Speed adjustment considerations

[Gumboots]

And, probability works like this, think of coin tosses. Flip once, you get heads or tails, while the average is 0.5 in the long run, the first result is 0 or 1, very far off from 0.5. The more you flip, the closer you will get to 0.5.

Teach yer grandma to suck eggs. I've had to explain it to other several times, although I didn't bother getting into standard deviations and all that (wanted to keep it simple).

So if breakdown chance is checked often, there shouldn't be the kind of variability I'm seeing. Your theory of one check per run, though it doesn't make sense to do it that way, actually makes perfect sense interpreting the results. Breakdown chance cannot be getting checked often. Then another random to determine where in the run it breaks down would explain everything.

Yup, I'm starting to think something like that is likely. The game does track total locomotive miles for oil, water and sand consumption, but it doesn't necessarily have to use total miles for breakdowns.

I almost always use maintenance spurs, and it's not uncommon for a loco to crap itself between the station and the facilities, or between the two facilities for that matter. This can be when it is full of oil, and is also obviously when it is hauling no load at all. Then other locos with the same rating will haul full consists reliably for yonks. So yeah, checking per trip is probably how it works.

One way of checking for this, in rough terms anyway, would be to test locos doing a lot of short hops against locos doing long runs. If it is only checked per trip, the locos doing long runs should give lumpier data. If it is checked against actual mileage, there should be no difference.

[low_grade]

It's in Lirio's pack, not sure where it came from, available in 1900, 120k cost, 11 maint, Above Ave in everything 68 top speed, not the best puller, looks like a Suburban.

Edit: did another test on the Zephyr to check oil consumption vs mph (it did a bit better than already creaming everything lol.) 126% oil consumption over the year * 621 miles to empty was it? = 782 miles, while 30mph average *2.35*12=846 miles, while this loco has a dead flat route and doesn't run into traffic, so i'd say the mph in the ledger is legit.

[Gumboots]

Weird. No idea what it's supposed to be.

[low_grade]

And I personally am not interested in checking that, since while annoying, crazy breakdown results don't appear to effect the bottom line too much, and nothing you can do about it, aside from on a very tenuous start save often and reload if the fates are against you. Crashes however have to be disabled!

[RulerofRails]

Thanks. I'm about to take a look at the data now.

Just for interests sake, in recent-ish games I noticed a few engines that broke down a moment before they entered a service facility or station. A notice of breakdown was produced but they escaped any stoppage penalty and the black smoke never appeared. They did their business as usual. Seems to be that if they roll into the station/facility while they are supposed to be slowing to a stop, they can escape.

Agree that reliability testing isn't worth a lot of effort. Even if successful, without being able to edit the exe, it would do little good.

[CeeBee]

Is the RRT3 v1.06 loco update beta 1 the most up to date version of Lirio's Balance and realism download? It's the only file I've found so far after looking through umpteen pages and posts.

[low_grade]

That't the one I'm using, and there are a few keepers in there. Not that I really want it, but I can't figure out how to get her Connie II into my main install...

[RulerofRails]

After looking through your data, I think I know why your fuel estimates are high: the time spent loading/unloading at the station doesn't count towards average speed. My test for that is in the very first post of this thread. Basically a full train takes approx. 1 month for a complete station stop (8 cars).

Faster trains will travel more. To accurately determine the number of station stops in a year we need to know the length of the run. In a real world application it's too troublesome to work it out. So it's a guess, but as you eloquently put it the other day, we can improve the odds of our guess.

In terms of the limits for consumption: a steamer needs water at 227 miles. I think we can agree that it's not typical to have steamers run out of water on a one-way trip. BTW, Oil limit is 610 miles. Forgot to reply on that.

The map for the current test has fairly sparse cities on average. I did a quick job at what were reasonably efficient connections to average 118 miles between cities. In an actual game I would probably route a bit longer to avoid more of the grades/ have a cleaner setup. In a real game I will probably want to run trains on some connections that are not the ones that appear to be the "shortest". This will be determined by cargo flow more than anything. So I'm guessing that the average figure may be in high 120's or even 130s for one of my plays. This is if I only run between the closest cities. If I skip cities for some runs, (don't know what routing you are using, but you are using just over 2x trains for the number of cities), this can potentially rise more, could be up to 50% (180-200).

Guessing an average of 180 between station stops, the fastest locos in your test will likely stop at stations at least 3 times a year. Using 9 instead of 12 in your formula (Average Speed*2.35*9) for the fast locos will make a lot more sense of the fuel cost prediction. Then obviously the slower locos will have less stops (really the average will be a decimal such as 3.2, 2.8, 2.1 etc.) so taking that into account should make things line up better.

---

One of the first things I looked at so far is profit per load carried. Lirio's Camelback is highest on $20.6 per load, followed by the Zephyr and P8 on $20.1.

There is very little relative spread in the revenue received per load of freight carried. $16.2-18.5.

For express (average 73% pax) we have a range of $28.5-33.5, With the exception being Lirio's Royal Hudson at $40.1, maybe an error in load count?

Which brings up a point on usage during a game, surely if I buy a few extra Zephyrs I can get a better economic result than with the GG1 (even if it wasn't necessary to buy electric track). They are close enough already that I don't for-see enough extra traffic inefficiencies being created to offset this.

[low_grade]

Ah, that makes sense, the train I keep watching is the one going from New Orleans to Mobile, starting the year about 1/3 of its way to Mobile. Slow trains will make it to Mobile and back to New Orleans but not quite back to Mobile (like 12mph locos). Very fast trains will Mobile-N.O-Mobile-N.O.-Mobile, though most only go Mobile-N.O.-Mobile and only just get started or fall just short of hitting N.O., or if a bit quicker hit N.O. but don't get back to Mobile. For distance, steamers hit a water tower before each station with about 20% water remaining, so about 182 miles between Mobile and N.O., and all trains hit the service station before N.O. with about 40% oil, so .6*610=366/2=183 miles between, okay both estimates are the same. So many trains are hitting 3 stations, so running 9 months, and covering 500-600 miles, say 500/2.35/9=23.7mph... yup memory serves, I had a bunch of locos average around 24mph.

And yes I'm sure there are a few data entry errors spoiling the results a bit... but actually I think I spotted most of those by now (it becomes obvious there's a problem when you see data so far out of alignment with everything else.)

[RulerofRails]

Yeah, mistakes will creep in. Hope I didn't sound critical, was actually going to go into the passenger appeal ratings so it made more sense to discard the outlier, but then realized it was probably just rambling. Pax appeal bonus is only applied after loading. The thing that caused me to re-think my anaylsis was that the Ugly rating means that some passengers will actually be loaded instead of freight. Couldn't think how to make this work without sorting into groups based on pax appeal. So I ditched that.

Just a question, is there a reason you didn't run the EF 66 (cheapest fuel cost in the default game)? Would be nice to know how it compares to the GG1. I forget if the intro date was changed in 1.06 or Lirio's stats? So maybe not C-era for those.

Just for confirmation, does your test use 100% double track?

[low_grade]

Taking a break from testing to try the Philippines map, but yeah, I tested everything available before 1950 (even if the 1949 entries are really D-era for most of their lives...) I have the EF66 available 1966. Yet to test the D era, and I may still do a test with more grades, but until I think of something else that I don't quite understand and need to test, I think I've learned about all I'm going to about the game from this round of testing. And I did learn a lot, thanks for the feedback!

Wait, so the events in Orient Express (assuming you don't "cheat" that map) which reduce station load/unload times are useless to boost average speed?!?

[Gumboots]

Wait, so the events in Orient Express (assuming you don't "cheat" that map) which reduce station load/unload times are useless to boost average speed?!?

Yes. They'll boost revenue, due to reduced turnaround time, but they have no effect whatsoever on average speed. Average speed is only ever counted between stations, and when hauling a load, so the relevant time is from when the "Loading" notice vanishes to when the "Unloading" notice appears.

Maintenance spurs don't get counted because there's no load being hauled between the station stops, which is why using maintenance spurs boosts average speed: you only get checked for the period when your loaded train is in motion, not the period when it's stopped while loaded.

[Gumboots]

I think we can agree that it's not typical to have steamers run out of water on a one-way trip.

I wouldn't say it's typical, but I would say it's common. Off the top of my head, on average I might have around 1/4 of my trains using almost all their water on a one-way trip. They won't necessarily run out, but they'll be so low that they're due for a top up anyway.

[sbaros]

5/ Fuel cost is completely unrelated to map economy, and that's that. It's one of RT3's simplifications and there's no point worrying about it, so I don't worry. I thoroughly recommend this brilliant strategy.

Maybe we could write formulae within scenarios where production / transport of the various fuels (and generated MWh as well) affects operating costs of the traction categories concerned?

[Gumboots]

You can't adjust fuel cost by event. The editor will let you write fuel cost events, but they don't actually work when playing. It's one of the known game bugs (acceleration is another one).

You can write events to be a proxy for fuel costs, based on your estimate of what the adjustment should be. I have used an adjustment to company overhead for a "fuel cost event", based on recorded fuel cost and overhead cost during testing.