Doublepost.
Back on subject. This has been discussed before.
The damage does not change aero right?
Back on subject. This has been discussed before.
The damage does not change aero right?
he is talking about the 3 dimensional draft. Normally cars provide a larger wake than just behind them. In NASCAR you will notice cars taking advantage of the side draft to slingshot at times after pulling out from behind another car.
Was talking about generality. I know this model quite a lot because of my real life job where I use it. It is basically only an example (and tbh the model that almost everyone use in automotive industry including car industry (peugeot & Renault) and motorsport aswell) to show how tyres modeling is difficult. Working already a lot about making tyres works irl to pay any serious attention on how it is set in LFS (btw this model seems to be quite close to Pacejka anyway, which is quite normal as they are both about tyres).
About the damage, I have never felt a diff in aero but sometimes it get some really nasty effect when some aero/body parts touch the ground.
Any let's be honest LFS physics are not perfect but LFS are still using the best physics I have ever tried and btw we don't have any of those at home to have a dynamic simulation like LFS provide. I don't know how far we are from the max we can do on computer with LFS but LFS physics are really good and tbh close enough to the reality for some motorsport fan like me that don't have any real driving skills.
Nice posts back there Momo92i.
We have kickass good community if we put sarcasm aside once and while.
Too bad we cant help out on the work being done to LFS.
We have kickass good community if we put sarcasm aside once and while.
Too bad we cant help out on the work being done to LFS.
You misinterperet. The current model you're all driving around on is an empirically based model that Scawen invented, perhaps similar to Pacejka (I don't know the details). I don't know if this is the same basic model that Scawen has been tweaking since 2002, or if he's ever started over.
The first new tyre model Scawen wrote was based on brush model theory. That model never made public as, although better in some areas, went very wrong when taken outside the areas covered by the assumptions and simplifications needed to make it work. So Scawen started a second new tyre model, based more on tread model theory, however it's ended up too slow for real time use in game, so there's now a third new tyre model, that aims to give results similar to the tread model, yet be sufficiently simpler to run in real time. That's no easy challenge.
For me a real simulation would emulate the environment of a race, and the aerodynamic / grip debate is a key case in point.
I realise I am in the minority around here with this view, but to me racing is about dog fighting and not lap times. I am therefor more interested in the simulation of a race than I am the simulation of a car.
If you go and watch real club/national level motor sports and watch some downforce cars racing you will notice a strange trait: The cars tend to be spaced apart by a fairly equal amount. Over the course of a race a few small gaps will emerge and the cars may break away from each other, but many will remain in 1 or 2 chasing groups and all the cars in those groups will maintain a fairly even split.
I've watched a lot of club and national level racing, and this trait is fairly common in downforce races.
In LFS any race involving cars with downforce will instantly begin to spread out.
This is partly due to a wider skill gap, but mostly it is because the aerodynamics are wrong. Following cars is too easy in corners, so overtaking is relatively easier, allowing faster drivers to pass and get clear of inexperienced drivers much easier than in real racing.
The feel of the cars in LFS isn't bad. If LFS' physics were perfect (which is Scawens current mission) then the physics would still not be the real thing, and much like the physics now - would feel "not bad".
So in my mind the real error in LFS' physics is that the final result, whilst it is "not bad", does not emulate a race.
Most of the cars in LFS I will never race for real anyway, I am therefor not too worried if the cars are perfect or just "reasonably good" in terms of their physics, instead I would rather be reacting to changes to air and track temperature and other dynamic elements of the race track which no sim yet bothers to simulate - and I would rather the physics allowed for the skills differential between gamers and created a final product which was more akin to real racing - whilst still keeping the physics feeling "more or less ok".
I am passionate about motor racing, and passion is only 1% reason. Car physics is "reason".
I realise I am in the minority around here with this view, but to me racing is about dog fighting and not lap times. I am therefor more interested in the simulation of a race than I am the simulation of a car.
If you go and watch real club/national level motor sports and watch some downforce cars racing you will notice a strange trait: The cars tend to be spaced apart by a fairly equal amount. Over the course of a race a few small gaps will emerge and the cars may break away from each other, but many will remain in 1 or 2 chasing groups and all the cars in those groups will maintain a fairly even split.
I've watched a lot of club and national level racing, and this trait is fairly common in downforce races.
In LFS any race involving cars with downforce will instantly begin to spread out.
This is partly due to a wider skill gap, but mostly it is because the aerodynamics are wrong. Following cars is too easy in corners, so overtaking is relatively easier, allowing faster drivers to pass and get clear of inexperienced drivers much easier than in real racing.
The feel of the cars in LFS isn't bad. If LFS' physics were perfect (which is Scawens current mission) then the physics would still not be the real thing, and much like the physics now - would feel "not bad".
So in my mind the real error in LFS' physics is that the final result, whilst it is "not bad", does not emulate a race.
Most of the cars in LFS I will never race for real anyway, I am therefor not too worried if the cars are perfect or just "reasonably good" in terms of their physics, instead I would rather be reacting to changes to air and track temperature and other dynamic elements of the race track which no sim yet bothers to simulate - and I would rather the physics allowed for the skills differential between gamers and created a final product which was more akin to real racing - whilst still keeping the physics feeling "more or less ok".
I am passionate about motor racing, and passion is only 1% reason. Car physics is "reason".
I have been thinking would it be possible to do a openwheel styled flip like in some indycar or f1 crashes (for example mike conway @ indy and webber @ valencia). I know its possible with ramps and at aston's downuphill part but would it be possible in a crash? Sure lfs needs wheel-to-wheel contact physics before even trying it. 
Like this? http://www.youtube.com/watch?v=Gzledv-37JM
Yea something like that but what I mean is a crash like this http://www.youtube.com/watch?v=iZy-tPkZ3W4
The cars front jumps a bit because of the wheel-to-wheel contact and then gets some air under it causing it to take an massive airborne.
The cars front jumps a bit because of the wheel-to-wheel contact and then gets some air under it causing it to take an massive airborne.
http://www.youtube.com/watch?v=g8XxQkXCmsU
Not open wheel, but getting too much air under the car is bad in any case.
Not open wheel, but getting too much air under the car is bad in any case.
Thoughts on RF2's approach? Seems to me like they're concerned about the same things that you are; and although the current Beta has incomplete things and some switches turned off, their intent and idea towards your ideology are apparent and indeed already drive-able to some degree. Though the tire physics are rather decent ATM as far as I can tell, so it is quite promising.
aerodynamics i agree with, grip i don't on the basis that its a rough draft of whats yet to come and as is their isn't enough or it isn't dynamic enough ( hence the need for a new tyre model )
otherwise, for a sub 1 gb game the phisycs are alright; its indie developed... if you paid more then 40 quid ( Cod/nfs/Bf3 moneys ) then you could have room to complain, but like minecraft and many other indie games, you just have to play till you get bored, then they patch it so you play it again - Repeat until game dies.
Being indie before indie was cool.
For the lepers.
Well... I've never had ridiculous things happen like this. Sure, we had the buggy barriers, but at least there was a logical explanation behind it.
Does an off-track barrier have anything to do with it?
What I mean is, is it possible they don't want cars leaving the track?
I don't have any experience with iRacing.
No you can jump as much as you like with the cars on the old tire model and some with the new tire model. The Ford GT use to have problems where the back of the car had so much downforce that once you became airborne, the back of the car would slam down and you could wheelie forever. However if you got enough of a jump to go upside down, then you'd soon be going faster (like 180+ mph) upside down until you hit a wall.
All the cars don't always "behave right" when crashing, as if they don't have weight, inertia, or something. Been that way for a while.
So this is iRacing! I thought it's supposed to be better, as developers are stating that it's the best simulator on the planet ever created....
I guess iRacing is trying to beat LFS in funny bug competition.
Well, they already won, not sure why they want so huge lead... perhaps they know something about LFS NTM?
1
2
How realistic are the physics?
(43 posts, started )
FGED GREDG RDFGDR GSFDG