Increase Fuel Display/Load Precision

[amp88]

Two suggestions here. Firstly to increase the precision of the fuel display (from the current 0.1% to 0.01%) and to increase the precision of the fuel load in the setup screen (from 1% to 0.1%).

On the fuel display we currently have a precision of 1 decimal place. This might at first seem to be precise enough, but if you do a large number of laps (~20+) significant differences can quickly creep in. For example, if your fuel display shows you're using 1.1% of fuel per lap this means you can be using anywhere from 1.05% to 1.149% (assuming the rounding is done in this 'traditional' Round half up fashion). Over a race of 50 laps, this means your fuel usage can be anywhere between 52.5% and 57.45%. Ok, so most people are going to say you should just play safe and put in a couple of percent more, but I don't think that's the way to solve the problem. Is it unrealistic to have fuel usage calculated accurately to 2 decimal places? With modern ECUs and electronics I'd imagine 2 decimal places accuracy is perfectly feasible, but I'm willing to take evidence to the contrary.

Secondly, in the garage we're able to set the fuel load at the start and pitstop to an accuracy of a whole percent. Depending on the car, 1% of the fuel load is between 0.2 litres (MRT5) and 1.25 litres (FO8). I don't think it's unrealistic in any sense to be able to determine fuel load in a more precise way than this. We've already seen in the latest test patches that Scawen is open to modifying the fuel system (by decreasing the minimum fuel loads to 1%), so how about making these changes too?

[ajp71]

I'd rather LFS scrapped the percentage system and just used proper units (or at least had an option to). 0.1% is very precise and TBH very few race cars actually have accurate fuel gauges, it's all done on guess work, they mostly have either just a warning light or nothing at all (and that includes GT cars). So having a reading accurate to 0.1% is really unrealistic if anything. Even F1 teams will not be able to predict fuel consumption that accurately. I'd suggest that the current system should be replaced with 1 litre increments.

[wark]

+1 for more precise fuel start level in setup screen (displayed in litres and percentages). It's not difficult to do in real life if you're starting with a completely empty tank.

Also, my real car seems to know exactly how much fuel it's using (live MPG readout and "miles left in tank" display), and of course the tank capacity is no mystery, so I don't really see why you shouldn't be able to see how much fuel you have in the tank, digitally, more precisely than 1 litre increments.

[dawesdust_12]

Two suggestions here. Firstly to increase the precision of the fuel display (from the current 0.1% to 0.01%) and to increase the precision of the fuel load in the setup screen (from 1% to 0.1%).

On the fuel display we currently have a precision of 1 decimal place. This might at first seem to be precise enough, but if you do a large number of laps (~20+) significant differences can quickly creep in. For example, if your fuel display shows you're using 1.1% of fuel per lap this means you can be using anywhere from 1.05% to 1.149% (assuming the rounding is done in this 'traditional' Round half up fashion). Over a race of 50 laps, this means your fuel usage can be anywhere between 52.5% and 57.45%. Ok, so most people are going to say you should just play safe and put in a couple of percent more, but I don't think that's the way to solve the problem. Is it unrealistic to have fuel usage calculated accurately to 2 decimal places? With modern ECUs and electronics I'd imagine 2 decimal places accuracy is perfectly feasible, but I'm willing to take evidence to the contrary.

Secondly, in the garage we're able to set the fuel load at the start and pitstop to an accuracy of a whole percent. Depending on the car, 1% of the fuel load is between 0.2 litres (MRT5) and 1.25 litres (FO8). I don't think it's unrealistic in any sense to be able to determine fuel load in a more precise way than this. We've already seen in the latest test patches that Scawen is open to modifying the fuel system (by decreasing the minimum fuel loads to 1%), so how about making these changes too?

If you have an insim appy, you can get the fuel load to 2 decimal places.

[deggis]

I've always wondered why it says "fuel load at pit stop: xx %" while it actually means "FILL tank to %". If Scawen is about making everything as newbie-friendly as possible, this is one of the things that definately is confusing for the first times you think pit strategy. Plus for litre units too, having percentages is just weird especially when all the other settings are in proper units.

[wark]

I've always wondered why it says "fuel load at pit stop: xx %" while it actually means "FILL tank to %". If Scawen is about making everything as newbie-friendly as possible, this is one of the things that definately is confusing for the first times you think pit strategy. Plus for litre units too, having percentages is just weird especially when all the other settings are in proper units.

Agreed--that is ambiguous.

[dawesdust_12]

*goes to edit my Language file...*

[danowat]

+1 for more precise fuel start level in setup screen (displayed in litres and percentages). It's not difficult to do in real life if you're starting with a completely empty tank.

Also, my real car seems to know exactly how much fuel it's using (live MPG readout and "miles left in tank" display), and of course the tank capacity is no mystery, so I don't really see why you shouldn't be able to see how much fuel you have in the tank, digitally, more precisely than 1 litre increments.

+1 from me, % is so not realistic, should be gallons :D

[ajp71]

+1 for more precise fuel start level in setup screen (displayed in litres and percentages). It's not difficult to do in real life if you're starting with a completely empty tank.


But there's never any need to fill it that accurately because fuel consumption cannot be predicted that accurately.


Also, my real car seems to know exactly how much fuel it's using (live MPG readout and "miles left in tank" display), and of course the tank capacity is no mystery, so I don't really see why you shouldn't be able to see how much fuel you have in the tank, digitally, more precisely than 1 litre increments.

The MPG/range readout in your car is a vague approximation fitted to a lot of production cars, it will be roughly right in a car driving at consistent speeds on a motorway, under extreme loads and changes of fuel consumption though (ie. when racing) it will be useless. Most racing cars do not have a conventional level fitted because the tank is filled with fuel tank foam.

[atledreier]

-1 for me too. I'd say we should get way less precision.

[three_jump]

+1 for using proper units.

+1 for changing the fuel on pitstop to add amount of fuel, not to fill to amount wanted (as that is far away from realism)

[tristancliffe]

We are doing the fuel in our car in kilograms, because that's what actually matters for performance, and we need to use fuel as ballast to reach the minimum weight. But Percentages don't bother me in LFS as they're just numbers. Who cares whether it's 15lites or 15%, all you are looking at is the 15 bit.

As for accuracy, the fuel in our car is put in with an accuracy of +-2 litres (ish, depending on how much we actually have in the tank!), so LFS's 1% is more than adequate.

[wark]

But there's never any need to fill it that accurately because fuel consumption cannot be predicted that accurately.

But the same could be said of tyre pressures and temp & load "prediction." If you're going to simulate filling a tank with liquid, why simulate a "lack of need" for accuracy? I think it would be unrealistic if I couldn't at least fill it up to the nearest cl.

The MPG/range readout in your car is a vague approximation fitted to a lot of production cars, it will be roughly right in a car driving at consistent speeds on a motorway, under extreme loads and changes of fuel consumption though (ie. when racing) it will be useless. Most racing cars do not have a conventional level fitted because the tank is filled with fuel tank foam.

It's live (updating several times a second) and fully dependent on throttle position/revs/engine load.

Nevertheless, I'm not going to argue for this cause, as I don't honestly care about anything above and beyond the analogue fuel dial. I just don't think the feature necessarily needs to be removed.

[wark]

We are doing the fuel in our car in kilograms, because that's what actually matters for performance, and we need to use fuel as ballast to reach the minimum weight. But Percentages don't bother me in LFS as they're just numbers. Who cares whether it's 15lites or 15%, all you are looking at is the 15 bit.

Hmm. Now (or at least when it becomes adjustable) we'll need the air temperature displayed next to the fuel mass/volume. :scratchch

[ajp71]

But the same could be said of tyre pressures and temp & load "prediction." If you're going to simulate filling a tank with liquid, why simulate a "lack of need" for accuracy? I think it would be unrealistic if I couldn't at least fill it up to the nearest cl.


IRL tire pressures are adjusted to the nearest half pound in club racing so to the nearest tenth is really believable for top level teams. Filling fuel to the nearest cl is ridiculous, nobody ever fills a car up that accurately because even in F1 fuel consumption is still an educated guess.


It's live (updating several times a second) and fully dependent on throttle position/revs/engine load.


Yes it takes an approximation of fuel consumption based on what the ECU mapping is telling the engine to do, of course the ECU cannot control exactly how much fuel is being consumed so actually a live economy gauge is a nice novelty but actually completely inaccurate. A long term average based on the time taken to get through a substantial chunk of your tank is far more useful.

[wark]

IRL tire pressures are adjusted to the nearest half pound in club racing so to the nearest tenth is really believable for top level teams. Filling fuel to the nearest cl is ridiculous, nobody ever fills a car up that accurately because even in F1 fuel consumption is still an educated guess.

Well, my point is, if you're going to fill it up from scratch, why not put exactly as much as you want in? why more or less? units of measurement are completely arbitrary. IRL you have a sort of almost infinite "slider." That's impossible to simulate, but you can get close with increments of very, very small units of measurement. Just because the amount is dealt with in units doesn't mean you're "measuring it," per se. IRL you can give your car a "splash" more fuel. That's the kind of reckless fine tuning (http://dictionary.reference.com/browse/oxymoron) I desire.

[ajp71]

^^ In the case that the amount of fuel added is precise then why not make the amount left in the tank displayed as acurate as a good RL prediction, ie. nearest 5 litres ;)

[Gunn]

Who cares whether it's 15 litres or 15%, all you are looking at is the 15 bit.Yep.

[wark]

^^ In the case that the amount of fuel added is precise then why not make the amount left in the tank displayed as acurate as a good RL prediction, ie. nearest 5 litres ;)

:Handshake

[amp88]

Just reading an Ayrton Senna biography and there's a small, interesting quote on the teams calculating fuel consumption. In 1988 the McLaren Honda test driver Geoff Lees said that their fuel consumption could be calculated to within 200 metres over a Grand Prix distance. That's 200 metres out of over 300 kilometres (0.067%).

[bbman]

In that case, LfS is bang on...

[frokki]

This should also depend on the car.

[amp88]

In that case, LfS is bang on...

It's an order of magniture out. (0.1 versus 0.06)

[Bob Smith]

I'd suggest that the current system should be replaced with 1 litre increments.
Agreed. That's what I use in my analyser, seems much more natural.

Are fuel cans measured by volume or the percentage of the tank they are to be poured into?

[bbman]

It's an order of magniture out. (0.1 versus 0.06)

No... If the tolerance limit is 0.06, the nearest you can really guess is 0.1 percent... ;)

[Forbin]

....

Also, my real car seems to know exactly how much fuel it's using (live MPG readout and "miles left in tank" display), and of course the tank capacity is no mystery, so I don't really see why you shouldn't be able to see how much fuel you have in the tank, digitally, more precisely than 1 litre increments.
I always thought the MPG gauge was essentially just a manifold pressure gauge that also took into account the car's speed.

[ajp71]

Just reading an Ayrton Senna biography and there's a small, interesting quote on the teams calculating fuel consumption. In 1988 the McLaren Honda test driver Geoff Lees said that their fuel consumption could be calculated to within 200 metres over a Grand Prix distance. That's 200 metres out of over 300 kilometres (0.067%).

Eddie Irvine also said he lapped round Silverstone in a Formula Ford faster than Jim Clark in a Lotus 49 (apart from they raced on completely different layouts there was no way he could have been anywhere near the 49, especially not with Clark driving it). My point is I don't believe one quote by a test driver over everything else I've seen and heard over this issue. Maybe in testing it was possible to predict fairly accurately but in a race you ultimately have far more potentially unexpected events, problems or changes of conditions that all change your driving style and fuel consumption slightly.

[Eldanor]

Who cares whether it's 15lites or 15%, all you are looking at is the 15 bit.

I think it's more natural, using standard units like liters or kilograms than using a relative unit like % of the tank capacity.

It's a minor detail and I can live with it, but it makes sense.