I am running LFS on the background.
I have an XRT running on the autocross loop. (the circly thing).
When I accelerate to 30km/h put it in 5th gear and wait, the car goes to 1/3-1/2 of its throttle to prevent stalling.

However, I am running this for 18 minutes now, and I haven't even lost a tenth of a percent of fuel.

When driving an other XRT with the same throttle, but actually physically doing it yourself (instead of what the car automatically does), you do lose fuel.

How and why is this?

It may be possible to explain it better with a replay, but I'm making one now, and I will run it for another few hours

sounds like a silly bug to me, if you loaded the replay that make it easy to find out and problem.

The car stopped in an hour, and I got an OOS-error (possible because of loading a second LFS at the time)

running for two hours now

Submit it as a bug.

Not exactly a game changing one, although it could be used to get to a finish line if you had a big enough lead, in a situation where normally you'd be screwed for not putting enough fuel in?

I just tried tweak.
I set the fuel tank to 0.1L.
When doing the same, you do see the fuel percentage decreasing..

E: Ah, it just takes very long

Using tweak: with a 6L tank it'll take 80-160 seconds to let the tank decrease a tenth of a percent
You would expect a 65L tank to take 11 times as much (~15-30 minutes) to decrease a tenth of a percent then.
But it seems that it takes far more time than that.


HALP

good fuel consumption apparently

Just correct me if I'm wrong here, But starting a car and letting it run wastes about a gallon every 2 hours (MY car) That's 0.016666666666666666666666666666667% Every minute.

If the game follows rounded math ( I'm pretty sure it does ) You would never loose fuel.

0.1\600min = 1.6666666666666666666666666666667e-4

Yeah, You would loose about... what.. 1 gallon every 5 days or something out of a guess?

But it's a racing simulator, I didn't know someone cared about full after 18 minutes of standing still, ARE YOU ONE TO WANT TO DELAY, SCIROCCO, FOR THIS?

Just had to.

I'm sorry, but your math is kinda off here. If a car needs 1 US gal per 2 hrs while idling, it will empty 60 litre tank in about 32 hours.


Now if you calculate how much fuel per second such car needs, you'll find it's something about 0.00053 l/sec. LFS recalculates physics every 1/100th second, so upon every recalculation you loose about 5.2 x 10^-6 litres of fuel. Standard 32-bit float in C++ can work with numbers as small as 10^-38, so maths precision is not the issue here.

Anyway, BlakjeKaas reported that to prevent stalling, LFS applies about 1/3 throttle and engine near the stall speed and such throttle position would make fuel consumption WAY higher compared that at idle. There is probably a bug at the function that calculates fuel consumption. It would also explain the OOS error.

I retried the experiment without opening a 2nd LFS while busy with the experiment.

I didn't get a OOS error this time.

The drawback is that you can't see the fuel percentage in replays =(

Since I have nothing better to do, I'm going to do some experiments of my own. I assume you had autoclutch and manual shifting enabled when you did this, right?

autoclutch disabled, else it would just clutch, stop and idle (full clutch)
manual clutch enabled

Quote from BlakjeKaas :

autoclutch disabled

Yeah, I realized that just after hitting Submit Reply

ah, ok.

FZR running for 100 minutes on 39km/h

99.9% in tank

I just spent last hour or so getting some raw data out of OutGauge to check what's going and it seems to me that the fuel consumption is just ridiculously low. I don't have the exact figures in l/100km or MPG but I'm working on it

Quote from MadCatX :

I just spent last hour or so getting some raw data out of OutGauge to check what's going and it seems to me that the fuel consumption is just ridiculously low. I don't have the exact figures in l/100km or MPG but I'm working on it

Nice, I'll just leave LFS running in the meantime.

It seems to me that most of it seems correct (idling costs fuel, antistall costs fuel), just the amounts of it and the calculation thereof seems off

Fine, I think I've got it. I used XFG for testing as it's engine is closest to those I have some experience with. When rolling flatout on 2nd - 4th gear it takes about 25 l/100km, when running on 5th gear at the end of the Blackwood straight consumption drops to some 14 l/100km which doesn't sound all that wrong to me.
However, when the engine is at idle or almost stalled, fuel consumption drops to zero, no matter if I measure consumption/100 km or consumption/hour. That's just plain wrong and it explains the problem you've discovered.

Below is the code I used for this, perhaps someone could check if it's any correct...

Quote from MadCatX :

I'm sorry, but your math is kinda off here. If a car needs 1 US gal per 2 hrs while idling, it will empty 60 litre tank in about 32 hours.

Eeeeyy.. Don't shoot the ... Me.. Uneducated game.. not developer person!

I was talking in LFS dear. No way in hell my car is idling for 5 days unless my car is using reserves off Washington, And that's just fishy. ( As if our government wasn't.. )

2:45 hours, and still 0.0% out of the tank.

It probably rounds it off.

In LFSTweak with a small tank it's a big enough factor to be rounded off upwards (I suppose)
In LFS itself it's a small enough factor to be rounded off downwards (to 0, that is)

Quote from TehPaws3D :

Eeeeyy.. Don't shoot the ... Me.. Uneducated game.. not developer person!

I was talking in LFS dear. No way in hell my car is idling for 5 days unless my car is using reserves off Washington, And that's just fishy. ( As if our government wasn't.. )

I didn't mean to offend you or anything, I just wanted to point out that there really is a problem with fuel consumption in LFS which can't be dismissed just as rounding or precision error. That XRT would clearly beat Prius in economy tests and I'd really want to have a 2.0l turbocharged engine able to pull that off...

Quote from BlakjeKaas :

2:45 hours, and still 0.0% out of the tank.

It probably rounds it off.

In LFSTweak with a small tank it's a big enough factor to be rounded off upwards (I suppose)
In LFS itself it's a small enough factor to be rounded off downwards (to 0, that is)

I don't think so, LFS measures level of fuel in tank pretty precisely. Look at the screenshots I attached (not exactly easy to read, I know). Those numbers you can see in the lower right corner are 1) amount of fuel in tank as LFS stores it 2) litres left in tank 3) Self-explanatory I don't think it's possible for an engine to run at 600 rpm, provide enough power to keep the car moving at 5th gear and use about 1l/100km. The second screenie shows consumption per hour and for all I know an engine like this should use about 1l/hr at idle.

so far it's doing 0L/100km (or close to that)

1L an hour should be showing up in percentages then, but after 3(!) hours, even with accelerating in between (I alt+tabbed and my wheel was in the incorrect position =() the consumption is near to nothing.
after 3 hours I had to accelerate, it's now at 99.6

so it's doing 0.20L/120km atm (with acceleration)

3 hours later, it's still at 99.6

I would kill for such a car.

I don't think it would take much to keep a car running, If the car isn't moving ( Please don't get me wrong, I'm just guessing and letting you guys answer ) Then shouldn't it really not take that much fuel? It's just pistons moving to keep the engine rev'd. Sooner or later of they where pushing enough power at one point ( Assuming you went to fifth gear correctly ) Then the engine could just use the kinetic force of the pistons to thrust up and down, And only rev back ( Stall limiter ) To keep the RPM's atleast somewhere up, And at that point if you are just parked. At around 1000 or 2000 RPM the car is doing nothing, I believe it would use more oil than gas, The way I see it using that little fuel could be possible ( Okay.. probably not, but something close to it. )

Just to test, I've let my car run all night ( Mercedes ĸompressor c230 ) No big changes in fuel.

Quote from TehPaws3D :

I don't think it would take much to keep a car running, If the car isn't moving ( Please don't get me wrong, I'm just guessing and letting you guys answer ) Then shouldn't it really not take that much fuel? It's just pistons moving to keep the engine rev'd. Sooner or later of they where pushing enough power at one point ( Assuming you went to fifth gear correctly ) Then the engine could just use the kinetic force of the pistons to thrust up and down, And only rev back ( Stall limiter ) To keep the RPM's atleast somewhere up, And at that point if you are just parked. At around 1000 or 2000 RPM the car is doing nothing, I believe it would use more oil than gas, The way I see it using that little fuel could be possible ( Okay.. probably not, but something close to it. )

I can see what you're getting at, but I think it's a bit more complex than that. You need energy to suck air into the piston, compress it and then blow out the exhaust gases. There is also some friction between the piston the cylinder.

I know that a 1.2 HTP engine (made by VW) needs about 0.8 l/hr on idle. If I turn on lights, AC an whatnot, it climbs to some 1.2 l/hr. Since onboard computers often talk claptrap when it comes to some exact numbers I used an awful lot of approximations and tried to calculate the idle consumption myself. I got something about 1.6l/hr, so I guess the computer was right this time. If you check the screenshots I posted yesterday, you'll see that LFS is way off when it comes to fuel consumption in low revs.

I can of course be totally wrong, I'm no car engineer

Quote :

Just to test, I've let my car run all night ( Mercedes ĸompressor c230 ) No big changes in fuel.

You really did that? I guess you might want to stay hidden under a table for a few days in case someone from Greenpeace shows up...

I once fell asleep in a started car and when I woke up (after an hour or so) I saw that the fuel needle was a bit lower...

Wouldn't the suction come from the piston lowering? It should suck itself back up right?

Yeah, as the piston goes down, it sucks air into the cylinder. Gas always tries to maximize it's volume, the more space it gets, the more space it takes up, so suction isn't much of a problem.
What's most energy-demanding in the compression. Piston won't move back up by itself 'cause the intake valve is closed.

Ever tried to fill a hypo with air, seal it and then push on the piston? It won't move quite that easily, will it? An engine has to face a similar task and it has to compress the gas to some 1/9th of the original volume... not an easy job.

My point is that engine really needs to produce some energy to keep running and that energy is not as low as it might seem.

yet the inertia of the other 3 cylinders in their cycle will keep things moving.