Do multiple header trains make pulling rolling stock faster?

[bigyihsuan]

I haven't been able to play around with trains yet, but do multiple locomotives at the front of a train make it go faster with more cars?

This is because it would be more logical that more locomotives = more horsepower = more wagons pulled at max speed.

[hitzu]

If they all are facing forward then they do. If two locomotives are facing opposite directions then "tail" loco does not help pulling rolling stock, but on a contrary acts like an additional weight which have to be pulled. In the other hand this train doesn't need loops.

[Xterminator]

If they all are facing forward then they do. If two locomotives are facing opposite directions then "tail" loco does not help pulling rolling stock, but on a contrary acts like an additional weight which have to be pulled. In the other hand this train doesn't need loops.

That's really good to know. I didn't actually know if more locomotives in the front, facing the same actually increased speed. Cool to know they do.
Here's another question. So would more locomotives in the front increase even the max speed of a train with 1 cargo wagon? Or does it really only do anything with heavier loads using like 4 wagons?

[hitzu]

So would more locomotives in the front increase even the max speed of a train with 1 cargo wagon? Or does it really only do anything with heavier loads using like 4 wagons?

I tried 2 locos facing forward with 1 wagon and can speed up to ~255 km/h. That is slightly slower than bare loco (259,9 km/h).

[iklarazu]

Two (or more) trains won't make the train go faster. They're still limited by the individual train.

Two trains facing the same way will however decrease the time it takes to get to full speed.

[sillyfly]

Two (or more) trains won't make the train go faster. They're still limited by the individual train.

Two trains facing the same way will however decrease the time it takes to get to full speed.

This may be true in real life, but in Factorio the maximum speed is a factor of the total wagons and the total locomotives. So a train with 2 locomotives and 4 wagons will have a faster maximum speed than one with 1 locomotive and 4 wagons.

[Choumiko]

This may be true in real life, but in Factorio the maximum speed is a factor of the total wagons and the total locomotives. So a train with 2 locomotives and 4 wagons will have a faster maximum speed than one with 1 locomotive and 4 wagons.

Isn't that true in real life too?

Code: Select all

speed(1Loco) == speed(2Loco) > speed(2Loco+4Wagons) > speed(1Loco+4Wagons)

seems like a comparison of max speed that should be true in reallife?

[psihius]

Trains max speed is equal no matter the locomotive count.

The train acceleration, on the other hand, is benifiting from more locomotives when there are wagons to pull. So a train of 4 wagons with 2 locomotives accelerates much faster than with 1 locomotive. Easy to check

[sillyfly]

Trains max speed is equal no matter the locomotive count.

The train acceleration, on the other hand, is benifiting from more locomotives when there are wagons to pull. So a train of 4 wagons with 2 locomotives accelerates much faster than with 1 locomotive. Easy to check

Indeed, it is very easy to check! So I just did - 1 locomotive, 4 wagons - max speed approx. 179 km/h. 2 locomotives, 4 wagons - max speed approx. 259 km/h.

[sillyfly]

Developer note for future readers: These values are for version 0.12 and lower.

So, I went and did some more measurements with 1 and 2 locomotives, and it seems the max speed is directly tied to the locomotive / cargo-wagon ratio:

Code: Select all

Locomotives / Wagons = Max Speed
1    /    0    =    259.2 
1    /  0.5    =    259.2 
1    /    1    =    259.2 
1    /  1.5    =    259.2 
1    /    2    =    259.2 
1    /  2.5    =    239.4
1    /    3    =    215.4
1    /  3.5    =    195.8
1    /    4    =    179.5
1    /  4.5    =    165.7
1    /    5    =    153.8
1    /    6    =    134.6
1    /    7    =    119.6

Where fractional numbers refer to tests with 2 locomotives, so - 1/3.5 means 2 loco. & 7 wagons, etc. whole-number tests up to 1/4 where done with both 1 and 2 locomotives.
You can see that up to 2 wagons per locomotive the maximum speed doesn't drop, but any ratio higher than that - the maximum speed gets lower.

[cube]

This part of Factorio is kind of realistic: each wagon adds friction, each locomotive adds power and friction, acceleration eats power. Easy. So max speed really depends on locomotive count / wagon count and so does acceleration.

[DerivePi]

This part of Factorio is kind of realistic

Let's ruin this concept with realism -
First, the mass is not constant. A loaded wagon should have a significantly higher mass than unloaded. Further, the mass changes as the locomotive consumes fuel (like rocket science). Mass is important for two things: - more mass means less acceleration where F is constant ( F = m a - of course F isn't constant either. More on that later). - more mass means a higher maximum tractive force. ie, the train can impart a higher force onto the rails without slipping.
m = nl x ml + nw x mw + mfuel + mcargo (nl = number of locomotives, ml = mass of locomotives, nw, mw = number and mass of wagons)
mfuel = mfuel initial - mfuel consumed
a = dx2 / d2t = F / m
F = Fengine - F windresist - F wheelresist; and Fengine <= Ftractive otherwise the wheels slip

Second, air friction is not constant. It is closely related to the velocity of the train. As the velocity increases, so does the air friction. The equation for air friction force should be something like:
F windresist = - (fb + nl x fl + nw x fw) x (dx/dt)

m = number of locomotives, n = number of wagons, fb, fl, and fw are the wind friction coefficient for the front and back face friction, locomotive side friction and wagon side friction, and dx/dt is the velocity of the train

Third, there is a friction force between the rails and wheels, F wheelresist. As mass increases, the wheels deform more and are not as efficient in rolling (related to mass and turning/velocity). Also, when turning, there is an additional resisting side shear/friction that also reduces the rolling efficiency of the wheels (related to change in direction over time, dphi/dt and mass).

Etc... blah, blah, blah. Or, as we learned in high school - "assume air friction is negligible"

Any consideration that the above should be taken seriously will be received with laughter.

[Choumiko]

You do justice to your avatar

First, the mass is not constant. A loaded wagon should have a significantly higher mass than unloaded.

That one would actually be nice i think

[ssilk]

You do justice to your avatar

First, the mass is not constant. A loaded wagon should have a significantly higher mass than unloaded.

That one would actually be nice i think

+1

And a bit slowering in curves.

[Nemoricus]

If the trains slow down as they go from empty to fully loaded, should the player slow down as they have more items in their inventory? Because if you do one, for consistency it should apply to all mobile entities that have an inventory. Cars, tanks, players...

[ssilk]

Of course. And because mass in Factorio isn't a problem (you have the stacks, which do compression based on some quite advanced hyperspace technique), this effect should work like so:
Any inventory has a "fullness" between 0 and 1. It depends on, if a stack is used. This factor is multiplied by some (very) small number, depending on the entity and subtracted from the acceleration-factor.

The result should be in my eyes, that a train with 3 full wagons agains one with 3 empty is slightly slower accelerated. I would say if both start at the same time, the full is after 100 tiles one train-length behind.

So this effect is not really strong, but remarkable.

[Kevin94]

Not really strong doesnt fit the numbers you proposed.
A wagon is about 6 tiles long, a train with 3 wagons and a loco is then about 25 tiles long (to lazy for exact counting).
When the the empty train traveled 100 tiles than the slower traveled 75 tiles in the same time. As acceleration is proportional to distance (x = a/2 * t^2) a full loaded train would have an about 1 quarter lower acceleration than an empty.
This implies a one third longer timer to reach full speed or a 7/9 (77%) longer distance traveled before reaching fullspeed.

[Nemoricus]

Of course. And because mass in Factorio isn't a problem (you have the stacks, which do compression based on some quite advanced hyperspace technique), this effect should work like so:
Any inventory has a "fullness" between 0 and 1. It depends on, if a stack is used. This factor is multiplied by some (very) small number, depending on the entity and subtracted from the acceleration-factor.

The result should be in my eyes, that a train with 3 full wagons agains one with 3 empty is slightly slower accelerated. I would say if both start at the same time, the full is after 100 tiles one train-length behind.

So this effect is not really strong, but remarkable.

If the effect is small, why have it? It seems like a complication that doesn't add much if that's the case. It's realistic, sure, but if the player has to do side by side comparison tests to even notice it, what value does it add?

Now, it is the little touches that make a game rich and immersive, true...

[ssilk]

Not really strong doesnt fit the numbers you proposed.
A wagon is about 6 tiles long, a train with 3 wagons and a loco is then about 25 tiles long (to lazy for exact counting).
When the the empty train traveled 100 tiles than the slower traveled 75 tiles in the same time. As acceleration is proportional to distance (x = a/2 * t^2) a full loaded train would have an about 1 quarter lower acceleration than an empty.
This implies a one third longer timer to reach full speed or a 7/9 (77%) longer distance traveled before reaching fullspeed.

Cool calculations. That was about my estimation. Perhaps a little bit less (50-70% more distance, but that depends on balancing). It should be a difference between a loco filled with one stack of fuel or three (the "fullness" is then only 1/3 instead of 1, and the train alone will accelerate visibly faster).


@Nemoricus: the little touches, yes! The effect on belts in the corners for example is also "not strong" but it becomes a big tread, and you can find tons of posts about that.

Edit: my thoughts about this (off topic): I think the basic idea is, that a full inventory is 1, an empty is 0 and any used stack is part of the inventory. That way a "mass concept" could be introduced, which fits in my opinion nice into this sometimes strange stack behavior (5 locomotives per stack), cause with hyperspace everything is possible, but a stack is it's realistic counterpart and that has some weight, if used.