noob question about pulling power in relation to weight

[jballar]

Is there some sort of ratio or rule of thumb about the weight of a train and the pulling power of the engine? What I am wanting to know is how to estimate how many cars to put on a train without affecting performance.

And this gets into another question, when laying track it shows the max speed. What is the best way to determine slope on a track? I am a Railroad Tycoon veteran and the track showed slope while laying.

thanks

[Saint Landwalker]

The short answer: "Not really, and technically every extra car you add will affect performance anyway."

Longer answer pieces:

You have to bear in mind a lot of different factors: Length, curvy-ness, and grades of the route the train is taking; likelihood it will have to stop at some point during its route (e.g. to wait for a junction to clear); what are you really using the train for, anyway?; etc.

Also consider that there's no one "weight-to-power" ratio that would work, as earlier locomotives just don't have enough power to maintain such a ratio the way that later locomotives do. I've had an A4 Mallard hauling ten BC4 passenger cars—a power-to-weight ratio of about 5.76 kW per metric ton. The power-to-weight ratio of the 4-4-2 Atlantic locomotive by itself is about 5.38. For a 4-4-0 American, it's 3.25. And then you start adding cars on there.

To be perfectly honest, as you play the game, you'll get a "feel" for what kinds of consists to give certain locomotives. I wouldn't think twice about giving an A4 Mallard that ten-BC4 consist, but I'd never even think about giving a 4-4-2 Atlantic anything remotely comparable to that—maybe five six-axle passenger cars if I'm feeling both generous and not-especially-in-a-hurry.

Of course, longer-distance hauls mean you have more flexibility—there's more time for the train to build up to and stay at its maximum speed, so you don't lose as much from the poor acceleration as you would lose on a line with frequent stops and starts.

As far as grade, unfortunately right now the only answer is "Just look at the track closely as you're building." I've heard people say things like "The steepest grade is equivalent to about 2.0%," but I don't know any data to back that up. As you're laying track, the "arrow" next to the confirm/reject buttons will give you an idea of how severe the grade is over that particular section of track, but it's still better safe than sorry—if grade is a concern, make sure you look closely as you're building.

[jballar]

thanks for your reply. I plan on spending a lot of time with this game, so I should get a better feel with this soon.

[toomuchbrew]

As was stated, there is no magic bullet due to variables. What would be interesting, and I may attempt this soon, is to experiment and draft a table using the following criteria.

Loco
Tractive effort
Power:weight
Avg pass wagon full weight
Avg cargo wag full weight
Grade <.5 (on avg)
Max wagons before 'intolerable performance impact'
Max wagons

I believe, and it's been years, rtycoon had something like this when choosing your consists

[Saint Landwalker]

Railroad Tycoon (I think all of them, but at least one of them) had for its locomotives a sort of grid that showed max speed according to number of cars (at an assumed weight per car) and grade of track (I think ranging from 0% to 6%). It was handy information to have.

One issue you might encounter with your planned experiment is defining "intolerable performance." Do you have any particular definitions in mind? "Seconds to reach max speed on flat ground," maybe? "Terminal velocity on incline of X severity"? Something else?

[toomuchbrew]

Originally posted by Landwalker04:

Railroad Tycoon (one of them, at least) had for its locomotives a sort of grid that showed max speed according to number of cars (at an assumed weight per car) and grade of track (I think ranging from 0% to 6%). It was handy information to have.

One issue you might encounter with your planned experiment is defining "intolerable performance." Do you have any particular definitions in mind? "Seconds to reach max speed on flat ground," maybe? "Terminal velocity on incline of X severity"? Something else?

Yeah, I was thinking about that. I think two stats actually

Less than say 75 or 80% of Max speed.

I think acceleration is not weighted as it should be. Most early to mid game routes are short to moderate length. I would say about 60 % slower acceleration than max would be an ok stat.

Those numbers are to just make a stand. People will have different thoughts on what is their lowest acceptable %. If only at the least, it's a baseline and then inferences and assumptions can be made with modicum of deviation.

[Saint Landwalker]

Originally posted by toomuchbrew:

Originally posted by Landwalker04:

Railroad Tycoon (one of them, at least) had for its locomotives a sort of grid that showed max speed according to number of cars (at an assumed weight per car) and grade of track (I think ranging from 0% to 6%). It was handy information to have.

One issue you might encounter with your planned experiment is defining "intolerable performance." Do you have any particular definitions in mind? "Seconds to reach max speed on flat ground," maybe? "Terminal velocity on incline of X severity"? Something else?

Yeah, I was thinking about that. I think two stats actually

Less than say 75 or 80% of Max speed.

I think acceleration is not weighted as it should be. Most early to mid game routes are short to moderate length. I would say about 60 % slower acceleration than max would be an ok stat.

Those numbers are to just make a stand. People will have different thoughts on what is their lowest acceptable %. If only at the least, it's a baseline and then inferences and assumptions can be made with modicum of deviation.

I think it would be a fascinating (and time consuming) project to run through every locomotive in the game, do test runs with them under a variety of load conditions, and chart it all out. Compare the time-to-max-speeds for a 2-8-0 Consolidation hauling 25, 50, 75, 100, 150 tons of cars. Maybe look at its terminal velocity for each of those conditions under each degree of slope. Of course then you're looking at something like 20 tests just for that one locomotive, extended across every locomotive in the game. And of course, the tonnage that would make sense to test isn't going to be the same across all of the locomotives—you wouldn't want to use the same load levels for the D 1/3 as you would use for the HHP 8.

Not that I'm going to do that, because I'm not at a point in life where I have that much masochism in my body. But maybe a project for somebody who isn't me. Or for me when I'm feeling particularly self-hateful.

[toomuchbrew]

Not that I'm going to do that, because I'm not at a point in life where I have that much masochism in my body. But maybe a project for somebody who isn't me. Or for me when I'm feeling particularly self-hateful.

I literally just spit out water all over my desk. Yeah, it's pretty lofty...

'i may attempt it' Just like when I said id go to the gym everyday, save money and be a better person. Lofty..

[Saint Landwalker]

Just a taste for those of you looking to commit a few hours of your life to being a stopwatch:

4-4-0 "The General"

Consist:   0 t — Time to Max Speed:  25 seconds
Consist:  20 t — Time to Max Speed:  39 seconds
Consist:  40 t — Time to Max Speed:  53 seconds
Consist:  60 t — Time to Max Speed:  69 seconds
Consist:  80 t — Time to Max Speed:  85 seconds
Consist: 100 t — Time to Max Speed: 103 seconds

I used 20-ton (metric, of course) increments for the consists, because the first-generation American passenger cars weigh 20 tons each, and most of the first-generation American freight cars weigh 10 tons each (stake cars only weigh 8 t). So that's your zero-grade breakdown for a 4-4-0 American pulling up to 5 passenger cars (total capacity 70) or 10 cargo wagons (or 12.5 stake cars).

[Paikia]

Nice, but if it was my project, I'd be digging in some physics or engineering websites, trying to learn what I need to find a set of formulas to do the job.

I'm guessing that once I have all the numbers (power, tractive effort, route length, average slope, and total weight), calculating an average speed is just a matter of knowing my sh!t.

Next up, I suppose, would be trying to figure out how payment is calculated in the game, in order to break it down to the most lucrative setup. Hmm... If only I didn't have to have a job... LOL

[Saint Landwalker]

Originally posted by Paikia:

Nice, but if it was my project, I'd be digging in some physics or engineering websites, trying to learn what I need to find a set of formulas to do the job.

Then you'd be making the very dangerous assumption that trains' acceleration in the game is based on actual physics formulas. Not saying it isn't. Just saying I wouldn't bet my own money that it is. =P

Originally posted by Paikia:

Next up, I suppose, would be trying to figure out how payment is calculated in the game, in order to break it down to the most lucrative setup.

That's actually pretty easy to answer: The most lucrative route for a single back-and-forth trip is a straight-line freight route that's fully-loaded going in both directions, and that has the highest possible theoretical maximum speed (even if it never reaches it). Revenue-per-agent-per-KM is based almost entirely on theoretical maximum speed of the vehicle, and although the game doesn't display it properly, freight is worth twice as much per KM as passengers are (presumably because of the assumption that most freight routes will only be fully loaded in one direction). Since distance is air-distance between the on-boarding station and the off-boarding station, you want the ground distance covered to be as close as possible to the air distance between the stations (hence, a dead-straight line).

[Paikia]

Originally posted by Landwalker04:

Then you'd be making the very dangerous assumption that trains' acceleration in the game is based on actual physics formulas. Not saying it isn't. Just saying I wouldn't bet my own money that it is. =P

I assume the game is probably using simplified formulas in order not to take too much of the system's resources, but it makes a lot of sense that those formulas would at least try to mimic real physics formulas, so we at least get some degree of realism. I'd love it if the physics were 100% accurate, but I know it's hardly the case in such games, and the severe stuttering and performance issues in Transport Fever keep reminding me why... :)

Originally posted by Landwalker04:

Originally posted by Paikia:

Next up, I suppose, would be trying to figure out how payment is calculated in the game, in order to break it down to the most lucrative setup.

That's actually pretty easy to answer: The most lucrative route for a single back-and-forth trip is a straight-line freight route that's fully-loaded going in both directions, and that has the highest possible theoretical maximum speed (even if it never reaches it). Revenue-per-agent-per-KM is based almost entirely on theoretical maximum speed of the vehicle, and although the game doesn't display it properly, freight is worth twice as much per KM as passengers are (presumably because of the assumption that most freight routes will only be fully loaded in one direction). Since distance is air-distance between the on-boarding station and the off-boarding station, you want the ground distance covered to be as close as possible to the air distance between the stations (hence, a dead-straight line).

Wait a second.... "Theoretical maximum speed"? Huh?
Are you telling me that the formula is based on a fixed statistic, the maximum speed limit of the loco, rather than the actual time it takes the train to deliver the cargo? Is that really what you meant? If so, it doesn't make any sense, since it means that all I have to do in order to maximize my profits, is take the loco with the highest maximum speed within my reach, slap on it as many wagons as it can carry, regardless of how long it takes it to get to the destination, and long live the king!

At this point, I really hope that I misunderstood you, or that you're wrong. :)

[Saint Landwalker]

Originally posted by Paikia:

Wait a second.... "Theoretical maximum speed"? Huh?
Are you telling me that the formula is based on a fixed statistic, the maximum speed limit of the loco, rather than the actual time it takes the train to deliver the cargo? Is that really what you meant?

You understood correctly, I'm afraid. Although it's "maximum speed limit of the vehicle" rather than "of the locomotive itself" (so if the limiting factor is the max speed of the wagons, that's what gets used).

By way of example, let's imagine you have a 4-4-2 Atlantic that's hauling a metric crapload (technical term) of wagons, so much so that it can't get over 50 km/h. But the Atlantic has a max speed of 100 km/h, and these hypothetical wagons have a max speed of over 100 km/h. Then somewhere else, you have a 2-6-0 Mogul hauling a relatively small number of wagons, so it can get up to its 75 km/h max speed and maintain that speed for most of its route.

The first train will make more per unit per km traveled than the second train will, even though the second train is traveling faster on average, because the first train has the higher theoretical max speed.

I've done a number of tests with this. It is so. Revenue per unit per km has nothing to do with how long it takes to cover the distance—it's a fixed number based on the vehicle's theoretical maximum speed.

(For another example, which I hit upon while doing some other experiments: If you have two truck stations that are 1 km apart, a line that runs on a straight road between them will make as much per unit delivered as a line that runs along a 10 km detour between the same two stations, as long as both lines are using trucks with the same max speed. Of course, the first line will make a lot more deliveries than the second, so it'll be more profitable per year*, but they will have the exact same revenue-per-unit-delivered. The fact that it takes the second line ten times as long to reach the station has no bearing on its per-unit-delivered revenue.)

* And this is really where the rub is. Do you use vehicles with a slower max speed, but which can reach and maintain that max speed more easily in order to make more deliveries per year at a lesser revenue-per-unit value, or do you use a vehicle with a higher max speed that can never reach it in order to make fewer deliveries at a higher revenue-per-unit value?

[Paikia]

Originally posted by Landwalker04:

You understood correctly, I'm afraid. Although it's "maximum speed limit of the vehicle" rather than "of the locomotive itself" (so if the limiting factor is the max speed of the wagons, that's what gets used).

Well, I'm not sure whether to call it "a design bug" or "a stupid decision made by lazy people in order to save themselves some work", but either way - this is, in my humble opinion, *not* how it should work. And unlike the need to simplify the physics formulas in order not to hurt performance too much, which is justified, I feel very strongly that simplifying this formula by using a known constant number (max vehicle speed) instead of a number calculated at that moment (the time it took the vehicle to reach the destination) simply hurts gameplay too much and is not worth whatever gains there are on the other side (performance, simpler coding or whatever).

Isn't everything too simplistic already?
Weight is assigned mostly to the wagons instead of to the cargo or the passengers, purchase and maintenance costs are divided wrongly between locos and wagons, fuel consumption and acceleration rates are only partly affected by weight of the train the conditions of the route, and those are just the ones I'm writing off the top of my head. Damn. That's just bad game design, in my opinion.

[Saint Landwalker]

On the one hand, as someone who was used to Railroad Tycoon's method of "Here's the most you can earn, and it depreciates over the time the longer it takes you to deliver it," I agree with the sentiment. On the other, I also can see the argument for "Well, you already get the benefit of faster delivery by translating that into delivering more stuff, so you don't need to also get paid for delivering that more stuff faster." Basically, on principle, I could go either way.

The part that gets me is the "theoretical max speed" part. It seems to be a matter of formulaic convenience that it's used the way it's used, but I'm also not a big fan of the "Slap twenty six-axle passenger cars on a 4-4-2 Atlantic and send it on its way, even if it takes all year to reach its destination" side effect of that approach. I feel like it marginalizes slower, high-power/TN locomotives that are supposed to specialize in heavy loads and harsh grades by punishing their per-unit-revenue relative to just using a locomotive that's wholly unsuited for the particular task just because it has a higher theoretical max speed.

[Paikia]

Originally posted by Landwalker04:

On the one hand, as someone who was used to Railroad Tycoon's method of "Here's the most you can earn, and it depreciates over the time the longer it takes you to deliver it," I agree with the sentiment. On the other, I also can see the argument for "Well, you already get the benefit of faster delivery by translating that into delivering more stuff, so you don't need to also get paid for delivering that more stuff faster." Basically, on principle, I could go either way.

The part that gets me is the "theoretical max speed" part. It seems to be a matter of formulaic convenience that it's used the way it's used, but I'm also not a big fan of the "Slap twenty six-axle passenger cars on a 4-4-2 Atlantic and send it on its way, even if it takes all year to reach its destination" side effect of that approach. I feel like it marginalizes slower, high-power/TN locomotives that are supposed to specialize in heavy loads and harsh grades by punishing their per-unit-revenue relative to just using a locomotive that's wholly unsuited for the particular task just because it has a higher theoretical max speed.

It's not the actual result that bothers me so much, as I know the game mechanics well enough to make money either way. What bothers me the most, is the simplistic approach to the formulas in the game. Why factor in time or speed at all, if you're going to use irrelevant or inaccurate data to measure it? The calculation might as well has been (income)=(price per unit)*(quantity)*(distance). If you factor-in speed - measure the speed correctly and accurately! Damn, those things piss me off!