automationgame.com

I'm not sure whether this should rather be called a bug report than a suggestion, but since I guess technicly the program is fine, I call it a suggestion.

When using an engine that pushes it's rev limit while having low camprofile one gets a horrible economy at high revs. So far so good. But obviously the engine isn't really meant to be driven in high rev's...
If one now sets gear spacing so that at 80 km/h you'd be at max. revs in first gear the economy jumps from like 10l/100km to 90l/100km and such.

I suggest using the highest gear that can still provide enough power to compensate for drag at 80 km/h for the economy calculation. Alternatively calculate for all gears and pick lowest value.

If it is not bugged it should do exactly what you are saying - so if it behaves like this it indeed is a bug we need to look into.

It does. Just checked it out with standalone B150502 and reliably reproduceable.

How many gears, how heavy is the car, and how much power does the engine have?

70.9 kW
3 Gears RWD
1399.3 kg

Here's the lua's:


just lower spacing one tick (to 20 I think) and watch economy jump from 14 to 90.

But I think it should be easily reproduceable with more or less any car and engine, since I've seen this many times before.

I have also seen this many times before, trying to find the optimal fuel economy settings.

For what it's worth, I've noticed a friend of mine skip straight from first to third when accelerating - I wonder if there's some assumptions in the fuel economy calculation code about gear-shift behavior that's causing the issue.

I've been looking what fuel consumption I get for what gear spacings ever since reporting this. Now I hope this can be helpful in tracking down the problem:

I've managed to break all (my) records in fuel consuption with this car:


I'd not even call this a car anymore, this is a fuel eliminating machine! It eliminates an amazing 1400+ l/100km (some 0.15 mpg), meaning it would need at least two refuels in order to complete Green Hell...

Drag at 80 km/h is something like 10kW (it's hard to exactly read out the graph). In both gears it would have more than enough power to make those 10 kW. So it should drive in first with something like let's say 1000g/kWh. For 100km it needs to drive for 1.25 hours at 80km/h. Now I simply multiply all that up to 12.5kg/100km. Specific weight of fuel is araound 0.75kg/l, so I divide by that, resulting in 16.667 l/100km. Not 1400.

So the game calculates 84 times as much as I do, so it can't possibly be because I misread drag or fuel consumption (I would have to have misread both by a factor of 9+).

Now I wonder what is the difference between my calculation and the one in the game, but clearly there is some difference.

Edit: Maybe it's driving in 2nd, which would explain some of this and maybe drag is considerably more... ...if I'm not mistaken I could get the drag coefficient out of the car's lua and try to perform the whole calculation...

In the final tab, under testtrack, it should say the Aerodynamic Efficiency of your car.

That value is frontal area * drag coefficient.

Also, IIRC the fuel economy is not just the fuel consumption at 80.

It actually calculates the values at 30/50/70/90 and 120 and then does some mathematical operations to calculate the value shown.

Although I don't really know the exact specifics beyond this, you will probably have to ask Killrob for this.

Edit: Checked you car and I think I found the problem.

From the lua file, the fuel consumption at 30kph is infinite for some reason, maybe due to not having enough power at that RPM. Because of this the maths behind the whole thing seem to break down.

This stays true until the top speed is lowered low enough for the fuel consumption to end up being a more sensible value of 30 something L/100km.

At that point if I check the lua file it shows a value for the fuel consumption at 30kph.

The fuel eco calcs are indeed bugged in the version you are playing, we just fixed them in the closed beta of the game so look forward to more realistic, less weird results