Openrct2: Peeps leaving after 5mins more often than in vanilla.

[zaxcav]

Following discussion originally posted in #2499 but is a separate issue.

player49256

This seems still not to be fixed, tested with 0.0.5.0 build 6b2082f

Load this savefile: https://github.com/OpenRCT2/OpenRCT2/files/369125/Chrisses.Wonderworld.zip

Move towards the north entrance, follow peoples who enter the park, watch how they move before they exit the park.
They seem to be stuck in the first mall, and even at the Information Kiosks at the beginning.

This did not happen in Vanilla (as guest count still plumits to 800 in OpenRCT2 but increase in Vanilla to 3600).

You can go to the thoughts panel, watch all who "Wants to go home" and check the map where they are, almost all are close to the north entrance.

I can provide pictures if requested.

zaxcav

The original problem reported in issue #2499 was with the pathfinding - meaning the peeps who had decided to do something (buy food/drink, go on a ride, leave the park) were having trouble getting to their destination (and thinking they are lost because it was taking too long to get there). Looking at the provided savegame I don't see any obvious pathfinding issues in the area indicated - people that have decided on an action are having no trouble getting to their destination.

I watched a few peeps (hungry and thirsty) enter the park, buy food and drink and then decide to leave. They did not think they got lost or have any problem getting to the shops they chose, nor getting to the exit.

If peeps are deciding to leave the park rather than staying and doing other things, then this is a different issue.

I don't know anything about the peep AI for making decisions on what to do while in the park.
Hopefully someone else can comment on this.

player49256

Just tried in Vanilla, and as you say, it seems to happen there aswell.
However, there's still a huge difference in guest count.

So I did some testing, and every single one of new guests that entered had full need for food and drink (red bar), some were happy and some mad.
Before they managed to finish both their food and drink, they wanted to leave park. They were inside park for max 5 min, except two who is still inside actually riding rides.
(That makes me wonder if all 6 guests would leave on OpenRCT2).

So that lead me to beleive some things are not how it's supposed to be:

1. Are every single guest supposed to have full bar in hunger and thirst when they enter?
2. Does the function of an actuall working map have any effect on this?

If you look at the picture below, the 6 first guests have just entered.
But the 7th guest in the bottom, have been in the park for 8(!) hours.

Maybe this is a seperate issue? But something is really off, which makes vanilla RCT2 able to keep above 3k guests inside the park, while OpenRCT2 barely manages 1k.

zaxcav

The effects of OpenRCT2 having working peep maps are:

1. peeps will choose things to do from anywhere on the map, rather than from only the nearby things + "big" rides (i.e. highest drop height exceeds a threshold or excitement >= 8.00)
2. the pathfinding limits are expanded compared to without a map.

As this is a different issue to #2499 this discussion should be moved to a new issue.
We can't expect other developers to follow this discussion when it's in a closed issue.

I looked into why your peeps are leaving the park. This is a brief description of how the OpenRCT2 code currently works. I know nothing more about this part of the code than what I've learned investigating this just now. I also don't know how the logic in OpenRCT2 differs from the vanilla logic.

All peeps that have been in the park for 5mins and haven't yet been on any ride are forced to pick a ride to go on.
If they don't pick a ride, they leave.
Peeps that are not walking (e.g. sitting) or have food will not pick a ride.

So, in order to keep guests in your park longer you need to encourage them to get onto a ride in the first 5 mins. I would suggest you move your food court further away from the entrance and/or put more rides between the entrance and the food court.

[duncanspumpkin]

Possibly something going wrong with the peep generator. Maybe not setting the correct variables on spawn

[zaxcav]

Initial hunger and thirst values are consistent with the scenario settings in the savegame, which are 92% for both hunger and thirst.

[zaxcav]

I played this saved game in vanilla RCT2 and OpenRCT2 for about the same amount of time to compare the two.
In both I basically just assigned all handymen a small patrol area, hiring additional staff until the paths in the park were covered (ending up with about 40-45 handymen in both), then kept vigil over the park for a bit longer, looking for litter and vandalism.

In vanilla RCT2 it _felt_ like it was very easy to keep the litter and vandalism under control - the handymen in their patrol routes kept up with the litter and there was no further vandalism. This left me with time to do other things. The park rating was higher and the guest numbers continued to climb.

In OpenRCT2 there _seemed_ to be more litter (vomit) and more vandalism and keeping it under control was much more of a challenge - I was often manually placing the handymen within their patrol area to clean up litter they hadn't got to yet and vandalism was still occasionally occurring. The park rating was lower and the guest numbers plateaued and then slowly started to drop.

These observations are all subjective, but it _seems_ that peeps throw up more in OpenRCT2 than in vanilla. Higher vandalism, a lower park rating and peeps leaving the park are all consequences consistent with this.

I did have a quick look at the logic for when peeps throw up, but it's not easy to see the dynamics of the nausea variables. The other problem is not knowing how vanilla works.
There's probably not much more I can do to investigate this at the moment.

[Gymnasiast]

You can try this: edit the OpenRCT2 source to prevent litter bins from filling up and see if it makes OpenRCT2's behaviour more akin to vanilla. RCT2 had a bug where bins would automatically be emptied every time a guest put something in it.

[zaxcav]

I tried reducing the likelihood of peeps throwing up so that the amount of vomit _seemed_ more in line with what I observed in vanilla (again, only my subjective assessment). This had a positive affect on the park rating (as you'd expect), but the number of peeps in the park still plateaued and then started to decrease. With all of the different aspects that affect these things it will be difficult to pinpoint what the difference (or differences) are.
The bin bug is a good suggestion though, so thanks for that. I'll report back when I've tried it out.

[zaxcav]

Issue #2574 looks like another report about the same issue(s).

[zaxcav]

Reintroducing the litter bug from vanilla didn't have any noticeable impact. Guest numbers still plateau and then start decreasing.
It could be that there are multiple minor issues contributing here, including excess litter from overflowing bins (vanilla bug fixed) and extra puke (my observations above).
Issue #3353 also looks relevant for this, so I'm going to see if I can figure that one out.

[zaxcav]

@player49256 This issue can be re-checked now that #4655 has been merged.

[zaxcav]

I've run the savegame in both RCT2 and OpenRCT2 with the current build to compare how the number of peeps in the park varies over time.
As originally noted, in RCT2 the peep numbers slowly increase without any player intervention.
In OpenRCT2 the peep numbers stay roughly level for quite some time (which is an improvement from past behaviour) but eventually still start to slowly drop as before - the time frame of my testing here is less than a full game year.
I've done this a couple of times with consistent results, so we probably can't simply put it down to random peep behaviour. Something is probably still slightly different in OpenRCT2 that's causing peeps to leave when in RCT2 they wouldn't.
This is quite a tough nut to crack.

[player49256]

I tested the park in OpenRCT2 with park maps/no park maps.
Using commit: c26b5ec -x86

I hired some staff, kicked every guest and closed all rides. Now the park has 0 guests and all rides closed.

1. Set price on park maps to MAX (20$)
2. open all rides
3. Wait... I wanted about 35 game years with no maps
4. Guest count flatted out at around 1600 guests

Now the same test again

1. Set price on park maps to FREE
2. open all rides
3. Wait... I waited about 25 game years this time
4. Guestcount peaked at 1400 but then held itself at around 1300.

So setting max price on park maps seemed to net +200-300 guests in park.
Of course, this would need more testing since only 1 test each was made.

[zaxcav]

@player49256 Great that you're running some tests. You get a thumbs up from me for that. Bravo!

I'm not sure setting the map price to max will ensure there are no maps sold - possibly some peeps still bought maps. I would expect lots of peeps had "I'm not paying that much for ..." thoughts regarding maps. I don't know the purchasing code very well, but I _believe_ there's some logic in there that considers whether the peeps find the price of each purchase to be good value or not and this has a slight immediate negative impact on happiness. I have no idea whether there is also a negative impact on happiness associated with "I'm not paying that much for ..." thoughts.
Can you confirm from the info kiosk stats whether any maps were actually sold?
Probably the only way to ensure there are NO maps sold is to close/demolish the info kiosks.

Comparing these two scenarios in OpenRCT2 is probably not very helpful for figuring out why quests in OpenRCT2 seem to leave sooner than in RCT2.
It would be much more useful to run the identical scenario in OpenRCT2 vs RCT2 to see what the difference is.

Definitely keep on testing and reporting your results here.

[player49256]

´So I've done some more testing.
Unfortunately, Vanilla RCT is crashing so cannot test that.

With same commit as previous test, I did a test with two different path layouts at the north entrance.
The first test was done with this layout, Info Kiosks was closed and guests started at zero. 10 Handymen and 5 Mechanics were hired:

At september next year (7?) I had around 490 guests.

Then I redid the layout at the entrance, and removed all Info Kiosks, hired same ammout of staff and started at Zero guests:

After about 1 year I had around 940 guests. Notice tho that in the picture above, guests have a tendency to walk to the first building, and then turn back towards the entrance/exit, just to turn back again, and so forth.

Below, I used latest commit (a29bf99)

So I did another experiment. I reloaded the save, redid the layout and used cheats to get around 3600 guests (which i managed in Vanilla), to see how they were moving inside the park:

Everyone went towards the "Food and Drinks" building, and turned back to the exit when done, a few seemed to enter any rides after eating and drinking.

Also, This ride took well over a year for a mechanic to fix, he seemed to get pretty stuck:

Hopefully this is of some help.
Regards

[zaxcav]

@player49256 Thanks for these tests reports.
A comparison with vanilla RCT2 is essential if we are to have a chance to figure out why guests are not staying as long in OpenRCT2. That RCT2 is crashing for you obviously doesn't help. Perhaps focus on getting RCT2 working again so you can run comparative tests. Worst case you can backup your savegames and custom objects and just reinstall RCT2.

None-the-less, there are a few things that you are observing that aren't working the same as how I _thought_ the AI was working. For example, aimless peeps have a 50% chance of continuing in the direction they are going at an intersection, so I would expect a lot more peeps to be walking into the Mini Bank queue path straight up the path from the entrance. ... unless the queue is full already.
I also note that while peeps will normally ignore short "useless" paths ("useless" meaning 1 tile width paths that don't branch and which are a dead end with either absolutely nothing connected or only a ride exit connected; and also meaning path tiles with the wide flag set) a path leading only to the entrance is not considered useless.
I'll look into these things.

Regarding Desert Fox - in my previous testing with the savegame that you uploaded I noticed that the path next to the ride exit is not connected to the ride exit (the hedge fence across the ride exit disconnected it). This used to be ok, because there was a hack (from vanilla RCT2 I assume) to let mechanics walk into disconnected ride exits. The reason for this hack was fixed recently and the hack removed, so paths to the ride exit must be properly connected now otherwise mechanics won't be able to find the ride exit. If you use the Tile Inspector you can reconnected the path manually through the hedge fence. Otherwise, delete the hedge fence and rebuild the path to connect it properly.

[player49256]

@zaxcav My reasoning in continuing with OpenRCT2 tests was because since things work in Vanilla you could try different components (Like park maps) to try and find how specific things affect the guest count.

I'll try to get Vanilla working again and see if I can do more tests, anything in particular you want tested?

UPDATE:
I Managed to get Vanilla RCT2 running, so i did same test with sending trams to the park. I notice immidietly that it feels like more guests are spreading in different directions on vanilla than on OpenRCT2, where basicly everyone headed for "Food and Drinks" building:

After a while many still wants to leave park, but also here you can see more spread, as where in OpenRCT2 everyone walks in a line towards the exit:

As for Desert Fox, you were right, havent seen that before, because this happened in Vanilla aswell. Some Jungle fence were accidentally placed there, blocking the exit.

So all in all, it feels like more guests just walks around and does things spontaniously while in OpenRCT2 more peoples walk straight to their destination (In this case, food, drinks and exit)

[zaxcav]

@player49256 Testing out different things in OpenRCT2 is great, but it will always be helpful to run the same test in RCT2 to see how it and OpenRCT2 compare. After all, we're trying to figure out why peeps in OpenRCT2 aren't staying in the park as long as in RCT2.
The observations you made in the last post about peeps in RCT2 not heading for food & drink stalls as soon are very interesting and give some possibilities about where to look.

[player49256]

Just tested the newly released stable version 0.0.5.0 build 84977fa.

The park now seems to be able to keep the current population for over a year, without decreasing. Start pop was around 2720, lowest was around 2650, higest around 2800. It kept going up and down inside these numbers.

Theres still a way to go compared to Vanilla, which managed to get 3600 in one year.

I havent been testing so much since i have not had time, just wanted to swing by and report above.

Cheers

[zaxcav]

@player49256 Thanks for the report.
I _think_ most of the improvement comes from peep satisfaction from rides being fixed.
There are known pathfinding bugs in the 0.0.5.0 build (most players only notice how mechanics are affected, but guests are affected too) that won't be helping with this issue. Fixes are waiting to be merged.
Otherwise, specific differences to vanilla still haven't been determined.

[janisozaur]

@zaxcav I don't recall if you hang out on gitter or not; you may want to check this out: http://reddit.com/r/rct/comments/5ko30v/rctc_guests_have_observable_waypoints_for/

[zaxcav]

@janisozaur Looks like it would be a useful feature to have for players and for attempting to debug peep behaviour such as this issue covers. I assume the peep thoughts & actions history have been extended in RCTC with timestamps and locations to support this.
I personally think the history period displayed is too long - it draws a tangled mess in the viewport, but this could be configurable and filterable. In any case, a nice idea.

[jorgechen1993]

While experimenting with something else in OpenRct2, I also noticed this correlation. i.e., the correlation between the hungry/thirsty peeps who first visit food/drink stalls & choose to sit to eat/drink, and the behavior of deciding to leave early without having considered any rides. Is it possible that what initially starving/thirsty peeps choose to consume is related to the issue? Burgers, chicken, and pizza take much longer to finish than other items (and make the peep thirstier), and that's time where the peep isn't enjoying any rides - they're just sitting. Is there a timer that decides how long a peep will stay in a park without seeing any rides?

[zaxcav]

@jorgechen1993 Yes, there is such a timer. A peep that is walking and has been in the park for 5mins or more and hasn't yet been on any rides and currently isn't heading for a ride will be forced to pick a ride to go on. If they don't choose a ride they decide to go home.
Another factor relevant here is that walking peeps holding food will not choose a ride, so not having enough seats for peeps that enter the park hungry to sit and eat will result in them coming, eating and leaving.

That's the basic logic involved.

[player49256]

Just wanted to update.
As of 0.2.0 this problem seems to have been partly solved. Guests stay more in park and there seem to be no problem getting above 3k anymore. Currently at 4400 guests with some more rides built than the save in this thread.

However, there are still some pathing issues, especially with staff not finding exits to rides or getting stuck at times in certain spots.

So, alot of progress has been made, but some issues still remain.