Bridging the Mississippi

[Orange46]

To me, it looks like the train just creeps between two close services, while it actually accelerates to a service further away. If it actually does creep, then you lose time if two stops are really close. I usually place services spaced a bit from each other, but don't hesitate to bunch them near a station if I am short on cash and need to send trains off their route to use the service.

The way to solve this isn't by math, but by actually saving a game and then building the services both ways and timing the same train doing the double service.

On another matter, it irks me whenever I need to squeeze the most out of services, only to discover that I set the service too close to a switch. The train then just ignores the service unless I set it as a stop.

[RulerofRails]

Re: bunching the stations. I'm realizing that I gave the wrong idea. I do normally put them a couple steps (no set rule, sometimes one or two, etc.) apart where possible. Sometimes if they are within city limits where there are other buildings restricting space or on a hill-side they will actually be hard up against each other. Sometimes I will see the 1mph, but not as a rule. At this point I am wondering if that only happens if both facilities are on the same map cell. I will test it, it's just a matter of when. . . .

[Wolverine@MSU]

I think someone found that "stops" (water/maint/station) need to be at least 4 track segments apart, although empirical testing would be the best way to figure out the optimal spacing.

[RulerofRails]

I did a quick little test session on service facility spacing. I couldn't get around the 1mph penalty by having the service tower on a different map cell to the maintenance shed.

Do you remember if the mention of the 4 track segments rule was on this site? Any more info to be had about it?

I have a suspicion that acceleration and top speed would play into optimum spacing. As such it might vary somewhat among the engines. What engine&cars load combination would you like to see? I can go and run a test or two, collect some data, but how should I display it?

[Hawk]

I did a quick little test session on service facility spacing. I couldn't get around the 1mph penalty by having the service tower on a different map cell to the maintenance shed.

Do you remember if the mention of the 4 track segments rule was on this site? Any more info to be had about it?

I have a suspicion that acceleration and top speed would play into optimum spacing. As such it might vary somewhat among the engines. What engine&cars load combination would you like to see? I can go and run a test or two, collect some data, but how should I display it?

Here's a post by EPH where he said he thought they should be at least 5 apart.

You might be able to find more by using maintenance facilities as a search term.

In this post, Wolvy said that trains won't stop at facilities if they're within 4 cells of a junction. Irrelevant other than there's a 4 there.

Here's a topic about the 1 MPH crawl when facilities are near a station.

Here's some more discussion about facilities by Mobius. - http://hawkdawg.com/rrt/rrt3/hp/an.htm and here - http://hawkdawg.com/rrt/rrt3/hp/ad.htm

I thought this was fixed in the 1.05 patch, or was it the 1.06 patch?

[Gumboots]

In this post, Wolvy said that trains won't stop at facilities if they're within 4 cells of a junction. Irrelevant other than there's a 4 there.

I've noticed another bug he mentions in that post:

This was discussed a bit in another thread, and from the limited testing I've done, I can't quite figure out the logic behind when a train will stop at "non-scheduled" facilities if there are "scheduled" stops in the itinerary. Trains sometimes WILL bypass services, even if they are below 50% (even completely empty) if there is a scheduled stop in the itinerary.

I've had this happen in several scenarios. Example: if I put a service tower and maintenance shed on a short sidetrack off the main line, and use a waypoint to make some (longer distance) trains go via the sidetrack, they will often skip the service tower and maintenance shed even if they are very low on oil, sand or water.

What seems to happen is that the scheduled waypoint overrides the need for a stop, so the train only "sees" the waypoint. Now that I think about it, I have usually done such setups with as short a sidetrack as possible. That means the waypoint is usually pretty much on top of the shed or tower. This may be another case of the "4 square bug", where trains apparently can't see towers within four squares of a junction/switch/whatever you want to call it. It may be that if I used a slighter longer sidetrack that enabled me to put the waypoint 5 squares away from the shed or tower, then the train might automatically stop at the shed or tower if it needed to. I should test that.

[Hawk]

In one of the links in my previous post, Steve (Mobius) Lorenz (you probably don't remember him. He might have been before your time) discusses testing with 1, 2, 3, and 5 cells between.

[Gumboots]

Yes, I read that, but he was talking about average speeds and turnaround times, not about waypoints borking automatic stopping at service facilities.

(Edited for typo)

[Hawk]

I thought it might give you some info that may be useful.

[RulerofRails]

Hawk, thanks for digging up that info!

Steve Mobius reported some interesting stats on his test and his report on the Eight-Wheeler was that you could only increase average speed by 2mph (dropped from 33mph) by putting service facilities at one end instead of in the middle the way the AI do. I sure wish that he had recorded the average speed for the Consolidation with the service facilities placed mid-way. With a lower acceleration rating, I'm pretty sure there would be more disadvantage. But it could be less than I expected.

It can be really confusing to lump things together. So will state that I understand that there is a difference between placing the service facilities a certain distance apart measured in "spaces" (1,3,5 etc.), and the "4 track cells" that Wolverine mentions. I think he is talking about the space between a junction and either of the service facilities. My understanding is that this may be an actual count of track cells, and because some junction angles will block positions for placement this cannot be measured in "spaces" as accurately.