Well, we are happy working this way. I really like doing the different sides of the code - physics, multiplayer, interface, graphics. Eric enjoys being a texture artist, a track designer, 3D modeller and so on. We like working from home and being able to have a day off in the middle of the week and work a Saturday instead, or in my case, go on a bike ride during working hours and work later in the evening. It's a life we have chosen to lead and it works. It is very different from the capitalist ideal of spending your life trying to make as much money as you possibly can, for no apparent reason. It is possible that Live for Speed doesn't reach the potential it could, but it sure does keep going on and on.

Let's say we decided to give up this way of life, restructure as a company and hire an office. At least one of us would need to sell our house and move nearer to the office, then start working 10am to 6pm, 5 days a week, hire people and spend our working days managing a team. We've been in that situation before, in more than one game company, and decided to do things differently. We have no desire to go back to that. But anyway, if we did (which we won't) then sure we could progress faster and then there would be these major setbacks, like the guy who did the physics leaves the company and no-one knows how the hell his code works, and then you need to find a replacement. The same reason why big commercial games are usually done quickly then they make another one.

Also let's say we somehow created a company for the sake of selling it to a corporation. They would require that we worked on for some years under their control. Again, not what we want at all. We don't want to sell it anyway, we just want to keep working on this thing we enjoy doing and which makes us a living. So that's what we are doing!

There were no changes in the physics of the cars. If any part of the car physics had changed then the old replays would not work. If you download some hotlaps, you will see they all still work as before.

If they are off then you are seeing the real vibrations of the car. Maybe it's worth softening the suspension or tyre pressure to reduce the harshness of vibrations.

I'm not sure I should say this but that's another thing that can work better in the new tyre physics (maybe depending on which car is selected, not sure about that). One thing we didn't like when we were trying to get the Scirocco ready for release, all those years ago, was that we had to use unrealistically hard suspension to make the car drive well. So maybe that's a characteristic of the current LFS tyre physics.

The current AI aren't deliberately slow, and if they are simply "turned up" they just crash. I did spend quite a while improving the AI in the new tyre physics system, years ago. So there is already some better AI waiting, with better racing lines and speed prediction code. I haven't yet done any work on the overtaking methods, but that can come later.

Quote from simon1234 :

I deleted all my blackwood pb's soon after the update is there any news if the online details will be reset. As at the minute im obviously getting false % off wr on lfs world.

I asked Victor recently and he said there is no such a thing as a WR based on online laps. If you type "/w wr" in game you get the hotlap WR. Online PB times are only for personal use and you can delete them yourself if you like.

I think the message you are seeing comes from Airio, as mentioned by cargame.nl, so I don't know about that.

Quote from sinbad :

I've never questioned it, but LFS records inputs as well as position, and there are some physics calculations at play. (I have questioned, though never asked, why distant cars might appear to lag in a MPR, yet when watching from that car or another close to it, the replay is solidly smooth. Even the MPR recorder's car might appear to lag when being watched from a distant car).

There are 3 levels of movement for remote cars.

1) The nearest 4 cars within 120m - full physics
2) More distant cars - low resolution physics (to save CPU)
3) Very distant cars - straight line movement since last update

I don't think number 3 is visible except when you switch to a distant car's view and you can see it moving in straight line for a second, with no car visible. I think that is disabled if you switch off the multiplayer speedup option. Sometimes you can see the inaccuracy in number 2 which is not optional.

Quote from Pasci :

Scawen: Is it possible that AI driver are less aggressive or reckless in the new version? Last race evening with AI drivers was really relaxed? :-D

I can not find any clues in the change list.

I haven't changed anything, just regenerated their KNW files for the new BL. Are you sure your AI were set to PRO?

The current tyre physics uses a slip-v-force curve that is made up to resemble the real thing, but not really based on physical principles. The lateral and longitudinal forces are combined in a made up way to produce reasonable results but again this is not based on physicsl principles.

So the result is that in various ways, the current LFS tyres don't react just like the real thing. We did make a lot of comparisons with real test data to try and get good results all those years ago and that's why it is not bad to drive.

I don't want to start discussing the new tyre model at this point but it is based on physical models of real tyres (with some assumptions and approximations) and compared with experimental data in better ways. The model is separated into the tread (contact patch) and the carcass (the belted, inflated, deformable structure).

My plan at the moment is to get the new tyre physics to a good condition where we are happy that it works better than the old one in all situations and then release it. There will always be more to be done, as it's not possible to make a perfect tyre model, so I expect there will be more improvements to come later, but without needing a complete rewrite.

That's something I'm interested in working on but it's a really big project to sort out. We need graphical frames in between physical frames, or alternatively 1000 Hz physics update rate. The solution will probably be connected with having physics and graphics on separate threads.

I'm interested in looking into this after the new tyre physics system is finished, which will allow many things to start happening.

Yeah, it should be nice to drive. Only with the current tyre physics, we had to use unrealistic setups to make the car handle well. That's the good thing, it showed up flaws in the tyre physics. The bad thing is it ended up taking years to develop the new tyre physics, and I took some wrong turns during that development. Also changed lifestyle in various ways during that time, LFS went through a quiet period.

Anyway it seems to me we are back on track, working hard on new things and LFS is having a gradual resurgence which will be helped by further developments we are working on and ones we aren't even working on yet.

Test Patch Q14 :

https://www.lfs.net/forum/thread/90424

This is intended as the final test patch!

Changes from 0.6Q12 to 0.6Q14 :

New Blackwood :
More fixes including some Z buffer issues and lod distances

Physics :
improved collision and tyre rolling on the 3D kerb objects

Fixes :
FIX : Removed message "Side mirror - point too central"

Last edited by Scawen, Fri 27 Jan 2017, 21:02. Reason : removed download link now that Q14 is available in the first post

Quote from Flame CZE :

Just tested it with FOX and it feels much better now! Does this only affect the new Blackwood kerbs or those kerbs on all tracks?

It only affects those kerbs at Blackwood. The new type of kerb that uses a 3D object for each kerb lump. They do not exist at any other tracks.

Quote from TurkemirtrZ :

So, new version (06R) has scirocco ? If has, i will in the shortest time buy a S3 licence, i used 2 years my friend S1 account . By the way you have any plans for Turkish translations?

As Flame said, the Scirocco will not be released in 0.6R. It will come with the new tyre physics which I am still working on but cannot give a release date estimate.

By the way, the Scirocco will be released as S1 content.

There is already a Turkish translation in LFS!

Quote from Ziomek21 :

Not really, tried FO8 at the KY2R buggy kerb, which had killed my car in the past, and it still shows no mercy.
Spr attached, the kerb action in question appears towards the end of the replay.

Sorry, this fix only affects the new kerbs at Blackwood. I understand the KY kerbs give a similar problem but I can't fix that in a similar way. Because the KY ones use track spline segments, not 3D modeller objects.

Quote from Xenix74 :

Btw: i would love to have these kerbs as layout objects

Good idea but we are trying to release this update tomorrow if at all possible. We don't want to divide the community online in two for any longer than necessary, at a time like this when there already aren't enough people racing online. And we want to get back to some other work.

Quote from Eri9k1 :

Will you open more input data for PostEffect shader?

Not now as for the same reasons, this update is all about releasing the new Blackwood and my head is really scrambled from several weeks of hard work.

Quote from BADLVBOY :

And ... why not increase the steering angle also to RAC, LX6, LX4 (also used for drifitng) and even RB4, FXO, basiclly for all road cars, people can choose whether they want/need more stering angle or not.

It was done for the rear wheel drive cars with macpherson strut suspension on the front, actually using a few lines of code from the new tyre physics version.

It is impossible to have such large steering angles for cars with front wheel drive and double wishbones can also get in the way of steering. So this was the simple and reasonably realistic thing I was able to do quickly.

Quote from Flame CZE :

Then I think the steering lock of the road cars should be increased from 720° to 900° or even 1080°, which should be closer to real life road cars. But that would require new driver hands animations, although I think the VWS has 900°, so pehaps the animation from there could be used? 45° at 720° sounds a bit too sensitive already.

This is already done for most cars in the new tyre physics system and I think it would be a bad idea to try to implement it in the public version now.

Really there is nothing more I want to do than get this update out there and get back to the tyre physics. Then we'll finally be able to release the Scirocco, the holy grail of sim racing!

Kerb Physics Test

Hi, I have attached an LFS.exe (Q13) which has a slightly differebt reaction on the kerb objects. The idea is to provide all the same bumps but avoid the large sideways forces that can unsettle the car more badly than in reality.

Please can you try it and see if it's safer on the kerbs?

NOTE : This is not intended to make kerb driving a lovely experience! It is an attempt to avoid some unrealistic actions, possibly leading to unexpected spins, particularly when your wheel goes over the top of the kerb and hits the steep edge of it.

0.6Q12 MUST ALREADY BE INSTALLED!

Eric is interested in doing those areas in a future update but he doesn't want to just enable physics, because it is not a proper job and has not been built as driveable areas. So it would be quite a lot of work and it is more important to continue with S3 tracks at this time.

Quote from UnknownMaster21 :

just same thing as on westhill, some tree hitboxes are infinitely tall... hehe

Can you point out some examples of this, maybe just two or three? They shouldn't be infinitely tall. I know that was a problem in the past but for Westhill I wrote some code to limit these "trunk" physics objects to be about the same height as the object. Though it does use the radius of the object's bounding sphere so for trees it is a bit higher than it should be, but should not be infinitely high.

WARNING : THIS IS A TEST

NOTE : THIS DOES NOT CONTAIN NEW TYRE PHYSICS


THIS PATCH CONTAINS A BLACKWOOD UPDATE AND IS INCOMPATIBLE WITH 0.6Q

YOU CANNOT REVERT THIS UPDATE!

IF YOU WANT TO STILL USE THE 0.6Q VERSION, MAKE A FRESH LFS INSTALLATION FOR THIS TEST


PLEASE TEST BEFORE YOU POST


Hello Racers,

Here is a new TEST PATCH : 0.6Q14

It contains a new version of Blackwood :

- New textures, geometry and detail improvements all around
- New Historic configuration (similar to GP but without chicane)
- New industrial estate area available when open config selected

Other incompatible changes :

XRG / XRT / FZ5 maximum steering angle increased to 45 degrees
Spawn position now disables "Could not join : Start is blocked"
Maximum number of layout objects increased to 2100 (was 1800)

Other changes :

There is an improvement in the car shadow system, which until now could cast shadows in the wrong direction.
Skid marks now remain remain when the race restarts and are visible on layout objects.
There is a new map view for open configurations and the track selection screen.
Please read the full list of changes below for more information.


0.6Q14 is NOT compatible with 0.6Q

- You CAN NOT connect online with 0.6Q
- You CAN play replays from 0.6Q excluding replays at Blackwood

You cannot upload hotlaps made with this test patch because it is only a test patch, not an official patch.


Changes from 0.6Q12 to 0.6Q14 :

New Blackwood :

More fixes including some Z buffer issues and lod distances

Physics :

improved collision and tyre rolling on the 3D kerb objects

Fixes :

FIX : Removed message "Side mirror - point too central"


Changes from 0.6Q10 to 0.6Q12 :

New Blackwood :

Various fixes for error spotted during testing e.g cracks and holes
Estate now uses two paths for echo maps and hidden object removal

InSim :

Reduced the frequency of IS_OBH packets due to blackwood kerbs
IS_OBH is always reported, as before, for movable objects
TYRE contacts no longer report IS_OBH for unmovable objects
BODY contacts also report unmovable objects played in a layout

Fixes :

FIX : Message text was faded on entering free view from escape menu


Changes from 0.6Q9 to 0.6Q10 :

New Blackwood environment :

- New textures, geometry and detail improvements all around
- New Historic configuration (similar to GP but without chicane)
- New industrial estate area available when open config selected

Other incompatible changes :

XRG / XRT / FZ5 maximum steering angle increased to 45 degrees
Spawn position now disables "Could not join : Start is blocked"
Maximum number of layout objects increased to 2100 (was 1800)


Changes from 0.6Q to 0.6Q9 :

Interface :

New small map view for open configs / car parks / autocross areas
Small map opacity can now be set for light and dark map colours
Improvements on track selection screen using the area maps

Graphics :

Shadows are no longer cast updwards or towards the light
Ambient (non-direct) shadow darkness now depends on distance
Skid marks now remain when race restarts (excluding replays)
Increased LOD distances of humans / drivers / marshalls
Skid marks are now visible on unmovable layout objects

Misc :

Westhill paths adjusted and regenerated to avoid graphical holes
Autocross paths regenerated and physical hole fixed at skid pad
New text command /lang X to load language (data\language\X.txt)

Fixes :

FIX : LFS could crash using /axload with a long layout name
FIX : Traffic lights did not work after texture resolution change


INSTALLATION INSTRUCTIONS :

A FULL version of LFS 0.6Q must already be installed

To install the PATCH using the SELF EXTRACTING ARCHIVE :

1) Move or save the patch into your main LFS folder
2) Double click the patch to extract it to that folder
3) When you see "Confirm File Replace" select "Yes to All"
4) Now you can start LFS in the normal way

NOTE : You can see if the patch is correctly installed when you run
the program (LFS.exe). At the bottom of the entry screen : 0.6Q14

THE FULL VERSION HAS NOW BEEN RELEASED SO YOU DON'T NEED THESE TEST PATCH DOWNLOADS - INSTEAD PLEASE FOLLOW THIS LINK : https://www.lfs.net/forum/thread/90458


DOWNLOADS :

IF YOU ALREADY HAVE 0.6Q :
PATCH 0.6Q TO 0.6Q14 (SELF EXTRACTING ARCHIVE)
www.lfs.net/file_lfs.php?name=LFS_PATCH_6Q_TO_6Q14.exe (132 MB)

IF YOU ALREADY HAVE 0.6Q10 (or later) :
PATCH 0.6Q10 TO 0.6Q14 (SELF EXTRACTING ARCHIVE)
www.lfs.net/file_lfs.php?name=LFS_PATCH_6Q10_TO_6Q14.exe (34 MB)



FOR HOSTING ONLY

DEDICATED HOST 0.6Q12 (non-graphical version for hosting only) :
www.lfs.net/file_lfs.php?name=LFS_S3_DCON_6Q12.zip (1.8 MB)

Last edited by Scawen, Sat 28 Jan 2017, 18:17.

Quote from Charlieh18 :

This patch stops RevBouncer from working from as far as i know

I don't know what that could be. I don't know of any changes affecting external programs (assuming they don't use some kind of exe hack).

Quote from Popughini :

They have to solve it, since it would not have to see the shadow through the concrete

The problem is that the concrete isn't casting a shadow. It's not really a bug that the car casts a shadow there. Really everything should cast a shadow, including the layout objects. That would be great, and I hope to do that one day, but it's a big project, not for this patch, and not until after the new tyre physics.

Yeah, I don't want to spoil the fun. I'll remove the code that switches off the physics in those areas, the map squares system will be the same size as before.

I'll need to figure out a different solution for the Westhill map, but it shouldn't be too complicated.

I wonder how many people drive outside the outer boundary fences at Westhill. I spent quite a while on this taking care not remove physics on any areas that you could drive to. I wasn't expecting to find out that people like to drive beyond the outer boundary fences!

Quote :

Also it seems that concrete slabs become nonsolid once they are out of map bounds although that was in previous patches aswell.

There is no physics on objects beyond the map squares system because there is no map square to register them on.

Map squares are a typical optimisation used in games, so you only need to check objects that are registered on nearby map squares, instead of running through huge lists of objects.

Yes, I did switch off the physics on some "out of bounds" surfaces at Westhill as they were increasing the size of the map squares system. It was related to the map because my task at the time was to restrict the physical area to the actual area we want to have mapped. That is important because the map only has a limited area available on the screen, and we don't want the map to appear too small.

It also has the side effect of making the map squares smaller which can have a performance benefit (though I have not tested this specifically).

I know what you mean but that area is really meant to be "scenery".

Quote from k_badam :

Shame about no tyre physics or anything.

Tyre physics just isn't ready. But I'm working on it.
The carcass works quite nicely and I'll be on the tread simulation soon.

Test Patch 0.6Q5 is now available with fixes and improvements!

Changes from 0.6Q3 to 0.6Q5 :

New map view for open configurations and track selection screen
Skid marks now remain when race restarts (excluding replays)
Skid marks are now visible on unmovable layout objects

FIX : Missing part of shadows at rear of LX4 and LX6
FIX : Shadows seen from a distance were not affected by haze effect

https://www.lfs.net/forum/thread/90208

Some other requests I still considering. Some requests, including the drift maximum lock increase, would cause an incompatibility issue. I do expect to release an incompatible version in the near future, so it looks like that should be possible, up to 45 degrees (no new tracks, cars or tyre physics in the incompatible version).

Quote from SergoMorello :

Sorry, for off topic, Scawen. Is it possible to make a little more adjustment for steering angle for rear wheel drive cars? As you already know how big drifting community is in LFS. Maybe just a few more degrees. It will not affect grip racing and mess with hotlaps. But it will help LFS drifters a lot, and will bring new emotions.

thank you

I don't know if this can be done without messing up all the hotlaps for those cars. But as it seems so popular, I can look into it. If it would mess up the hotlaps then I won't do it, an increase in steering range would have to wait for the new tyre physics, which I have been working on.

But, to help me, can you tell me exactly which cars need more steering angle to help with drift and how much that extra angle should be? It really must remain within the realms of "realistic" and I guess that means it should not include any cars with front drive.

Quote from mbutcher :

Is this going to be released as a stable build soon or is there more to come?

There is more to come. Explanation below.

Quote from pasibrzuch :

So shadows dissapeared from places they shouldn't be.

Will they come up to where they should be? e.g. yellow lines in pit entry/exit areas, layout surfaces or some 3d objects?

ps. btw of lyt surface, any texture for it + visible skid marks? :]

At the moment the shadows do not affect transparency, and I didn't do anything to change that. It's a good point and I haven't even thought about that! I think with all the other things I am working on, the shadows on alpha surfaces will not be changed in this update.

The problem with skid marks on concrete is if the concrete object disappears, we could have floating skid marks. I haven't yet worked out a system to solve this. Though I haven't put much time into it. I suppose it's possible by registering every bit of skid on an object, so they can vanish if the object is deleted... again, probably not in this update.

Quote from k_badam :

Can we see more of what is to be expected from this patch? It has been almost a month with no changes...

I've been diverted onto some updates on the tyre physics, because I needed to send Eric an update with new editor functions, but the new tyre physics were in an "intermediate" state. Now I've brought them to a driveable state, I can update Eric's dev version and I can get back to some patch things.

But I don't want to try and say what else will be in the patch.

Quote from Jacobys :

Dear Scawen, since everyone is waiting and expecting "wow" content I cannot do something else but stop and ask myself , what can we do to help other than suggesting and demanding improvements? What is our part other than contributing with the licence upgrade?

Yeah, normally the test patch threads are more lively and I keep adding more updates. This time, the shadows were tested, no serious bugs exist and then I've been diverted onto other things and there was also two weeks with the kids at home.

I'm back to work now so hopefully there should be more to test soon!

Quote from UnknownMaster21 :

( People are waiting more for fixes and addons for insim )

What are the things they are waiting for?

I don't follow all the discussions here, specially while I'm trying to focus on tyre physics, so I have probably missed things. But it would be good to know if InSim needs fixes or important updates.

Quote from FPVaaron :

It's a shame really, that major updates now days are otherwise "small fixes" from version 1.12 to version 1.13 in any other barely reputable game.

That is such a ridiculous statement that you have simply made up in your mind and then decided to say it, without comparing it with any actual facts to check what you are saying.

If you want to make a point, you are welcome to do so, positive or negative. But if you want to state such negative viewpoints, you should support it with some kind of evidence. Otherwise you are just spreading the fantasy that exists in your mind.

If you look at the new version announcements over the past few years, you will see major improvements.

Bringing up the Scirocco is quite pathetic really. The one specific announcement in all the years we have been developing Live for Speed, that hasn't yet been released. You think it's sensible to attack us for it, several years later, because just once we believed we could release something, then later discovered the tyre physics weren't ready for it. You have so little to support the negative view you want to spread for your own amusement, that you come up with this worn out old beating stick that has been used against us for so many years and we have issued so many apologies about. All because of one medium powered front wheel drive car, that people have blown up into some kind of Holy Grail of sim racing.

I've been working mainly on the tyre physics.

Eric has been working on tracks.

Sorry but there's not much more to say at this point.

Anyway, here is a new compatible test patch with some small improvements. Most notably the car shadows are no longer cast towards the light. See the attached screenshot in the Test Patch 0.6Q3 thread.
https://www.lfs.net/forum/thread/90208