You probably know what I mean with "Time advancing damage", for those who don't I'll write an example.

A wanna drive a fast two initial laps in some long race in Blackwood. In second lap I push my XRT too much and I fly over a kerb in S curve. I landed so hard and I've damaged my suspension. Not too seriously, only a little, but enough to make time adv. damage effect. Because I wanna win, I am still pushing and damaged spring is not able to take the heavy loadement. After some laps is suspension damage getting worse, but I am still keeping floored gas. After 20 laps spring definitly breaks and I am forced to retire due to serious suspension damage.

I hope everybody understand now what I mean by Time advancig damage.

I know that damage model is not completed yet and I know also that not every damage can scale how the time goes(bodywork damage etc). I didn't use as an example engine damage, because I don't want to get "Engine damage is not implemented al all, noob" type answers. If it were on me, I would prefer Time advancig damage to engine damage.(I almost lost the race with someone who has damaged suspension from first curve, and I wasn't driving bad)

Quote from thisnameistaken :

I think it would be hard to do.

I don't think so. There is no need to calculate exactly every multiplied force. Only thing that Scawen have to do is define Damage advance map for parts where can this effect appear. Damage advance map will calculate with damage seriousness and do all the necessary "work". You don't need to simulate damage advance (D.A.) EXACTLY, because every part of cars in LFS are fictious. You cannot say IT IS NOT REAL, because you can't know it. So we'll must familiarize with any further(I hope further) D.A. characteristics, I know that devs will do it properly.

Yes, this is a nice suggestion but i'd wait for S3 for these kind of things.
Wether you see it or not, this is more complicated stuff and is more
of a detail. Details will be in S3. Blur effects, real-time lighting/weather,
chrome wheels, blue led windshield washer nozzles, you name it.

For this to be coded, the base damage code must be close to perfect.
Most of these basic systems are being coded as we speak, this is S2.
S2 is not finished yet, we are priviledged in a way to be able to play it
before it's even done

Personally i'd like to see fuel consumption reflect throttle useage
more, but again, that's also a detail.

This could be added in the future, but the more obvious flaws in the physics engine such as diffs, aero ride height tricks and chassis flex should be sorted first.

I would think that if the damage model is done correctly and completely that "time advanced" modeling wouldn't be needed at all. This seems like something that would be used to fudge damage modeling when there isn't realistic enough modeling to deal with the circumstances.

Quote from Cue-Ball :

I would think that if the damage model is done correctly and completely that "time advanced" modeling wouldn't be needed at all.

You mean this that way the "D.A." will ensue from the damage model and there'll be no need to do it "extra"? If you mean it this way, you might be right, but I'm worry that SO REAL damage we won't have for a while(for a LOOONG while). And as I alredy said, we don't need 150% real damage calculations, because all the cars in LFS are fictious and we can't argue with devs if their damage modelling is real or not.

Try giving the fxr relentless abuse for about ten laps then come back and say what you want changed about the progressive nature of the simulated engine damage.

I think the general idea behind this is great, but IMO "time advancing" is a bit of a misnomer. Damage just doesn't happen just by itself, so time doesn't actually play a role. What does happen is that parts get more prone to damage if they are already damaged, up to the point where parts are so sensitive, that extremely weak forces, like just moving the car, already damages them.

This could be expressed in curves for each part, so lets say the dampers still hold together with 30% damage and need quite a beating to get further damaged, but from there on it goes downhill and just going off track could already damage them.

The only real "time advancing" damage should be applied to constantly active parts, like the engine for example, after a certain amount of damage occured of course, not already at 1%.

Remembers me of the Raikkonen Crash at Nuerburgring. He had a flat point on his right front tyre and due to the vibrations the whole suspension threw apart on last lap. But before we get that in sims we pretty sure will all be old

Quote from AndroidXP :

I think the general idea behind this is great, but IMO "time advancing" is a bit of a misnomer. Damage just doesn't happen just by itself, so time doesn't actually play a role. What does happen is that parts get more prone to damage if they are already damaged, up to the point where parts are so sensitive, that extremely weak forces, like just moving the car, already damages them.

THANK YOU. You said it just exactly........ We can call it Fatigue of material, can't we?
(LOL, I didn't expected that someone'll do the SEARCH and dig up my million years old idea)

Quote from Fonnybone :

Personally i'd like to see fuel consumption reflect throttle useage
more, but again, that's also a detail.

Do you mean instead of based on actual power output?

Personally I'd rather wait for a detailed S2 Final with the systems they've put in place already refined somewhat... I think S3 should be the glitz & glammour etc...

I think the progressive damage is a good idea

My 2 cents

I would like to see blowing tires as a result of hitting something. It's quite strange that a car hitting a wall at 120 mph can continue racing if suspension was luckily damaged symetricaly (me sometimes). With tyres blown it would be impossible. Indestructable tyres are also someway unrealistic. It would be quite easy to implement at this stage of developement of LFS.

There are too few situations in LFS when player has to visit pits to continue racing (at least in demo races). With more of them races would be much more thrilling IMHO.

What do you think?

Sorry for english.

Even Toca RD 2 had that kind of tyre damage But c'mon, each thing at his own time, let's enjoy S2 Alpha, I'm sure there will be some surprises for the final version, and some other in the ocassional patches . The complete damage model can wait until S3. IMO of course.

Quote from himself :

I would like to see blowing tires as a result of hitting something. It's quite strange that a car hitting a wall at 120 mph can continue racing if suspension was luckily damaged symetricaly (me sometimes). With tyres blown it would be impossible. Indestructable tyres are also someway unrealistic. It would be quite easy to implement at this stage of developement of LFS.

There are too few situations in LFS when player has to visit pits to continue racing (at least in demo races). With more of them races would be much more thrilling IMHO.

What do you think?

Sorry for english.

I agree...

Seems like the whole damage engine could benefit from just being much more "sensitive" than it is now, although it's probably not that simple

Impacts do affect tires already; if you have tiny amount of rubber left (IE 2 laps worth maybe) and you nail a curb then they will pop...

As for tires bursting on impact, in real life they do not magically wear down and rip to pieces, instead the wheel is physically bent, causing the tire to no longer sit correctly on it. This kind of damage will likely never be simulated, it's not necessary as if you've actually damaged it this badly you ain't going very far.

As for time advancing damage, that is misleading, I hope we do not get, keep going after picking up damage and the car just gets damaged more, pre-scripted events are a thing for GT, not in sims. As for damaged suspension components working loose and causing more damage that is a very good idea but I'd leave it until after some others have been addressed.

The most important 3 damage things should be:

1. Overlapping wheels in single seaters, it would actually cut down on crashes IMO as they'd no longer be driven like touring cars, essential for any realistic simulation of single seaters. Could be a server option to cut chaos in beginer servers.

2. Aerodynamic/Body damage a big thing to impliment, but actual straight line speed/advanced claculations are not needed. Just a simple loss of downforce when you modify your wings would also clean up single seater racing.

3. Kerbing actually damaging the car, I've seen big accidents caused from damaged suspension components due to this. IRL you would never ever consider taking the FE Green chicane flat in a Formula Renault, the only one place you'll gain from that is a bed in A&E.

Other things could be implimented are enigne damage as well as a whole manner of visual stuff such as detaching parts, smashing windscreens etc. but I just don't think the average users PC could handle the visual things and a physics model as detailed as LFS.

Quote from ajp71 :

...keep going after picking up damage and the car just gets damaged more, pre-scripted events are a thing for GT, not in sims...

You're right with your opinion about scripting. I know I have suggested it, but that was only for simplification(I didn't want to get 'That will be CPU killer' answers). Of course that scripting is lowering the realism factor, but BEFORE we will be able to get 100% calculated damage, some scripts in certain situations(window smash etc.) will not hurt IMO. I revised my suggestion about prescripted damage advance, and I realized that it isn't the best way how to model that....., you're right ajp71...

Well i don't think that most poeple, who read this, thought of something like pre-scripted
events. They probably meant calculated damage. The title is somewhat misleading there, but i think no one wants pre-scripted events.
As for calculated damage, imo this is something that is hard to emplement, don't know if it's even doable at this time.

I really think were're going to have to wait until the average user has a CPU that will be able to simulate suspension components moving in plains and axis they weren't designed to. Whilst it may be alright for some (and I reckon my P4 could cope) there's no point in needlessly isolating people.

Yes, current top systems could probably cope with it, actually, with calculating it for one (read: your) car. If you had to calculate it for the AI cars too... hello single digit fps range.

Right now every joint and axle is just a "logical" point on the car, interacting with other logical parts through the joints (kinda). If you wanted to simulate suspension damage accurately with parts going in places they don't belong to, you'd have to give everyone of these axes a own collision box and own physics.

Right now not even the collisions between cars and tyre+road edges works 100% correctly, so how should that ever work with 50+ additional independent objects per car. :faint:

I don't really think that it is so CPU intensive to simulate the forces acting in the suspension components. I mean, there is no need to simulate every bearing, axle bolt, etc... We already have (somewhat basic but) working model of suspension damage.The problem with current system is that it only calculates the loads parallel to suspension arms. Ot so it seems. This is why the famous "F08 suspension bending at high speeds causes the tire to gain some additional camber". The suspension arms just strech. Of course they bend somewhat too as the wheel position can change longnitudally (spelling) too. I don't know if torsional transitions are possible...

The problem with the current system is that the suspension remains quite functional even with heavily damaged components. Also:
- wheel can't drop off
- suspension components can't break, they just strech
- wheels don't bend
- suspension itself is very durable and can take quite massive hits with little damage

I am not saying that it is bad system, but if you look at what kind of suspension damage they have in rally for example, it mostly is some suspension arm broken or some joint broken. Maybe the springs or the shocks broken. Or the steering arm (right word?) is broken so the suspension is still almost prefectly functional, you just can't turn the wheel anymore. It has its own mind and it decides to turn when ever and where ever it wants

Basically suspension damage is that the parts take hits during the driving and the system handles these hits and as time progresses there are more hits, so the joints and arms wear and get weaker. If there is a bigger hit, some damage is done somewhere in the suspension system and the wearing gets much faster. Like when Räikkönen had that badly damaged front tire, the suspension had to take the excessive loads for many laps, with the loads growing all the time...And when he started to brake before the corner, the load had gotten too big for the suspension arms to handle (= possibly a crack in suspension arm or similar. It was imho really more about vibrations than braking or steering forces that broke the suspension...).

When we are talking about damage caused by dynamic stress (=vibration, variable load) the part that takes the loads, wear and the wear is basically a crack "growing" in the part. When this crack gets big enough, it is a critical point as the part gets much weaker after this. (of course plastics and different materials act little differently but...). So the model can be time advancing but maybe the real term to use is "hit amount" advancing suspension damage (). So with every hit the suspension gets weaker and the suspension wears all the time as it moves.

This stuff is quite self explanatory (spelling again) but let me continue a bit

Of course this makes me ask what kind of suspension related damage should be modeled in LFS?

• Vibrations? Maybe. Except for extreme situations vibrations don't break suspension parts in racing cars. The time is too short as the parts get changed before there is actual change of breaking down.
  
• Arms bending? Maybe. Though different materials bend very differently and therefore modeling the bending of an carbon fibre suspension arm is pointless. Sure high strenght steels bend less than pure iron. But the bending can seriously ruin the suspension geometrics.
  
• Arms breaking? Defenately.
  
• Suspension joint breaking/getting loose? Yes. No big difference if an arm gets loose from a joint than the arm breaking in the middle. But there are two/three joints per suspension arm.
  
• Drivetrain axles broken. Yes. Even though it may be more about bad engineering than that axles are easily breakable. But the effect of losing one is quite remarkable
  
• Wheels bending/breaking. Maybe not. Very little effect and happens usually when there is big hit which hits directly to the rim instead of the rubber, like driving over rock. And happen extremely rarely in racing on paved roads.
  

But I really don't believe that the current system is final, so there may be better one coming

That's it in nutshell

EDIT: I don't think collison detection is needed. It just makes the system too complicated. ?
EDIT: (a) typo

No, what you said is perfectly fine (and would work without too much CPU killing). I responded to ajp71's post of "things going where they don't belong". I may actually have misinterpreted it though.

I fully agree with what you said.