Quote from hazaky :

Wat, where's the difference? LFS wins makes hes ePenis longer? Its the same as some people can do 10x headshots in CS, but in real life theyr gonna crap theyr pants. Maybe not really in lfs, but u know what i mean.

At the moment I would say LFS is only good for teaching you the dynamics of racing, rather than actually making you better skilled at handling a real car. Although I can put some good laps together in LFS and hold my own wheel-to-wheel when I get into a Kart it makes no difference other than I know HOW to race rather than actually being able get the most out of the car.

LFS + Teaching = Win. LFS + Real Racing = Fail. Think what you will of that.

Quote from legoflamb :

It could go either way actually. Putting toe-in on the rear limits the tire's slip-angle. While grip is maintained, the car is more stable. However, the car will break grip with less angle of the whole car in a corner. While toe-out will allow more angle of the whole car in a corner, without breaking grip; but make it more unstable.

It would go either way depending on the conditions. Toe-out will create more oversteer in the middle and exit of a corner more so with throttle. Toe-in as you said will make it more stable in the entry but less so in the middle and exit as again as you said the reduced slip angle will cause the rear to lose grip earlier than if it had toe-out.

Quote from blackbird04217 :

Sure I don't know Java syntax directly but I think I would be able to pickup the ideas. I finished reading the NN chapter in that book; all except the source code implementation. I like some of the ideas, but would certainly need to test the idea of this outside of realms of the racing simulation; on something very simple and then one step further than simple without over complicating things. I now see that the implementation of NN is considered the easy part of the process, and that the hard part, time consuming part and challenging part is to get the required results.

I do understand that X,Y,Z inputs are patterns which if you answer with A will always lead to B; but when it comes to handling traffic, and other things I think this will lead to less than desired results. In other words, I can't think of any way to accomplish training the NN; which is the hard part of NN's!

Of course, as I said I can give you a simple practical tutorial where you create a small ANN to recognise 10 letters of the alphabet.

As long as you train the ANN to deal with traffic it would be capable of interacting with other cars on the circuit no more or less than if we raced each other.

At the moment I am speaking in pipe-dream terms and I think a good starting point would be to get a simple 2D car to navigate around an oval circuit with the two constrain parameters of crashes and lap time.

After a few google searches it appears this technique has been used before - http://www.ectri.org/YRS07/Papiers/Session-2/Booth.pdf

http://togelius.blogspot.com/2 ... ry-car-racing-videos.html

Last edited by Dac, Thu 28 Jan 2010, 15:49.

It is most certainly useful I can assure, every racing driver who ever lived had a brain that worked functionally as ANN do. There is always a pattern, and it's the NN's job to find it. Otherwise they are no use. All of the inputs I mentioned above are extracts from the environment that the car is travelling through hence that is the pattern.

The only concern I would have is in the implementation as you said. The only experience I have is with the JNNS which allows you to create, run and test neural networks very quickly. How you would go about mapping the input and output nodes to the simulation environment is I am hoping your expertise.

If you'd like I can send you the program and quick tutorial I was given to implement and run a simple NN.

Edit: A few google searches yields this http://www.ra.cs.uni-tuebingen ... tware/snns/welcome_e.html output's C-programs?

Last edited by Dac, Thu 28 Jan 2010, 15:31.

Please don't worry about wasting my time, I am quite passionate about improving driving simulators especially the AI.

A brief explanation for you. The hidden layer, yes that is where the meat is. NN work unlike any other system you have seen before, the only other example is biological brains. Every brain is built up of neurons. They take two input signals, add them together and decide whether or not to signal to the nodes connected to it. Since they are all connected together you have a network of individual and extremely simple neurons. Hence a neural network.

The magic happens when you join all these little building blocks together. You have input nodes which you activate, such as you could have a neuron for 'CarInFront'. When there is a car in front, this node would be switched to 'ON'. This activation would be connected to all the other neurons in the hidden layer, which are then connected to the output nodes. When the CarInFront node is activated it cascades into all the other neurons in the network affecting their activations, which in turn affect the activations on the output nodes, such as 'SteeringAngle'.

So you can vaguely see from this example how the inputs influences the hidden layer nodes which in turn affect the driving of the car.

This is an area where ANN can help you greatly. My interpretations of your goals is to make them operate on the basis that a human does. And for this ANN are just so perfect. I can understand how you are wary of them since you haven't researched them and from a few google searches they may seem too abstract and not very practical, but bare with me here and so you can see the same light I do.

Your method is a cure of the symptoms rather than a cure for the cause. You can go down this route but it will never be as genuine or appropriate as ANN. For instance, reaction times have a lot less to do with racing than you might think. Jeremy Clarkson I believe did a documentary on it and found that F1 drivers actually have the same reaction times as the rest of us. Where they differ is in their experience, just like a professional football (soccer) goalkeeper reacts to the smallest of cues in order to predict where the ball will go. This is working on experience to judge BEFORE the event happens the best course of action rather than working at the limit with superhuman reaction or slightly before as is your technique to try and 'cover up' the superhuman effect. As ANN work on experience there is no need to cover up their ability because they are genuinely functioning as a human would. So the problem is eliminated. If you see what I mean.

Let me just run you through the basics of it. You have 3 layers, input nodes, hidden layer nodes and output nodes. They are self explanatory. The input nodes correspond to the inputs a human would have, the hidden layer is the 'brain' which encodes experience into a purely parallel processing unit which uses the output nodes (hands, feet etc) to produce the action or in this case control the car.

Here is what immediately strikes me as necessary;

Input;
Vision
Feeling - Steering Wheel, G-forces, Pedals, Seat,
Sound - Engine, Tyres [FL, FR, RL, RR],

Output;
Steering Movement - Steering Angle
Pedal Movement - Accelerator Pressure, Brake Pressure
Head Movement - Left/Right
Gearbox - Shift Up, Shift Down

I have not included the hidden layer because in practice it is not an exact science, just like engine manufacturers find ignition timing maps by trial and error so do ANN find the best configuration for the hidden layer. Don't get disheartened there are only a few variables to control namely, the number of nodes and the learning rate (how big you want the changes to be within trial and error). You start off with the minimum number (i.e. 1) and then slowly increase until the desired behaviour is acceptable.

The concept is very easy to understand, just like the gates on a CPU work extremely simply so do ANN, but what is astonishing is when you put them in action you can see how easily the human psyche can be modelled. In fact it gave me a lot of food for thought about my own existence, and whether or not a good ANN would indeed be concious.

This area is highly worth pursing for you if you can spend a few hours researching it.

Quote from blackbird04217 :

Hmm. I should probably do some reading on NN's then, I've heard of them including mentioned once already in this thread; though my objective is not for an AI that 'learns'. Teaching the AI is not the idea here, and I don't see it fitting very well, however where you say that very human like output is achieved than I should at least give it a read over. Hopefully if it is useful it fits into the design of my AI already since I've spent some time on that.

I can see that your an open minded guy but I'm still sort of struggling with your goals. Myself I would like to develop AI that can drive on the track as good as the real deal and/or without you knowing they aren't human. Just like the Turing test where if they can be perceived as being human through the game then in this context they perform as good as humans would. That would open the doors for many possibilities from very clean racing all the time every time to be able to race against the legends. If we are on the same path then I'd like to continue this conversation.

Forgive me there has been some very thorough posts and it's quite a long thread so I only skimmed the first few pages. Interesting you mention this because I have been thinking or rather 'day-dreaming' about making AI cars not only drive like a human but to model top drivers to see how you compare against them, such as Fangio, Clark, Schumacher etc.

Also, last semester I did a AI module which was very interesting. The two dominant programming methods are artificial Neural Networks and Evolutionary Algorithms. The former I created for coursework using Java Neural Network Simulator (JNNS), not only are they extremely easy to create and train but produce scarily 'humanoid' type output.

I would strongly recommend you create a base Neural Network to learn how to drive over very many training examples (epochs) rather than picking out a corner-by-corner algorithm which is doomed to fail IMO. Using a NN the drivers would be able to use their experience to race against you which is the key to success in racing.

What's more, once you have a trained NN it is extremely fast to produce output and each one will be unique. So thinking aloud you could even pick out those which more/less experience for pro/novice difficulties or even those which show driving traits similar to the pros. You may even find that by presenting them certain driving conditions they learn how to drive with a certain style. Not only that but you could pit Schumacher against Fangio or whatever.

Another important reason I would say is that the NN must learn from your driving, so if you tend to cut them up and or move around in the braking area they will become aware of it take necessary action before the event occurs.

edit: Seems the idea has already been suggested and you dismissed it. All I can say is I wish you luck and that using the ANN route would provide vastly more fruitful results from your labour.

Last edited by Dac, Thu 28 Jan 2010, 12:33.

Quote from legoflamb :

I thought toe-out in the rear induces more axial rotation. And that rotational momentum would cause over-steer.

Yes that's supposed to be 'toe-out'. Will edit it. Problems with posting to early!

More toe in = More responsive steering and visa versa. Toe out on the rear has the same effect but since the wheels do not steer like the fronts it will be more inclined to oversteer.

Last edited by Dac, Thu 28 Jan 2010, 11:32.

I think a one car per race would be a good idea. Hate it when n00bs cause an accident then just re-spawn with a new car as if nothing has happened

Well this thread is being allowed to run it's course so I guess that's something, and we'd be foolish to believe the developers aren't aware of this arguing.

Quote from Vertic :

yeah I can't use punctuation correctly, sorry for that, I'm not a native speaker and I rarely write english.

But I agree, that is too long. Most people would like to see progress much more often, even if the steps are smaller and the released stuff is not perfect. But that is up to Scawen and co. to decide, after all it's their own product, they do it how they feel it suits them. And let's not forget that when people payed for the licence, there were no promises for updates, so basically any update you get is a gift. If you say you would be even willing to pay for the updates, I don't think that would help either, for the reasons I wrote in my previous post.

Scawen and co have developed a very good product with a large user base, so it's only natural that they want it too continue. It's like building a very good car and then deciding not making spare parts and accessories for it.

The problem we have is that we know there is enough money in LFS for it too become a lot more, but the developers refusal to employ more coders is always going to hinder progress and make it second best. Without updates LFS will go downhill because the competitors are competing, and the majority wont stick around to play a dinosaur like they will the bigger sims.

Quote from Vertic :

I don't think communicating the progress (or the lack of it) would do any good. Why are most people frustrated? Because updates have been promised, even a release date was communicated for the upcoming contents, and - not so much later - it turned out that the release will be delayed. For whatever reasons. People are now puzzled, how can the delay be that long? After all, iirc developers put up a release date a month or so before the original estimated date, and that has passed more than a year ago. That is, you have to admit, disappointing. Even if there are a lot of other extra stuff that has been, and will be done in addition to the original plan.

I, for one, don't care, because I play LFS only occasionally, and can't even remember the last time I went online, but will try the new patches whenever they come out.

As a software developer myself, I know how this stuff works, and I am surprised that, after all these years, Scawen and co. can still work on it. That alone, is a big achievement I think. And I wouldn't be surprised if, after being at it that long, many other things in life became more important than developing this single piece of software (no matter how much they like doing it), like family and so on. That, and the ever increasing complexity of the software slow down the development. You can always bump into a problem that you think you can solve in two days and then turns out to be two weeks. There will always be people who can't accept these facts and start to moan about the lack of progress here in the forums. Developing a software with this complexity is a huge task, and if only one person is doing it, it takes a lot of time.

Progress reports don't speed up the development - if there is no customer or, even better, co-developers to provide some pressure (which can help productivity to some degree, but can be counter-productive as well) - and don't make impatient people calm down either, so they're really pointless, unless the developer feels like writing one. Regular progress reports can help keep up the interest but, not for very long. More than a year, imho, is very long.

There always will be moaners, no matter how good or bad the communication is. Even if the patch comes out, there will be people saying: "that's what we've been waiting for this long? that's sh!t, try 'xyz' they're doing this and that so much better!" People who are bored, can't be satisfied, and grow angry over time, and there is nothing you or anybody can do to change that. Except for themselves: they need to find something else to do to cure the boredom.

What we, are, bothered about, is, not just, the fact that, they have been promised, but the lack of content, which hasn't been updated since 2007.

Quote from J03130 :

im sorry but why has every twat in this forum sniffed out this 'argument' and tried going against me????????? if your not involved dont get involved!

It's a public forum which means people are entitled to post. And if you weren't making such a fool out of yourself maybe you wouldn't be in this situation.

Also, how about tidying up your spelling and grammar but more importantly your attitude cause they all stink!

Quote from J03130 :

If track days count then yes i do race in real life. not my car obviously i dont wanna mess that up.

Track days are not races, therefore you know 'nowt',

Maybe LFS could have it's own investigations unit to seek out these kind of people and take appropriate action. Just the thought that there is an LFS 'Police' that they can be reported too would probably make these idiots think twice.

Quote from Bawbag :

2015

I like your optimism.

If the RB4 didn't look like the result of kinky sex with a Rhino and Crocodile id like it a lot more.

Quote from ACCAkut :

22: When you search for the buttons on the steering wheel if you want to look behind you

23: When you try to shift with your non existing shift paddels

After 2 on LFS I went to get a McD's and at the roundabout I actually went to press the steering wheel button to look right

Very good

I still can't fathom why anybody would need a cruise server :S

------------1

This is definitely something I approve of, sometimes you wan't to just go for it and take the penalty when you hit traffic.

Does anybody know just how much work went into making those cars though? Did they pull them apart like in Gran Turismo or just go for CoG, weight, etc?

Quote from z-ro 8 :

yeah, but then everyone will complain cuz they only have one real car and one real track....

The FBM is real

I honestly think once we have some real tracks there'l be a lot more people join just to race on it, and *hopefully* it will encourage the development team to go for more real tracks