Long time no see

[bombardiere]

Somehow I dug up this old thread and during the weekend I did do some calculations. (I like stats but I am terrible in the math. )

Note: The original formula has an error. The correct line is following:

((Cylinder diameter^2 * Cylinder stroke * boiler pressure * speed) / (driver diameter*375))/4

Now I am even more convinced that Chris was on to something. Now I strongly believe that Poptop based its #2 on Cylindar horsepower.

But of course, so that it would no to be too easy, I say that they only based it on horsepower. While I did my counting, I notice that the speed must a factor. Just as I speculated in my last post over ten years ago.

I have come to conclusion that they had a speed baseline around 50 - 55 mph. So this would be a "standard" horsepower and if a loco is faster or slower, then #2 would be adjusted accordingly. This basespeed is to be divided by the game top speed and that result is a multiplier for Cylindar horsepower calculation.

For example, if the loco's top speed is 100 mph then the result of Chris's formula must be multiplied by 0.5. And if speed is 25 mph, then the multiplier is 2.0.

I realise that this is not a scientific method, but for me it gives a very good baseline on which base new locostats. I am aware that user made locos have wildly different stats. Including ones I have made. So I think it could be useful to have a common base on which extra locos are made. I know that this will not give perfect results and even after top speed adjustment, some freight and early locos will still need some help.

Another thing is that I realised that Poptop did indeed looked at the tractive effort. #1 is very much based on the tractive effort. But again they did not do it too simple for us. Firstly, #1 have a correlation to tractive effort of kilos. I mean that if I divide metric tractive effort by 100 then I get close to what Poptop has used on #1.

I found this very odd that why Poptop, which used imperial measurement, would suddenly use metric on tractive effort. Then I realised that it is not 100 kilos, but imperial tons (2000 ib) divided by 10. Because the maximum number of cars is 8, perhaps they though that using the full tractive effort on RRT3 was too much and then decided to divide tons figure by 10 to fit it better in the game environment. In similar fashion that Chris discovered that cylinder horsepower was divided by 1000.

Again tractive effort is not an exact #1. It is also to be adjusted. Either by speed or wheel arrangement. I am not sure.

[Gumboots]

Have you read this?

Setting the record straight on RT3 locomotive pulling power

I think Lirio may have been right, but TBH I also think it's irrelevant these days. I really don't care how they derived their numbers. All I care about is getting balanced rosters for various regions of the world.

Ok, say pulling power is basically tractive effort. In RT3 you have one set of cargo cars to do all trains, yet in real life freight and express train weights varied dramatically between countries and regions. US trains, for example, historically made UK trains look like toys. This means that, as I've posted before, if you want to have accurate tractive effort figures then you will need different cargo car sets to match those figures. It's pointless giving a Big Boy and a UK 0-6-0 accurate hauling power and then expecting them to be useful in RT3 while hauling the same consists. It would be as daft as a cat trying to bury its **** on a concrete floor.

Theoretically we could make a range of cargo cars that were matched to actual train weights in various regions, but in practice it's unlikely any of us will ever be bothered doing it. What we want is one set of cargo cars and a bunch of locos to haul them. This means they have to be balanced so that UK locos are unrealistically strong relative to US locos, otherwise it won't work.

[bombardiere]

Oh. No I have not read it. Ten years of absence and I have missed a lot. Thanks.

Ok that was interesting reading. So Poptop didn't have exactly a speed modifier, but they pick a speed "rating". Whatever, but it has same implication that I thought.

In fact, the most interesting thing which arose from my own calculations, was a formula, that if I drop the high speed of No 999 or Mallard, then how much I must raise the #2 to keep rough same ratio.

You are right about what you write about the practical meaning of these formulas in the game's environment. Very little, as it is a game, it simulates real environments very lightly. If at all. Still I find some use of this. Especially now that Lirio explained what that formula actually is and how shorten it.

Because while making a new engine, it is useful to know where the ballpark is. That doesn't not mean that #2 must be exactly as it comes from calculation / tractive effort, but at least for me it is good to know what is rough value. So from where to start looking a good game balance.

Still for me it would not be an issues if UK locos are "toyish" in comparison to US ones. Perhaps it would encourage different strategies. Use a larger number of shorter trains. Ok, in many UK / Europe scenarios that means that oil, water etc should be adjusted accordly. Due shorter geographical distances.

I will be reading more...

[Gumboots]

Still for me it would not be an issues if UK locos are "toyish" in comparison to US ones. Perhaps it would encourage different strategies. Use a larger number of shorter trains. Ok, in many UK / Europe scenarios that means that oil, water etc should be adjusted accordingly. Due shorter geographical distances.

Geographical distances don't matter in RT3. All that matters is map scale. That Latvia map I've been messing around with is larger than the PopTop South East Australia map. I know which area is bigger in real life but RT3 works out distances on the basis of map pixels.

The larger number of shorter trains would require other adjustments, due to higher running costs. So you might want lower purchase price and maintenance costs, for example, or you might want more double track and therefore lower track costs.

[Firefly 2-2-2]

UK Locos are really a different sort of beast. The 2-2-2 Firefly, Crampton, Duke and Stirling combo is really good for Express Service, but they're gas guzzlers and are notoriously unreliable. What that means is that you'll often get more Express income, but will have to pay more attention to oil and water levels and don't force them atop too steep grades like you would with Consolidations.

As it was said, some maps of Europe are bigger than certain maps of Australia and the US. That doesn't matter in the long run as long as the terrain is tidy, comfy and flat. Case in point is Netherlands scenario, where the terrain is almost fully flat - you don't need Consolidations or any other rugged train in here, rather it's the Fireflies and the Stirlings, with their shiny "SUV" like gas guzzling, pretty boy style and their lower reliability ratings will truly shine.

In other maps, eg. Brazil with the very steep grades, that wouldn't really work. And train coolness really makes a difference, not only for Express but if *every* train in your inventory is ultra cool, you get a boost in income. OF course, that makes quite a lot of difference. Even though with the later America trains, like eg. the Pacific and the Northern, you're still able to purchase ultra cool stuff for your Express lines, European trains have that advantage in that they're both functional and good looking. The Class A1 is a case in point: it's flexible enough to serve both as a Workhorse and an Express train. Some specific steam lines, if you decide to electrify, might also be retained for Express service using the best available steam engines.

[Gumboots]

The 2-2-2 Firefly, Crampton, Duke and Stirling combo is really good for Express Service, but they're gas guzzlers and are notoriously unreliable.

Not really. The Firefly will explode if you even look at it sideways (I throw them away after three years) but the Crampton is no less reliable than several other locos of that era. The Duke is as reliable as the Connie, and has the same running costs for an equivalent consist and mileage. The Stirling is less reliable in RT3 (wasn't in real life) but still has the same running costs.

And terrain doesn't affect reliability. We checked that. It's purely dependent on loco age, reliability rating, oil level and consist weight.

And train coolness really makes a difference, not only for Express but if *every* train in your inventory is ultra cool, you get a boost in income.

No, you don't. Passenger appeal ratings only affect the prices paid for loads of passengers. They have no effect on prices for loads of troops or mail, or loads of any freight. If you aren't hauling passengers, it doesn't matter how ugly the loco is.

And there's another trick to it, in that the passenger appeal boost to price is only applied after loading. It's not considered when auto-loading an "Any express" consist. If mail is $23k/load and passengers are $22k/load the train will load mail preferentially, even if the loco has an Ultra Cool rating which would boost the pax price to $26k/load. In that situation, the only way to get it to auto-load pax would be by using a pax-only consist.

[Firefly 2-2-2]

... Later in the game, when I have several hubs set up, I make custom consist Express trains with 6 passenger cars + dining car + caboose. Sometimes I throw away the caboose, anyway.

An Ultra Cool train running a relatively long distance between two hubs, with a speed record and dining cars can reap a lot of money under this scheme. I know this because I had one such Northern 4-8-4 running under this. I generally leave mail for the "All Cargo" workhorse trains.

I'm also gonna try to implement this in one of my future games: building an all-passenger, very very flat grade line with almost no curves and use only high speed trains on it, starting with the Shinkansen, connecting four or more relatively distant hubs. It's doable and might be very profitable.

The Firefly isn't *that* unreliable. You can safely replace it with Cramptons or even the Beuth, but for Express it's as good as it gets until the late 1870's. Mind you, as a "second generation" train in-game (I have my own quirks for classifying engines in-game), coming after the Patentee and the Planet, and even compared with the Beuth, it's an extremely fast train and that makes a lot of difference.

In my sort of classification, steam engines in the 19th century would be roughly like this:

1st Generation: Planet, Adler
2nd Generation: Firefly, Beuth, Crampton
3rd Generation: Iron Duke, Stirling, Fairlie

etc...

[Gumboots]

... Later in the game, when I have several hubs set up, I make custom consist Express trains with 6 passenger cars + dining car + caboose. Sometimes I throw away the caboose, anyway.

An Ultra Cool train running a relatively long distance between two hubs, with a speed record and dining cars can reap a lot of money under this scheme.

Yeah, sure, but that's because it's hauling pax. For trains that aren't hauling pax, coolness makes no difference.

The Firefly isn't *that* unreliable.

Oh yes it bloody is. The reliability rating curve is parabolic (literally). The Very Poor rating of the Firefly gives it a 25% higher chance of breakdowns, when brand new, compared to a loco with a Poor rating (Planet, Beuth, etc). Not only that, but its breakdown chance escalates faster with age. That doesn't mean it can't be useful, but my preference is to restrict it to high-paying routes where I want the faster turnaround and am prepared to pay the greater cost in loco replacement.

You can safely replace it with Cramptons or even the Beuth, but for Express it's as good as it gets until the late 1870's.

Depends on the situation. On dead flat terrain the Crampton is 25% faster when hauling 7 post-1850 pax cars + dining car, but up a 2% grade the speeds are equal. The Firefly has a distinct advantage climbing a 4% grade, but OTOH what goes up must come down and the Crampton will be faster downhill. The Stirling is even faster on flat terrain with the same consist, being 50% faster than the Firefly under those conditions, but again it's no faster up a 2% grade and it's slower up a 4% grade. But it's faster downhill too, and both of these locos are substantially more reliable than the Firefly.

In my sort of classification, steam engines in the 19th century would be roughly like this:

1st Generation: Planet, Adler
2nd Generation: Firefly, Beuth, Crampton
3rd Generation: Iron Duke, Stirling, Fairlie

etc...

That's about right, but the game's Duke isn't an Iron Duke. The Iron Duke was an 1847 GWR class of 4-2-2's. The game's Duke is this one: https://en.wikipedia.org/wiki/Highland_Railway_F_Class

[Firefly 2-2-2]

Yeah, sure, but that's because it's hauling pax. For trains that aren't hauling pax, coolness makes no difference.

Yes. I usually haul my Mail Express on freight trains, anyway. Or the general cargo trains. It makes no difference, really, but separating mail from passengers can be really profitable when making custom express routes (think of them as luxury passenger liners). They can be optimized for one special role only.

Depends on the situation. On dead flat terrain the Crampton is 25% faster when hauling 7 post-1850 pax cars + dining car, but up a 2% grade the speeds are equal. The Firefly has a distinct advantage climbing a 4% grade, but OTOH what goes up must come down and the Crampton will be faster downhill. The Stirling is even faster on flat terrain with the same consist, being 50% faster than the Firefly under those conditions, but again it's no faster up a 2% grade and it's slower up a 4% grade. But it's faster downhill too, and both of these locos are substantially more reliable than the Firefly.

I think I concur with your assessments, anyway. Still, I very much like it since I used to play British Isles a lot and that was the best train available after years of 1st generation slow engines. Plus it's still cool, even 160 years later.

https://www.youtube.com/watch?v=a3509eeqcrE&t=142s
https://www.youtube.com/watch?v=urAwB6i1qzU

Lol my name really betrays my soft spot for that engine.

That's about right, but the game's Duke isn't an Iron Duke. The Iron Duke was an 1847 GWR class of 4-2-2's. The game's Duke is this one: https://en.wikipedia.org/wiki/Highland_Railway_F_Class

Yep, easy to mistake 'em both.

[RulerofRails]

Something that might make the Crampton seem less Reliable is that it is introduced after the 1850 consist weight change (for freight 5->10 tons). So generally consists that you see when you first buy it will be twice as heavy as those when you first buy the Firefly. Obviously Firefly in 1850-1900 is going to have to haul the same amount. But often we base our game decisions on observation which isn't always reliable compared to the data.

I'm here idly wondering that while breakdown chance as displayed in the readout doesn't from my memory change with difficulty level, perhaps the application of how often a train will actually breakdown is less on the lower levels?

Firefly, just to let you know that most of our experience is with Expert difficulty level. If there does happen to be fewer breakdowns on the easier levels we are not used to that.


BTW, Gumboots congrats on 3,500 posts!

[Firefly 2-2-2]

I really don't know. Really. The thing is, until the Stirling comes, my argument was that the Firefly could still see some special pax service.

I've had 80% of my trains being made of Fireflies in a Netherlands playthru and never had that much problem with breakdowns. Bottomline is, I dunno if Difficulty really impacts it, BUT, I always place plenty of Service Towers and Maintenance sheds, especially near hubs, and I never see one of my trains run out of oil. *Besides* I always add Cabooses to all of my trains, especially in the 19th century (after the 20th, when some "Outstanding" Reliability trains come up, I ditch the Caboose for them). All of this might have a considerable effect upon breakdown/crash frequency.

[bombardiere]

Oh, the fact is that Firefly is the most unreliable of all 1.05Ctc engines. It is only locomotive with Very Poor reliability. Crompton, Alder, Planet and Beuth all are Poor.

Still, sure I use Firefly. I kind of like it even though I have to wear a helmet as shrapnels are flying from exploding boilers.

[Firefly 2-2-2]

Lol, that might be true.

[Gumboots]

Yeah, sure, but that's because it's hauling pax. For trains that aren't hauling pax, coolness makes no difference.

Yes. I usually haul my Mail Express on freight trains, anyway. Or the general cargo trains. It makes no difference, really, but separating mail from passengers can be really profitable when making custom express routes (think of them as luxury passenger liners). They can be optimized for one special role only.

Yep, that's definitely a good way to go if the map has the pax traffic to sustain it. Mail is one of the faster rotting cargoes though, along with troops, produce, meat and a couple of others I can't recall offhand. These should ideally be hauled on medium priority, with the more durable cargoes (iron, coal, etc) on lowest priority. But then this also depends on your actual supply needs for industry, and spot prices for any cargoes.

And I get the appeal of the Firefly. They were a very good unit for their day, and in real life they weren't as unreliable as RT3 makes out.

[Gumboots]

I really don't know. Really. The thing is, until the Stirling comes, my argument was that the Firefly could still see some special pax service.

Yes, but it would be better suited to graded routes. The Crampton is a better option on flat terrain, IMO.

I've had 80% of my trains being made of Fireflies in a Netherlands playthru and never had that much problem with breakdowns. Bottomline is, I dunno if Difficulty really impacts it, BUT, I always place plenty of Service Towers and Maintenance sheds, especially near hubs, and I never see one of my trains run out of oil. *Besides* I always add Cabooses to all of my trains, especially in the 19th century (after the 20th, when some "Outstanding" Reliability trains come up, I ditch the Caboose for them). All of this might have a considerable effect upon breakdown/crash frequency.

Oil level has a significant effect on reliability, and having too many service facilities is a good thing in that respect. It also has the advantage that each facility will tend to be less congested, which can make quite a difference on routes that have heavy traffic. It's the same logic as having double stations in a city to ease congestion and speed up turnaround times. For not much extra capital outlay you can get a good improvement in traffic flow.

The caboose will halve the breakdown chance for any locomotive. The RT3 manual says that, which is obviously a reason to doubt it but in this case it is actually true. We've tested it to make sure. You'll get (on average, over enough examples) twice as many breakdowns if you don't use a caboose. The caboose has no effect on crashes though. Crashes appear to be done by a random number algorithm as long as oil level is greater than zero. I think it increases the probability of a crash if you run the loco out of oil, but I haven't done enough targeted testing to confirm that.

One thing to be aware of is that the default dining car and caboose are both much heavier than default express cars. They are even heavier than default freight cars. Default A era (pre-1850) express cars are 3 tons each (pax, mail and troops all the same). Default A era freight cars are all 5 tons. The default dining car and caboose are both 7 tons.

This means that if you are hauling 8 express cars the consist weight is 24 tons, but if you are hauling 6 express + dining car + caboose the consist weight will be 32 tons, for a 33% increase in weight to be hauled. This will adversely impact both fuel costs and reliability, although the reliability with the caboose will still be higher than 7 express + dining car and no caboose. The extra consist weight will obviously also be detrimental to speed, particularly up grades.

For the B era (1850-1899) the figures are 7 tons express, 10 tons freight, 13 tons caboose and dining car.
For the C era (1900-1949) the figures are 13 tons express, 20 tons freight, 27 tons caboose and dining car.
For the D era (1950-9999) the figures are 27 tons express, 40 tons freight, 53 tons caboose and dining car.

[Firefly 2-2-2]

The Dining Car is still worth it, especially since you theoretically ain't gonna carry 8 pax all the time with your Express loads. Suppose a consist only loads 5 pax, in this case the Dining Car will make a difference in profitability. I never ever let my trains run without a Caboose, ok it might be ok to ditch the Caboose after 1914 for some high reliability Express trains, but still...

I mean for most trains this doesn't make a difference. Case in point: once the Duke Class gets past 1900 it starts to struggle, but the Pacific, the A1, the Northern and the Mallard are well capable of handling the extra weight and still fly very fast. These are the ideal Express trains for the period - I usually go for Duke or Stirling -> Pacific -> A1 or Northern, and then keep these until high speed trains come for Express. The Mallard is also top notch, but it's not reliable enough to run one without a Caboose.