been balanced, so it's possible that the costs will need adjusting - but balancing will be a big job and will have to be done in a coherent way - after the next release. However, it's not necessarily a sign of trouble if an aircraft, especially not an aircraft not built to be particularly economical, cannot make a profit over such a short run. From what I understand, that particular type of aircraft were not in fact used on such short runs, and, in the 1920s and 1930s, small islands were generally served by boat.

I am very glad that you like the new features and that they help balance - the planned addition (after the next release) of the new city growth/passenger generation algorithms should help even more.

What do you mean about players losing control of city growth, however; what control did players have in previous versions that they do not have now?
Title: Re: [10.26] Runway length not calculated correctly
Post by: killwater on May 08, 2013, 06:21:51 PM
Well cities were mainly growing when you were serving them with a decent transportation. Now they appear to grow on they own even without any kind of public transport. Or maybe I am wrong? But it is a good thing. World should not be player centered I think.
Title: Re: [10.26] Runway length not calculated correctly
Post by: jamespetts on May 08, 2013, 09:38:42 PM
I see: people will walk to their destination. Cities well served will still grow faster than ones not served, however.