Usual stuff, quick to judge and not quite thinking it all through.
My real point was that there's not a single RL RWD car I've driven that spins or oversteers in such weird manners. IRL, a gentle squeeze of the throttle should gently shift weight to the rear and help stabilize the car, especially if your car has LSD. Before patch U in LFS, cars would simply snap.
As for snap oversteering the XR GT, it was WAY too easy then. All you need is just half throttle coupled with a tiny bit of constant steering angle.
Try drifting a 2 ton RWD SUV in the rain with half worn tyres. If you can do that consistantly and safely on a narrow 2-lane road filled with random puddles of water, THEN you're free to call me a no driving skill hack.
What I'm REALLY saying is that no RL RWD car I've driven with anything from low to moderately high levels of power, snap oversteers the same way as pre-patch U LFS cars did. LFS could really use some real tire data, which it MIGHT have acquired and implemented in patch U. In essence, patch U RWD cars handle much more closely to their RL counterparts.
My real point was that there's not a single RL RWD car I've driven that spins or oversteers in such weird manners. IRL, a gentle squeeze of the throttle should gently shift weight to the rear and help stabilize the car, especially if your car has LSD. Before patch U in LFS, cars would simply snap.
As for snap oversteering the XR GT, it was WAY too easy then. All you need is just half throttle coupled with a tiny bit of constant steering angle.
Try drifting a 2 ton RWD SUV in the rain with half worn tyres. If you can do that consistantly and safely on a narrow 2-lane road filled with random puddles of water, THEN you're free to call me a no driving skill hack.
What I'm REALLY saying is that no RL RWD car I've driven with anything from low to moderately high levels of power, snap oversteers the same way as pre-patch U LFS cars did. LFS could really use some real tire data, which it MIGHT have acquired and implemented in patch U. In essence, patch U RWD cars handle much more closely to their RL counterparts.
FYI, I don't use throttle or brake help at all. And in motorsport circles, TC is considered a "driving aid". I was speaking in terms of RL motorsport practice, no in a game sense.
And no, I don't use TC for my FZ50.
Well, if there's a "LFS Steering wheel aid fund", then no one would complain about lack of steering wheels and the fact that current examples are still pathetically expensive for their less than realistic feel and function. Alas, this is no charity...
Or our developers may come up with something like the vaunted G25 at under 250AUD, then much more poeple would actually WANT to use proper controls. Alas, that's not happening too.
And no, use of a clutch pedal should not mean that you should be faster. There is no logical reason at all for a "it's seemingly more difficult and thus must be faster" rule. There seems to be an unwritten rule or dogma here that states that if it's more difficult, it must be more sim. This is an urban myth that must be eliminated.
Remeber pre-patch U tyre physics? AMAZING lack of longitudinal grip. It was much more "difficult" (to be honest it was complete lottery) to drive, especially the RWD cars, so it MUST be more sim than it currently is now right? WRONG! An under powered RWD like XR GT struggling with power understeer all the time? Now that's a sick joke.
Driver aids must mean slower? Which millenia are you from anyway? If you can consistantly beat TC equiped RL mdoern F-1 car with another such car with TC off for just 10 laps, THEN I'll agree that driver aids must mean slower. If you can shift faster than an SMG-II equiped car can, THEN I'll agree that driving aids must mean slower because "IT'S MORE SIM/REALISTIC". I can go on forever on this, but that's not the point.
Hyperactive, if you could somehow manage to build quality 3 pedal H-gate and pedal shift steering assemblies that provide unprecedented levels of realism, while still selling this to fellow LFS players at prices no greater than 200AUD, THEN you can say whatever you want without a single thought for so many fellow LFS players who are a bit less fortunate than you are. Otherwise, a touch of thoughtfulness goes a long way to welcoming more people to LFS.
As a final point, as long as quality 3 pedal sets aren't the norm and clutch simulation remains unfinished, there MUST be no changes that force the use of 3 pedal sets to achieve competitive and fun racing. The real problem here isn't the driving aids and the auto-clutch. The real problem here is flat shifting, something that can only be fixed by properly modelling engine and drivetrain damage. Otherwise, this whole thread is absolutely pointless.
Just been trying to race online just a week ago. Incesssant disconnects for unexplained reasons. 
Couldn't do a single race properly thanks to unexplained lag and disconnects. This at 2 digit pings.
Couldn't do a single race properly thanks to unexplained lag and disconnects. This at 2 digit pings.
Auto blip or manual lifting are both acceptable ATM due to obvious limitations stemming from LFS's unfinished clutch simulation. The only thing that is REALLY unfair here is flat-shifting with a button clutch, which is imply wrong. As I've already said, this can be at least partially remedied by implelemting much realistic engine, gearbox and drivetrain damage. Flat shift = major mechanical trauma = possible gearbox or clutch destruction, especially for higher powered/torqued cars. That's how it is IRL and that's how it should be in LFS. IRL, race sequential dogboxes can be shifted clutchlessly, but this requires precise throttle lifts/blips, something that's beyond the average noob.
And don't quote drag cars. They're only good for a few passes before anything blows up. LFS is predominantly a racing sim, not a drag racing game.
And don't quote drag cars. They're only good for a few passes before anything blows up. LFS is predominantly a racing sim, not a drag racing game.
Last time I checked, clutch use is facing obsolescence in the top level racing world. There's no way a human can shift in under 200ms. A sim is supposed to approximate reality as closely as possible, isn't it? Hence clutch use is ideal for the road cars, but just plain silly for a car like the BMW Sauber F-1.
I've wished for WRC style rally stages in LFS for a long while, but ATM, even the gravel terrain simulation is quite off. Wish it'll happen, but unfortunately, it's still some way off.
Well, that's exactly what I was thinking. Glad you're on the side of sensibility.
What unfair advantage? The only unfair advantage here is the absurd button clutch on-lift shits. Major engine, clutch, gearbox and driveshaft trauma. The auto clutch shifts pretty much as quickly as any skilled driver with a good gearbox and maybe a short shift kit could shift IRL anyway.
And last time I checked, very good and resonably priced 3-pedalled steering sets aren't exactly standard issue, so "forcing" clutch use is just plain stupid. I simply refuse to waste hard-earned cash until excellent and well-priced products dominate the market.
Besides, LFS is still devoid of ANY clutch simulation anyway. RL clutches have all sorts of intricacies such as slip points, grab points, etc. Does LFS punish ham-fisted use of the clutch as RL cars do? No. For instance, it's still impossible to stall a car in LFS through less than smooth clutch and throttle inputs.
I'm all for realism, and IRL, good auto-clutch sequential gearboxes such as Ferrari's Magnetti Marelli systems always outperform full manuals. Well, the LFS auto-clutch can't quite shift in under 200ms as RL Ferrari Enzos could, so there is no real unfair advantage. Unless, of course, you can't use full manuals properly (aka 95+% of the current driving population). If LFS simulates gearbox damage and wear, no-lift shifts will be obsolete for races of decent lengths real soon.
AS far as I'm concerned, there should be only 2 choices:
1. Auto-clutch
2. Properly simulated analogue clutch
As for a car like the BF1, auto-clutch MUST be mandatory, since the RL car is has a semi-automatic sequential gearbox. All the dedicated race cars (aka cars with slicks as a tire option) should be possible to shift clutchlessly with no damage like RL racing dogboxes, but ham-fisted operation MUST be punished with gearbox damage.
Point is, theres is no point to mess around with auto-clutch now. Just make a properly working clutch for LFS. Besides, our developeres have much more important physics issues to correct right now. MAJOR issues such as aero modelling and turbo modelling.
I just tested the RB4 yesterday to check the exact behavior of the LFS turbos. I started out stationary in 1st and started off in 1st (not drop clutch racing launch). It takes 5000rpms to fully spool the turbo to its amazingly small 12psi or so amount of boost. When was the last time I saw a 2.0L turbi I4 with only 243hp and a TINY turbo take so long to spool up? Never. I repeated the test for all 5 gears and the boost vs rpm behavior remained constant.
Just goes to show how patheticly off turbo modelling in LFS still is. With such fundamental physics issues such as this still unsolved, the last thing Scawen should do is spend time on messing with the auto-clutch.
And last time I checked, very good and resonably priced 3-pedalled steering sets aren't exactly standard issue, so "forcing" clutch use is just plain stupid. I simply refuse to waste hard-earned cash until excellent and well-priced products dominate the market.
Besides, LFS is still devoid of ANY clutch simulation anyway. RL clutches have all sorts of intricacies such as slip points, grab points, etc. Does LFS punish ham-fisted use of the clutch as RL cars do? No. For instance, it's still impossible to stall a car in LFS through less than smooth clutch and throttle inputs.
I'm all for realism, and IRL, good auto-clutch sequential gearboxes such as Ferrari's Magnetti Marelli systems always outperform full manuals. Well, the LFS auto-clutch can't quite shift in under 200ms as RL Ferrari Enzos could, so there is no real unfair advantage. Unless, of course, you can't use full manuals properly (aka 95+% of the current driving population). If LFS simulates gearbox damage and wear, no-lift shifts will be obsolete for races of decent lengths real soon.
AS far as I'm concerned, there should be only 2 choices:
1. Auto-clutch
2. Properly simulated analogue clutch
As for a car like the BF1, auto-clutch MUST be mandatory, since the RL car is has a semi-automatic sequential gearbox. All the dedicated race cars (aka cars with slicks as a tire option) should be possible to shift clutchlessly with no damage like RL racing dogboxes, but ham-fisted operation MUST be punished with gearbox damage.
Point is, theres is no point to mess around with auto-clutch now. Just make a properly working clutch for LFS. Besides, our developeres have much more important physics issues to correct right now. MAJOR issues such as aero modelling and turbo modelling.
I just tested the RB4 yesterday to check the exact behavior of the LFS turbos. I started out stationary in 1st and started off in 1st (not drop clutch racing launch). It takes 5000rpms to fully spool the turbo to its amazingly small 12psi or so amount of boost. When was the last time I saw a 2.0L turbi I4 with only 243hp and a TINY turbo take so long to spool up? Never. I repeated the test for all 5 gears and the boost vs rpm behavior remained constant.
Just goes to show how patheticly off turbo modelling in LFS still is. With such fundamental physics issues such as this still unsolved, the last thing Scawen should do is spend time on messing with the auto-clutch.
Last edited by Jamexing, .
Not completely true. Yes, it's expensive, but 2 way adjustable suspensions are very common for road cars used on the race track. Last time I checked, Toperformance could make custom 2 way KONI sport coilover dampers for my Mitsubishi Tredia for 300+AUD per side if I wanted to. Of course, the spring stiffnesses would be bespoke to the customer's specifications.
And last time I checked, i could easily increase the camber adjustments of my Mitsubishi Tredia by +-1.5 degrees with a simple set of camber bolts from whiteline suspensions. And they cost less than 80AUD!
The one adjsutment that defies realism in LFS is the brake torque. No car IRL could adjust braking torque exactly as such. As for the brake balance, just get a brake balance bar form AP Racing or some other reputable manufacturer. They aren't horrifically expensive anyway and provide an excellent degree of adjustability as you upgrade your brakes.
Conclusively, LFS road cars aren't OEM, but are simply examples of what road cars with serious track use in mind would be.
Last edited by Jamexing, .
Mass increases by the cube of linear dimesional increase. In essence, if you compare a 1mx1mx1m block of solid steel compare it with a 2mx2mx2m block of identical density, the latter will weigh EIGHT times more.
Unfortunately, area increases by the square of dimensional increase, so the bigger it is, the samller the strength to weight ratio. Since stresses increase linearly with increase of mass, if an engine was simply scaled up in every to increase capacity, its revving ability MUST fall due to the reasons stated above.
For a RL example, if elephants were to possess the agility of cats, they would need to be over 90% bone. So bigger is absolutely sronger, but weaker relative to one's own mass.
Unfortunately, area increases by the square of dimensional increase, so the bigger it is, the samller the strength to weight ratio. Since stresses increase linearly with increase of mass, if an engine was simply scaled up in every to increase capacity, its revving ability MUST fall due to the reasons stated above.
For a RL example, if elephants were to possess the agility of cats, they would need to be over 90% bone. So bigger is absolutely sronger, but weaker relative to one's own mass.
Last edited by Jamexing, .
Last time I checked, 4G62 (1.8L turbo) equiped Mitsubishi Cordias could easily rev to 6500rpm without poblems. In fact, that's the optimal shift point, since it makes maximum power at 6000rpm.
Well, the original 4G63 equiped Mitsubishi Starions had 7000rpm redlines anyway, though they can easily be revved to 7500rpm with no problems.
Well, it seems that all this LFS cars rev too high business is absolutely obesessed with the XF Gti. For a RL comparison, a 1.3L NA econobox engine form a Mitsubishi lancer revs to 6000rpm and makes 88hp at around 5000, though that's in its OEM state. Though 7000+rpm 1.3L engines are possible to build and operate IRL, there aren't any cases that I know of yet. Well, to keep the XF GTi in the same class as the XR, its needs its 120 or so hp, but to do so while remaining NA requires increased revs, since torque production won't skyrocket without forced induction.
If we take an oversimplistic view and simply shift the peak power point to higher revs while maintaning the torque generated at maximum power, then the XF would need to rev to 6800rpm to generate 120hp, if the engine I talked about just now is used as a template. So given the power required for the XF and its remarkably small engine, its current redline isn't TOO bad. Granted, it is technically possible to tune 1.3L NA I4 engines to rev and generate such power reliably, though it won't be a cheap exercise to get 90+hp per liter out of it. Expensive, but doable, especially for a track used road car such as the XF GTi.
Well, the original 4G63 equiped Mitsubishi Starions had 7000rpm redlines anyway, though they can easily be revved to 7500rpm with no problems.
Well, it seems that all this LFS cars rev too high business is absolutely obesessed with the XF Gti. For a RL comparison, a 1.3L NA econobox engine form a Mitsubishi lancer revs to 6000rpm and makes 88hp at around 5000, though that's in its OEM state. Though 7000+rpm 1.3L engines are possible to build and operate IRL, there aren't any cases that I know of yet. Well, to keep the XF GTi in the same class as the XR, its needs its 120 or so hp, but to do so while remaining NA requires increased revs, since torque production won't skyrocket without forced induction.
If we take an oversimplistic view and simply shift the peak power point to higher revs while maintaning the torque generated at maximum power, then the XF would need to rev to 6800rpm to generate 120hp, if the engine I talked about just now is used as a template. So given the power required for the XF and its remarkably small engine, its current redline isn't TOO bad. Granted, it is technically possible to tune 1.3L NA I4 engines to rev and generate such power reliably, though it won't be a cheap exercise to get 90+hp per liter out of it. Expensive, but doable, especially for a track used road car such as the XF GTi.
Last edited by Jamexing, .
No, the truth is MOST LFS cars do NOT rev too high. For instance, the turbo GTRs make maximum power at about 6000rpm. Last time I checked, the 600hp twin turbo 3.6L engine of the Porsche GT1 reved to about 7500rpm, though unrestricted ones would go to about 700hp at similiar peak rpms. Though the LFS 2.0L GTR engines are based on the principle of making massive torque with the use of massive turbos and around 3bars of absolute inlet pressure, so it isn't too bad.
Most cars actually rev perfectly, but have absurdly bad powerbands. The classic case is the RB4. It revs easily to 7500rpm, which is reasonable for its kind of engine, but maximum power comes at only 6000rpm. Again, there are serious turbo modelling issues here. Yes, the RB4 will rev to the 9000rpm peak of its rev counter, though this is as far as it will actually go with no rev limiter and doesn't do any good. For a RL comparison, the 4G763 2.0L turbo would rev to 9000rpm if the rev limiter was removed, though then again such excessive revving does it no good anyway. In fact, the engine can safely withstand short transients to that rpm, though it is not advisable.
Next is a RL DIESEL case study. I happen to be familiar with the 4D56 diesel engine. It's pretty revvy for a diesel engine, redlining at 4500-4700rpm, depending on tune, with peak power usually between 4000 to 4200rpm. Some fellow Pajero drivers that use the same engine have actually reved it to its 6000rpm peak for a couple of seconds without too much incident, though the timing belts, etc and other drive belts tend to go before the engine suffers really significant damage. So yes, it CAN overrev to such rpms, though it makes no power there, so it's pointless.
The only car I've REALLY noticed to have a silly redline is the FZ50. Last time I checked, such flat sixes from RL Porches redline a around 8200rpm, and it's the latest 997 model GT3 with about 415hp! Maximum power is at around 7600rpm anyway. That 9000rpm redline for such an "underpowered" (360hp) car is just absurd. Hmm, must be a revvy and torqueless unit, but it certainly doesn't drive like one. I can reasonably accept that a 9000rpm redline is a necessity for a 3.6L NA to achieve 490hp, but a 9000rpm redline from such a relatively large engine that somehow makes only 360hp is a bit silly. Maybe the LFS tuning division used MASSIVE bores and tiny strokes...
Well, at least the TBO cars have realistic redlines. So do all the single seaters. The GTRs aren't far off considering what they are. The LX cars are also good, given that they make great power from such small engines. Last time I checked, my friend's Corolla Sportivo has a 8200rpm redline with a peak power of 191hp@7800rpm at the output shaft, though RL tests confirm the value to be around 180hp at similiar revs with all the accessory belt drives on. So the 9000rpm redline for a 190hp 1.8L engine is perfectly realistic. As for the RA, it's powerband values and rpms are copies of the real thing and thus, just fine. Just need to fix that absurd turbo model that requires even a less than 1bar boost turbo an eternity to spool up and the RA would be as great a drive as it should be.
Most cars actually rev perfectly, but have absurdly bad powerbands. The classic case is the RB4. It revs easily to 7500rpm, which is reasonable for its kind of engine, but maximum power comes at only 6000rpm. Again, there are serious turbo modelling issues here. Yes, the RB4 will rev to the 9000rpm peak of its rev counter, though this is as far as it will actually go with no rev limiter and doesn't do any good. For a RL comparison, the 4G763 2.0L turbo would rev to 9000rpm if the rev limiter was removed, though then again such excessive revving does it no good anyway. In fact, the engine can safely withstand short transients to that rpm, though it is not advisable.
Next is a RL DIESEL case study. I happen to be familiar with the 4D56 diesel engine. It's pretty revvy for a diesel engine, redlining at 4500-4700rpm, depending on tune, with peak power usually between 4000 to 4200rpm. Some fellow Pajero drivers that use the same engine have actually reved it to its 6000rpm peak for a couple of seconds without too much incident, though the timing belts, etc and other drive belts tend to go before the engine suffers really significant damage. So yes, it CAN overrev to such rpms, though it makes no power there, so it's pointless.
The only car I've REALLY noticed to have a silly redline is the FZ50. Last time I checked, such flat sixes from RL Porches redline a around 8200rpm, and it's the latest 997 model GT3 with about 415hp! Maximum power is at around 7600rpm anyway. That 9000rpm redline for such an "underpowered" (360hp) car is just absurd. Hmm, must be a revvy and torqueless unit, but it certainly doesn't drive like one. I can reasonably accept that a 9000rpm redline is a necessity for a 3.6L NA to achieve 490hp, but a 9000rpm redline from such a relatively large engine that somehow makes only 360hp is a bit silly. Maybe the LFS tuning division used MASSIVE bores and tiny strokes...
Well, at least the TBO cars have realistic redlines. So do all the single seaters. The GTRs aren't far off considering what they are. The LX cars are also good, given that they make great power from such small engines. Last time I checked, my friend's Corolla Sportivo has a 8200rpm redline with a peak power of 191hp@7800rpm at the output shaft, though RL tests confirm the value to be around 180hp at similiar revs with all the accessory belt drives on. So the 9000rpm redline for a 190hp 1.8L engine is perfectly realistic. As for the RA, it's powerband values and rpms are copies of the real thing and thus, just fine. Just need to fix that absurd turbo model that requires even a less than 1bar boost turbo an eternity to spool up and the RA would be as great a drive as it should be.
Last edited by Jamexing, .
I meant pie as in the constant that relates the circumference of a circle to its diameter. Its exact value is unexpressable since it takes infinite decimal points and is in fact numerically represented as an infinite series, but the approximae number used for pie is 3.14159.
Well, the proper SI units for power and torque are Nm and watts or kW for cars, since engines simply make 1000s of watts anyway. And if the SI units for angular velocity and torque are used, the following equation for power in watts holds true:
Power = Torque x Angular Velocity
Whereby power is in watts, torque is in Nm and angular velocity is in rads/s aka RPM/60x2xpie.
Power = Torque x Angular Velocity
Whereby power is in watts, torque is in Nm and angular velocity is in rads/s aka RPM/60x2xpie.
RL RX-8s redline at 9000rpm. Case closed.
Notice that LFS's simulation of realistic car behavior is computationally intensive since it calculates and executes the behavior of a car in RT (real time). So whenever another car is shown in close proximity, computational demands increase accordingly. And I doubt any single PC could confortably handle mathematically simulating 40 cars at the same time and in close proximity to each other with LFS levels of realism. Some games do it via various means, and some so-called sims don't do much real-time computations, since the calculations are either hghly simplified via the use of lookup tables. Either way, the game lags.
Currently, my PC barely manages to handle 3 cars on screen at once (my own car included). Any more means MAJOR lag.
Currently, my PC barely manages to handle 3 cars on screen at once (my own car included). Any more means MAJOR lag.
Agreed, nikimere, but one could always wish. This is an improvement suggestions forum anyway. Could be an S3 or S2 final thing as far as I'm concerned.
Those FXO wheels look a lot like RL OZ Superturismos. Anyway, the RB4 would look a LOT better with 17 inch BBS forged multispoke wheels, like those on the Mitsubishi Lancer Evolution XI MR.
And yes, race oreinted wheels only.
Last edited by Jamexing, .
Wish our developers could somehow make a similiar sort of deal as they had with BMW Sauber F1 and the Michelin Tires. BBS wheels are RL race wheels that look good and perform even better. Besides, it adds to realism in a similiar way the Michelins do.
Last time I saw an alfa it spent more time in repair than on the road. Some European cars are heavily overhyped. For instance, the Ferrari F50 was claimed by the factory to do 202mph, but the last time I checked, the gearing won't allow it without going over its 8500rpm redline to just over 9000rpm. Only possible if you remove the rev limiter, something that isn't such a great idea. At redline, it was actually geared to go 190mph only.
Yes, LFS does have a sheer lack of tire noise. For instance, the road cars in LFS aren't THAT loud, but they all lack the relatively low frequency sound that sounds like a basketball bouncing off the floor at the instance they are perturbed, say when running over a washboard surface or over a curb.
There's always the VERY easy way (weigh down the FXO enough to slow it down) or there's always the more optimal solution (make other cars as fast as they should be for RL cars with their power to weight ratios). If we improve the others JUST enough to compete well with the FXO, there's no danger of the TBO class being faster than it should be. ATM, TBO cars aren't exactly the most popular, and to slow the whole field down anymore would only discourage their use. Improving the other 2 cars until they perform as they should is more difficult than just slowing the whole class down, but doing things the better or best way never is.
Since some fundamental issues with turbo engines and aero aren't quite copmlete, I agree that slowing the FXO down is the best band aid solution for now. I'm just trying to be realistic. In the long run though, slight upgrades to RB4 and maybe XR GTT tires would be required if the TBO is to be as much fun as it could be. We don't even need FXO sized tires for the other 2 cars to closley match the FXO's performance.
For instance, our developers could simply cheat on the physics and use numerical tables like some other sims, but it's not the BEST way. The sheer insistance of our developers to properly model car handling with proper mathematical based models is one reason I purchased LFS2: the pursuit of the ultimate PC car racing/driving simulation possible on a PC.
Since some fundamental issues with turbo engines and aero aren't quite copmlete, I agree that slowing the FXO down is the best band aid solution for now. I'm just trying to be realistic. In the long run though, slight upgrades to RB4 and maybe XR GTT tires would be required if the TBO is to be as much fun as it could be. We don't even need FXO sized tires for the other 2 cars to closley match the FXO's performance.
For instance, our developers could simply cheat on the physics and use numerical tables like some other sims, but it's not the BEST way. The sheer insistance of our developers to properly model car handling with proper mathematical based models is one reason I purchased LFS2: the pursuit of the ultimate PC car racing/driving simulation possible on a PC.
Last edited by Jamexing, .
That's true IRL professional dedicated racing, but the TBO cars aren't exactly full blown race cars.
FGED GREDG RDFGDR GSFDG