Patch: New passenger destinations

[dasy2k1]

sounds like you may be having the same issue as me....

if you have 2 ways of getting from A to B all the passengers will wait for the fastest way regardles of the fact that it takes 4 times longer counting waiting times.

[Stefanouz]

The patch is working, because at other trajects it work, but some trajects it don't.
That the passengers want to take an other route is also not possible, because there isn't.

[Gekko]

Can we get this patch updated to the latest trunk please?

[Roest]

Ok i updated it. It seems to work. I just tested it with 4 bus stations. No guarantees if it really works.

[jub]

Does anybody encounter following behaviour: the connection between two stations is created asymetrical - A has B as a next station, but B doesn't have A as a one?

[Roest]

A connection is established when a vehicle goes from A to B. So if you don't have a route B to A, there wont be a connection.

[trainmaster611]

I'm having a little trouble with this patch. I'm running trains on an ABCD style route (and reverse). I have about 4 trains with identical orders. However, when a train arrives at Station A, and there are about 15 people headed for Station B, only about 2 or 3 will get on the train, and the rest stay in the station.

I don't understand. Why will they not board a train that lists there destination in its orders? All the trains follow the same route and have the same engine, so it should be the same speed.

[Yexo]

I'm having a little trouble with this patch. I'm running trains on an ABCD style route (and reverse). I have about 4 trains with identical orders. However, when a train arrives at Station A, and there are about 15 people headed for Station B, only about 2 or 3 will get on the train, and the rest stay in the station.

I don't understand. Why will they not board a train that lists there destination in its orders? All the trains follow the same route and have the same engine, so it should be the same speed.

Do your trains have shared orders? If not, the patch will detect is as 4 different routes, and thus only 1/4 of the passengers goes on your train.

[Roest]

Also i doubt it makes much sense to report any errors/bugs. The original author of this patch has gone missing and i have no clue what's going on deep inside it. The code has changed quite a bit since the last released version of this patch and honestly i'm surprised it still compiled and does what it does.

[BlueEagle_nl]

Is it possible to get this up to trunk again? Compiling on r12904 gave some errors in station.cmd.cpp, also read the post in your Patchpack for that...

[MJS]

How is the performance of this patch, generally? On my singlecore Pentium at 3 GHz with just 1GB of RAM, I am experiencing jolts at the start of every new day. They last up to 4 seconds. My guess is that's due to recalculation of the passenger destinations. It also occurs with passengers off but mail on, but not with both off. (By the way, I have the daylength patch on as well - would that influence the amount of time needed for calculations?)

Does anyone have suggestions to improve performance? Enabling or disabling the walking passengers helps only a bit, unfortunately.

[PhilSophus]

(By the way, I have the daylength patch on as well - would that influence the amount of time needed for calculations?)

With daylength patch applied the autosave interval defaults to daily instead of monthly (which is not a good idea for low daylength factors). Maybe what you're noticing is the delay due to the autosave?

[MJS]

I don't think so - the delay disappears when I turn paxdest off. Besides, I have manually set autosave to a lower interval, and it displays a message in the statusbar when saving, which isn't there during the daily delays... But thanks for your suggestion.

[Touqen]

I'm working on bringing the patch back up to trunk. It's mostly there just ironing out a few kinks. It should be available in a day or two.

I know that there were some outstanding issues with the patch, but I haven't read through all the posts in this thread so if those could be summarized I could get to work on fixing/implementing them after the patch is up to trunk.

[prissi]

I honestly doubt that this patch will make it into the trunk, since it is even larger, eats more CPU and creates more questions than the last two attempts on this subject. Worse of all are the desync in multiplayer, which were also a show stopper for the last patch.

NB: The patch was updated by keeping obselete stuff alive. I have the impression, currently a major overhaul would be needed.

[XeryusTC]

I think this patch needs a complete rewrite for it to get even a remote chance to be included in trunk.

[Touqen]

Yea. It definitely needs a lot of help to get it trunk worthy. I'm just getting it up to trunk so that people can enjoy it and then will embark on the task of overhauling it so that it performs better and actually adheres to the coding guidelines.

[PouncingAnt]

Personally I think if people stopped updating this patch, it would give everyone more motivation to create it again from scratch.

[Touqen]

Personally I think if people stopped updating this patch, it would give everyone more motivation to create it again from scratch.

In that case, I'll just do that.

[bokkie]

I do like the walking passengers from the last attempt, because in this way it's possible to connect networks from different competitors. On the other hand, it should have a lower priority than getting a working and trunk-ready paxdestpatch and it can always be added later on.