if development is slow to the point where no new people are buying, or people are leaving the game(which ultimately hurts the overal game). They won't be able to live off that lowered income.
They are not sharing their selling. without all of you who bought the game, it would not be possible

i said the request is valid, didn't say it should jump to the top of the priority. At this point asking for anything reasonable, is valid, where it lands in terms of priority is another discussion.

No one should have to master the game to give criticism of ask for new content. At the end of the day people play this game to have FUN. if you feel that a new track would increase the amount of fun you would have, its a valid request.

Last edited by lalathegreat, Thu 18 Oct 2007, 06:45.

Quote from Shotglass :

hes wrong though ... only under vista youll find the video memory mapped into a uniform address space on all other systems the vram has nothing to do with the 4 gigs of address space and isaddressed differently if im not completely mistaken

the implications have been discussed on anandtech at length

i am pretty sure i am correct the GPU is memory mapped using the CPu memory space thus limiting it

15% of that performance prob comes from the extra memory addresses. Some people might not know, but 4Gb of address space does not mean you can use 4gb of ram. some of that goes to that GPU so if you have a 8800GTX with 768 ram and 4gbs of ram installed only 3.2Gb is going to show up in windows, even though u have 4 installed. also address space is also taking up for pci cards as well eg sound card.

where are the boobs

of your latest engine Sim, which part is more computationally intensive. Also did you design the sim with a larger respect for it generating audio.

Quote :

i think the harsh truth is that realistic real time engine simulation is not gonna happen anytime soon and quite possibly not before lfs exceeds its lifespan

I disagree, if we can have real time tire simulation that meets our standards, then there is no reason for it not to be possible.

Quote from Woz :

Just shows what happens to turbo cars though when boost is gone. A 1.6 Diesel can beat a 400bhp car if the monster car is a turbo and the turbo is not spooled.

To bad that the situation needed for that to happen is unlikely to ever happen in a race

well he took it literally, when i say real time engine simulation i mean best possible simulation that can be done in real time. not everything simulated

Quote from tristancliffe :

I'd love real time engine simulation. The fact that decent non-real time engine simulation is only just possible outside of big car companies obviously isn't the point. Anyone here used WAVE? It takes quite a long time to do very basic throttle-power plots, and that's on a single cylinder normally aspirated engine!

So the question is, how close do you go? Closer than now, sure, but not fully accurate which isn't feasible.

I'm sure the engines will get an update at some point in terms of how they produce their power. Coughing and flames, or even backfires, are probably an S3 thing.

But the aero work is weaker than the engine simulation overall, so I'd say that takes priority at the moment. If the basic turbo model could be improved at around the same time I think we'd be a bit happier still.

I'm sure Scawen has already mentally priorities the various options, and he'll get on with it soon. He's very quiet at the moment, so I think he's probably doing AI at the moment quite heavily. Once that's done who knows what he'll do next.

I think its can be done really good in real time. I would be interested in what you think is needed and why you think it can't be done in real time, decently at least

Quote from z3r0c00l :

I look forward to driving cars in LFS with engines that have personality. The amount of complexity of engine modelling shown here, which I believe is possible, given the quality of the tyre model, will be great.

At the moment I find LFS's engines a bit too smooth, and linear, and generally artificial. I would imagine with the complexity of the engine sound simulation, and accurate modelling of the engine itself, we will find all sort of behaviours emerge that are hard coded in other games, like a nice burble on the over-run of cars with carbs, and if they can be bothered to model exhaust temperatures and such, backfires and other immersive behaviours, like a little black smoke from an engine that is too hot (don't drive a suzuki bandit 600 hard in a country which is at 40 degrees c! the radiator is far too small!).

I can't wait for the first race I see where the guy in first place blows a turbo flat shifting infront, causing the guy in second to spin out from all the oil deposited on the track, leaving me, tactically in third place, to take all the glory, with non of the effort.... brilliant

ive always been pro real time engine sim. but some aren't prospect of having 20 cars coming into t1 all with different sounds and randomly backfiring would warm my heart. problem with lfs is that most people will disagree because they rather something else take higher priority. eg someone wants better aero so they go -1

see your assuming that the turbo can't make the same manifold pressure with the plate on. all you have to do is keep the waste gate closed longer until the desired boost is reached what the plate does it make the turbo have to spin faster to create that kind of pressure

Everyone seems to drive slow cars.
Regardless i have a 96 civic hatch chassis with 2.0 liters 250Whp Na

Quote from Ball Bearing Turbo :

You guys are missing the point:

Explain to me how a GT1 FXR and a GT2 FXR (Unrestricted in the former and let's use 25% for the latter) can both run 24lbs of boost and have different outputs. (!)

If you want to keep the plate for some reason, then fine, but the whole point of a restrictor plate is to flow less air - which means the turbo won't be able to force enough air into the manifold to produce the same 24lbs, otherwise the output wouldn't change anyway.

The engine should really run out of boost a lot easier with a plate, at the very least. I'd have to think of how that would work.

You don't take an existing turbo car, and throw a plate on it. Maybe you can redesign one with a plate to meet rules (as in WRC), but that's different. Then you know the flow requirements ahead of time and can attain a suitable turbocharger.

It would seem hard to get your head around it but its possible. boost is not what really creates power. it is quite possible to have the same car running 10psi with one turbo and 10psi with another and having one flow more air. In this situation It really comes down efficiency of the compressor. If your put a plate on the intake in order to get the same manifold pressure. the turbo is going to have to spin faster. in real life if your slapped one on you would prob take the turbo out of its efficiency area on the compressor map resulting in less power.

Now if you've played around with LFSTweak at some point. you can clearly see that LFS Engine Simulation isnt really a simulation, its more of a Formula used to create a torque curve. (With the coefficients being all the variables you play around with.)what that means is just because in game it mentions restrictor dosen't mean thats whats actually happening in LFS. it might just be taking .005 away from general power.

developers have a pretty big to do list

no one who is against it is truly against it. The point is as much as we would love to have it, it can't be done to the extent that would justify it. whats the point of driving on monoco when its not accurate.

you can generate boost in neutral just not practically. as said before if you revved in neutral you'll hit the red line in a second. At redline Fuel and ignition are cut if you kept it there it would slowly build boost. but seldom would you ever want to launch at red line. if you have a two step launch control you could build boost in at a lower RPM. Eg a STi with launch control set at 4500 Rpm will make 8 PSi and but no more.

think the point being made here is, unless you want to pay for Eric to fly to Germany or to what ever track. Pay to have it shutdown for a week so they could collect the above data. simply not possibly

Quote from tristancliffe :

And Forza and GT don't exactly have realistic representations of the tracks. Sure the corners are in vaguely the right place and of the right radius/widths, but that's only the first 10% of building a circuit (imo).

quick break down on the other 90%

Quote from tristancliffe :

I'm in the minority here, but I don't think real tracks will make LFS better. We'd just moan about how unrealistic the lap times are, either because of LFS physics limitations or, more overpoweringly, how poor the track is. Unless it has every bump, surface change, kerb profile & condition, dusty area etc then it's not going to produce realistic races or driving experiences. It would just cause complaints.

I have to agree with him. absolutely no way they could make real tracks. Think for a second, to make a track you actually have alot of data about it. Wdiths, angles of corners blah blah. you also need to know where all the surface bumps and stuff are. I believe the GT team and Forza actually walked around the tracks collecting that sort of data about bumps and such. This team simply doesn't have the resources to implement a real track @ 2007 standards. and if it doesn't have that then it might as well be a fake track.

Quote from Shotglass :

so in other words their efficiency is rubbish => turn it down a little => problem solved ... exactly what i said

ok but their efficiency is only rubbish when on boost

Quote from Shotglass :

nobody argued it was that simple in real life
ultimatle an engine is just a device to convert energy stored in fuel into mechanic energy so guesstimating its fuel consumption via its power and efficiency is perfectly reasonable
all youd have to do to take your points into account is different efficiencies for different cars and youre done with it

but the point i was making is that high powered Turbo cars use more fuel than they need in order to keep their EGT down.

Quote from jtw62074 :

Sure it does. In reality the efficiency value there varies a bit with throttle and rpm, but fundamentally this approach is good enough for its purpose. To get nitpicky you'd want to use BSFC curves, but what difference would it really make to game play in LFS, and then how would one argue for or against the accuracy of the BSFC curves when the engines are completely fictitious, anyway?

If the mileage is too high or low on some cars, a quick tweak to the fuel efficiency value for each car is all that'd be needed. 30% is rather high for fuel conversion efficiency, in general, iirc.

There was a debate in rec.autos.driving about fuel efficiency versus fuel economy a couple years ago, where someone was insisting that the best fuel economy came at 40% of redline. That's not specifically what people here are arguing about, of course, but some of you might find it interesting as it gets into some calculations using real world engine data that is both throttle position and rpm dependent, and considers both in variation with a car's speed. The posting was pretty off the cuff by this point in the thread as the guy we were talking too was exceptionally dense (high volumetric efficiency? ), so my language and attitude is a bit flame-like.

The specific post is here:

http://groups.google.com/group ... ving/msg/4349618c20e0f5b4?

The rest of the thread is here for context:

http://groups.google.com/group ... p;rnum=1#44a65441d60ee4dc

I suppose it does work in an ideal situation.
from what experienced in real life its not as simple as the explanation the developer gave.

In real life when you tune an ECU Fuel map(S) its usually done based Manifold pressure vs RPM. an Na Engine tends to make more power running slightly rich around 13.5 AFR. But thats at WOT in the cruising area were fuel economy is concern you can pull the AFR back till its lean how lean depends, Exhaust Gas temps.

the algoritm in the game to me is fine for NA engines but Turbo charged car you have to run the AFR richer at WOT (around 11~12 ish depends on how much boost). you would actually make more power running a leaner mixture but your can't because the EGT would be too high. So at WOT you would actually be injecting more fuel than necessary and making less power for the amount you injected. so i think were the algorithm is going to break is when you have a NA car vs Turbocharged one the turbocharged one should consume more fuel.

So for me i really don't see power generated as the best way to judge fuel consumption. it would be interesting to see what MPG the GTR versions of the car get.

Quote :

- I didn't "invent" the fuel consumption figures, the fuel being used in any physics update comes from the current power output of the engine and the amount of energy in a litre of petrol, modified by an efficiency value (30% - most of the energy is wasted as heat).

well that explains the results shown. thats clearly not how it works in real life.