Aircraft - DC3 - not profitable

Topic: Aircraft - DC3 - not profitable (Read 4149 times) previous topic - next topic

Aircraft - DC3 - not profitable

August 10, 2009, 02:57:11 pm

I am testing 6.3 with the standard Pac64 set and timeline on. Year is 1954. Map is 255 - 765. I have 4 airports with 3 routes. One route is flying 100% full.

The problem, non of the DC3s make any money. Even the one flying 100% full.

Most interesting is that the DC3 flying 100% full replace bus chain with 3 transfers. Every bus in this slow travel solution made money.

Also of interest is the load times on plains. I have a tram that loads 130 p****engers in a little less that 2 hours at a stop. The DC3 is tacking about two days to load 33 p****engers. This seems very odd.

Re: Aircraft - DC3 - not profitable

Reply #1 – August 10, 2009, 06:17:28 pm

Jeffatsqui,

thank you very much for your feedback: that is most helpful. It is very useful to understand what is happening with balancing issues. Could I trouble you to upload a saved game?

A preliminary suggestion would be to try using the pak128.Britain-Experimental speedbonus.tab file here: Simutrans-Experimental is not balanced to have different speed bonuses for different types of vehicles in the way that Simutrans-Standard is. That will mean that fast aircraft will earn a higher revenue per p****enger per tile than slow 'buses. If you do that, however, remember to re-set your speed_bonus_multiplier_percent in simuconf.tab to 100.

Also, how far does your aircraft travel, and to what is your scale set? Just as in real life, aircraft in Simutrans-Experimental will not be profitable except for very long journeys, several hundred kilometres at least. If you are using an aircraft to fly the equivalent distance of, say, 80km at whatever scale that you are using, then it is not surprising that it will not make much of a profit, as the revenue advantage of the aircraft's higher speed is only fully realised on longer journeys. Also, a fully compatible pakset would likely set the comfort rating of an aircraft to be higher than the default of 100.

As to the loading times, the loading time for an aircraft is in Simutrans-Experimental, just as in real life, many orders of magnitude longer than that of a 'bus. That is another factor that makes them uneconomical on very short routes, and that should be borne in mind when using them. Remember, in Simutrans (both Standard and Experimental) there are two inconsistent time scales: the macro scale in which years are measured, and the micro scale in which speeds are measured. Waiting times are calibrated to the micro time scale, not the macro time scale, and are not treated as being the equivalent of hours for a 'bus or days for an aircraft.

Incidentally, the loading times of individual vehicles can be customised by the pakset author. The default (of 30 seconds of real time for an aircraft, and 2 seconds of real time for a 'bus) is applied only when the pakset does not specify a specific loading time (as will be the case with any Simutrans-Standard pakset).

Re: Aircraft - DC3 - not profitable

Reply #2 – August 11, 2009, 03:13:39 pm

Thanks jamespetts,

Using your pak128.Britain-Experimental speedbonus.tab make the behavior of the DC3 very realistic. They now behave like other transportation (bus) vehicles. At high load factor the DC3s now make money, at about 50-60% they are break even, and if it is not making money I can put in a min load to control costs. This is very exciting.

On your other questions:
* I am using the default .25 K per square
* The long axis of the map is just under 200 Kilometers
* Airports are 50K, 75K, and 100K apart
* DC3s on the longer two routes make money

I have not tried the larger planes coming on line. I will continue to play with this map to see if the newer equipment also behaves with your speedbonis settings.

On a side issue, we at OSNV.org and the Reno Lions Club use Simutrans as an economics lab for kids. We provide the kid with older PC that we reconfigure with VectorLinux. Because of the CPU and graphics limits we use Pac64. We have hundreds of these retask PCs with Simutran out to the kids.

Because of your excellent work on improving the economics of play I am moving to the Experimental Simutrans. However, I need to know that Pac64 will be fully supported as we can not move to 128 on the older PCs.

Re: Aircraft - DC3 - not profitable

Reply #3 – August 11, 2009, 07:38:16 pm

Jeffatsqi,

thank you very much for your response - feedback like that is extremely helpful. I am very glad that using the updated speedbonus.tab made things work for you. Please do let me know if there are any further issues :-)

As to Pak64 and Simutrans-Experimental, as you know, Simutrans-Experimental can work with any pakset, but only a pakset set up with Simutrans-Experimental specific settings can take full advantage of the specific features of Simutrans-Experimental, such as comfort, loading times, catering, upgrading, way constraints, etc.. Updating a pakset takes a great deal of time and effort. I am currently working on Pak128.Britain-Experimental, which will be a fully supported branch of Pak128.Britain. I am not able to work on more than one pakset because of time constraints.

Pak64 is, however, open source, meaning that anyone minded to do so could make a Simutrans-Experimental branch of it. I know that Bernd Gabriel has started work on a Simutrans-Experimental branch with 19th century vehicles, but I do not know how completely that that takes advantage of Simutrans-Experimental features, nor whether he plans to extend it any further. Details on Simutrans-Experimental compatible paksets can be found here. If you were minded to do so, you could produce a fully compatible version of Pak64 yourself, using Bernd Gabriel's version as a starting point. I could certainly ****ist by providing guidance as to the best way to calibrate things, but would not have time actually to edit the .dat files myself.

As to the hardware requirements, I know that Simutrans-Experimental is in and of itself more demanding than Simutrans-Standard, pakset aside (although it has been made more efficient recently): if you are using older computers, it might help to check them first before switching them all over. Might I ask what the specifications are of the computers that you use? One thing that makes a big difference with Simutrans-Experimental is the amount of memory, and you might find it relatively inexpensive to upgrade the memory on older computers. At least 1Gb is recommended for a large map with Simutrans-Experimental.

In any event, I am most honoured that you would consider using Simutrans-Experimental for your economics project, which is doubtless most worthwhile. Thank you again for your feedback, and do let me know whether you have any other queries or issues.