These articles are also very good (make sure to click "Next article" at the bottom when you're done, at least the first 3 are worth reading):

https://gafferongames.com/post/integration_basics/

Quote from Marino108 :

Please help.

This is a useful paper that explains how constrained rigid-body models could be simulated. It's not the only way to do it, but it's a good starting point to understand how complex it is modelling rigid body dynamics.

https://www.cl.cam.ac.uk/techreports/UCAM-CL-TR-683.pdf

The main purpose of me posting that link is, if you're not familiar enough with math/linear algebra/mechanics to understand most of it, then you'll need to get that sorted first.

+1

This is the only forum I use regularly that I can't get on tapatalk.

You could make one using Selective Laser Sintering (SLS), we got some parts made about similar size to a model car for around €60. It's made from powdered nylon sintered together and it's pretty tough. The company we used in the UK was called 3TRPD, you can just send them a common 3D file format and they'll make it for you, it's pretty cool.

Quote from lerts :

oh come on
if you can block wheels braking with left foot is like shooting penalties with left foot, you just dont have the sensibility to brake on the limit

You could charge millions by teaching that "fact" to most of the F1 drivers!

Quote from vitali.88 :

Perfect example!

Yep, that is what you are trying to acheive, but beware the driver taking the inside line can block this move by braking a bit more than needed in the corner, you will then be forced to slow or stay on the outside, if you stay on the outside you will risk being forced off onto the grass as he takes the normal racing line, or if you try and take the inside the driver in front can just defend that line because he is going slow enough now to make a tighter exit to block you wherever you try to go.

What I like to do when I see a driver taking the inside line is to force him to take the tightest line possible, ie by driving very close to the inside line myself behind him, or if alongside then edging over towards him (of course still leaving him plenty of room). Then just before the normal braking point you can dart over to the outside line and hopefully he will be going so slow you can just pass him around the outside before he has realised, or if not it likely won't be able to hold the inside line on the exit and you can take him there like in the clip above with MS against JPM.

Quote from scipy :

i dont think so.. when the engine is coasting there is almost no compression beeing made, cylinders are running almost on vacuum with intake vales beeing steady and exhaust valves opening slightly. what you are using to slow down is: let's say you're on 6000 rpm in 6th gear, start braking and downshift without a throttle blip, the engine has already started to slow down - it had some kinetic energy on 6000 rpm, and that energy is going down, also taking into consideration the rotating mass of the flywheel and the clutch assembly.. when the clutch reingages it will spin that mass once again, and the thing that opposes you is friction, not compression..

I see what you are saying, that is called "inertia" not "friction" - you will get the same effect if the system was perfectly frictionless because you still need energy to spin-up the components of the system. BTW, you do realise that friction is a constant force and not proportional to speed?

But, that doesn't explain why you get more engine braking at 6000rpm compared to 2000rpm (forget changing gear, just lift off at different rpms).

Quote from scipy :

btw you're not actually using the gearbox, it's the engine friction (and not engine compressions as it's popullary belived).

That's incorrect. Firstly, friction forces are roughly constant with respect to the speed of the surfaces in contact. So if what you said was true, the engine braking would be the same torque no matter if the engine was turning at 2000rpm or 8000rpm - obviously this is not true.

Secondly, have you ever felt how hard it is to turn an engine over with spark plugs in? Then with them removed (ie no compression)? The difference is huge, because it's the compression that makes it hard to turn over, and hence creates the engine braking.

Hehe scipy i remember trying to explain the exact same thing to another guy on my engineering course. He just could not understand that if you are braking at 100% using the pedal there is no way you can slow down any quicker by using the engine as well.

In a perfect world with perfect tracks, drivers and cars you would never need to use engine braking, *but*, there are 2 reasons IRL why engine braking can be useful.

1) In low-grip situations (wet grass, snow, ice) when the engine is connected to the steered wheels, it can prevent lock-up (at least help keep you around the point of maximum deceleration), using just the pedal it is very easy to slide off the peak and lock the wheels - then wasting a lot of distance getting the wheels turning again so you can steer.

2) It can dynamically control the brake bias, giving more control to the (skilled) driver to keep all 4 tyres as close as possible to the maximum deceleration. At most tracks the optimum brake bias will not be the same at every corner, using engine braking you can make up for the variations.

I found http://en.wikipedia.org/wiki/British_racing_green in about 2 seconds on Google.

EDIT: Damn, beaten to it by a few seconds!

Quote from Bob Smith :

Don't forget that any car brakes quickest at high speed, since you not only have the tyres grip to slow you down, but air drag. With the massive drag of the wings/undertray on the BF1, combined with the low weight, I think you can easily get over 1g of braking, before even touching the slow pedal.

Indeed, if you are at the top speed using all 700 horses, as soon as you lift off you have 700 horses of drag pushing back on you. Just tried the BF1 at about 200mph (no wing), and sticking it in neutral gives 0.6g of "braking".

Quote from AdamW :

hate to split hairs, but F1 cars don't stop so fast because 'the brakes are good' but because they have gigantic eff-off tyres. Just about any modern car (certainly any modern _racing_ car) has brakes that are more than sufficiently 'good' to make the contact patches of the tyres the limiting factor in stopping power.

It's because of the downforce actually. There is no way you would be able to brake at anything like 4g without the aero aids, the tyres alone wouldn't get you above 1.5g, if that.

Just a quick movie to show off some code I've been writing to control the camera position. I know the three views are not perfectly sync'd, but it was thrown together in a bit of a rush!

http://www.scottboham.co.uk/lfs.avi
[10 MB]

OK I fixed it. What I think the problem was is that when I send the init packet to InSim, I set the reply port to 0 (which means send replies back to where they came from). For some reason when OutSim sees this, it does not send the packets to the same port (maybe it sends them to port 0, is that possible?).

Anyway, I set up two separate sockets (one for tx and one for rx) and it works fine now. I hope this answer will be useful to someone else at some point!

I'm using InSim from C++ and all is working fine, but I need to get the position of the car for my particular app. I sent the "SSP" command and inside LFS I get the message "Outsim enabled", but then I cannot detect any received OutSim packets from my app, ever.

I'm probably doing something obvious wrong, but am I right in thinking that when OutSim is enabled from InSim it (OutSim) will send the packets to the same place as InSim sends its replies?

Secondly (this was asked before on here but no answer), in InSim.txt it says you can detect outsim packets by length 64 and first byte zero. But in Readme.txt it says the first "int" is the time in ms, and the toal length is 68 bytes (including Game ID). My assumption is that when enabling OutSim from InSim, GameID won't be sent (making it 64 bytes) and the first byte of the time will always be zero. Is this correct?

So I am left unable to work out why I'm not getting any packets from OutSim. Any ideas or something I can test?

Yes, the setting does need some work :-) it does look a bit like an RC car now you mention it!

My first try at rendering the 3D models, didn't realise this was possible or that people were interested in it!