I'm quite sure it's not as simple as changing one variable in one line of code. If it were, Scawen would have already done it. However, the suggestion is reasonable and we need support from more people in community.
Anyway, start of April is very close and I'd rather Scawen gets his hands on tyre physics, than to get sidetracked again by these "small and quick updates". There's an infinite amount of them and there will allways be, can you "just do this one" before..
Trucks usually have many gears because the engine rpm range where it has best efficiency is very short. Therefore, many gears are required for the truck to cover wide moving speed range. F1 cars have 8 gears now days. Most modern road cars have up to 6 gears, some supercars have up to 7 gears. So, stats wise 6 is enough, but some odd balls need more.
Given the variety of mods, there seems to be the need for more than 6 gears to replicate transmission systems. My oppinion is that allowing 8 gears should cover mostly everything, but it's up to Scawen to decide.
Given the variety of mods, there seems to be the need for more than 6 gears to replicate transmission systems. My oppinion is that allowing 8 gears should cover mostly everything, but it's up to Scawen to decide.
These pedals have their own usb game cotroller, so if you plug them directly into a PC, you should see a separate controller in LFS. If you were to plug them into the TM wheelbase then everything is detected as one controller. You can give it a try, although I don't recommend as you will only have 10bit pedals. To utilize full 12bit res of your pedals, you have to plug them in PC. Especialy now that LFS supports high res controllers (patch E9 and above), this is a no brainer.
The trick for getting them to work is not to select any controller as a primary one. You just have to bind all axis to your LFS controls. You may need to locate TM controller profiles in LFS folder and delete them first.
The trick for getting them to work is not to select any controller as a primary one. You just have to bind all axis to your LFS controls. You may need to locate TM controller profiles in LFS folder and delete them first.
In a normal car, this kind of smooth start is a real art. The more you drive the smoother you get. Once you start paying for your own fuel, then you also start to learn all possible ways of gentle driving and fuel saving.
What are the units of values in Y-axis? This is a prediction of what exactly? 
Well, yeah, it is what it is at the moment.
I think we should let Scawen to start working on other things. The clutch bite point is not of a big priority anyway and it's not related to this test patch. It's an old legacy thing, we can disscus it in another thread for "launch control".
I think we should let Scawen to start working on other things. The clutch bite point is not of a big priority anyway and it's not related to this test patch. It's an old legacy thing, we can disscus it in another thread for "launch control".
Just change the upper limit to 100%, the default value is 65%. That should streach the bite point a little bit.
Last edited by rane_nbg, .
For mouse steering all you need is sensitivity adjustment.That you have already with multipliers, just allow range of values like from 0.1 to 10.0 at least. Mouse is different than controller axis, no need to unify it. Its sensitivity depends on many things, DPI, screen resolution, windows sensitivity slider. Having independent X,Y mouse multipliers in LFS is more than enough.
Last edited by rane_nbg, .
Tnx for these skins, both look very good. I'd pick black if I had to choose only one.
Rather than thinking what's next, I'd focus my attention to the updates that are already cooking in the pot. Scawen mentioned that after this last patch E9 he released today, he will work on tire physics, so we're closer than ever to the big one.
But if I had to play a fortune teller a bit, I have to say the next big thing would be track editor, not counting all sorts of small quality of life updates for LFS that will take place along side it. It's just a guess.
But if I had to play a fortune teller a bit, I have to say the next big thing would be track editor, not counting all sorts of small quality of life updates for LFS that will take place along side it. It's just a guess.
It's not because Xbox controller reports the 2 triggers under a single Z axis. We explained in more details above.
LFS has the option to combine 2 axis into one, but not the vice versa.
LFS has the option to combine 2 axis into one, but not the vice versa.
It has already been suggested before and it's in the summed up list. This is still far away due to level of difficulty on the game engine side.
First of all, thank you very much Scawen for considering my suggestion, and also big thanks for asking us for our thoughts about it.
As someone who developed an FFB wheel firmware and also my own GUI, where I had to deal with exactly these questions how to make it transparent for the user and how to make it simple and intuitive to use.
It seems that working in % of axis range is best for setting calibration ranges. So instead of having +-1000 I would go with +-100%, with a step of 1% adjustment, which should be ok.
About axis sensitivity (non-linearity), I would leave this out and simply keep a linear axis between the user-defined min and max limits in %. If you really want you can add a 3rd point - center, but not really necessary at all. Can be done with a 3rd party software like DxTweak.
In principle, whether you choose +-1, +-100% or +-1000 I don't mind, as long as in this range there are as many steps as controller reports by HID.
It would be good to see somewhere the raw axis value as reported by DirectInput. It should fit into graphics in FF/axes, I got used to it and it is simple and effective, so I don't think any changes are needed there. So instead of showing a momentary value of the axis between -1000 to 1000, I would put there a raw HID value for that axis, like 0-65535, or 0-4095, 0-1023, 0-255. With this, a user can directly see the resolution of his controller for each axis and at the same time an axis value.
As someone who developed an FFB wheel firmware and also my own GUI, where I had to deal with exactly these questions how to make it transparent for the user and how to make it simple and intuitive to use.
It seems that working in % of axis range is best for setting calibration ranges. So instead of having +-1000 I would go with +-100%, with a step of 1% adjustment, which should be ok.
About axis sensitivity (non-linearity), I would leave this out and simply keep a linear axis between the user-defined min and max limits in %. If you really want you can add a 3rd point - center, but not really necessary at all. Can be done with a 3rd party software like DxTweak.
In principle, whether you choose +-1, +-100% or +-1000 I don't mind, as long as in this range there are as many steps as controller reports by HID.
It would be good to see somewhere the raw axis value as reported by DirectInput. It should fit into graphics in FF/axes, I got used to it and it is simple and effective, so I don't think any changes are needed there. So instead of showing a momentary value of the axis between -1000 to 1000, I would put there a raw HID value for that axis, like 0-65535, or 0-4095, 0-1023, 0-255. With this, a user can directly see the resolution of his controller for each axis and at the same time an axis value.
We did, it was nice, then it was not nice due to off-topic and personal disputes, and then it was closed.
Hmm, not sure man, I'd say probably no cars with TC in S1.
You're quite new here, but just browse around a forum and you will find an infinite amount of suggestions, discussions, and whatever about LFS updates and future development.
Our guy is Scawen - game developer, team lead, manager, PR, tech support, director of LFS, name it however you want, the guy does it all. Eric is a 3D artist, and creator of cars (demo, S1, S2) and all tracks. Victor is working on the internet side of things, LFS world, and more.
So you see, a very small independent team, but they do great stuff.
Our guy is Scawen - game developer, team lead, manager, PR, tech support, director of LFS, name it however you want, the guy does it all. Eric is a 3D artist, and creator of cars (demo, S1, S2) and all tracks. Victor is working on the internet side of things, LFS world, and more.
So you see, a very small independent team, but they do great stuff.
Can't really name them all, but many in S2 and much more in S3.
We have suggested this many many years ago and much more, but there is only one developer doing all these things, so it's going there eventually. At least we hope
You do realize that there exists an entire car mods system with 800+ cars and objects in S3?
Welcome to the club.
What I'm asking is equivalent to: my monitor is Full HD and supports 1920x1080. In the game, I can set screen res only to 1280x720, so I ask to implement support for whatever my monitor is maximally capable of -just in the game controller world.
Tnx for support.
Here's a simple explanation of how the two are different.
Let's say your car is stationary and the wheel is centered, you are not touching it at all. Then, from the side comes another car at a moderate speed that hits your car. Physics does its magic, calculates all the forces, and sends the FFB to a wheel. The resolution of that FFB signal has nothing to do with the game controller, it's inherent to LFS physics. Now, what the FFB wheel will do with that signal is another thing and it is very important for a final FFB feel. In the same way, you would hear a better sound with better and more powerful speakers. The song is the same, but what you hear in the end depends on the device you listen to.
Now, let's say FFB is set to 0. You drive a car around, there the resolution of the steering axis dictates what will be the minimal input you can give that the game can still sense and act upon. LFS can still send the full res of FFB even if the steering axis is very low res. But we talk here about how sensitive the car is to the steering input. It may be ok for a GTi car to have 2000 steps, but something more serious like a BF1 or similar will benefit from utilizing the full available res of the controller (16bit usually for wheel, 8 or 10bit for gamepads).
FFB consists of many effects (forces), some are dependent on the steering input, some not. For example, track shape and all imperfections get picked up by a tire and suspension system, and that causes some forces to act on a steering column and a steering wheel. Whether you move it or not, they're gonna be there as long as the car is moving.
Here's a simple explanation of how the two are different.
Let's say your car is stationary and the wheel is centered, you are not touching it at all. Then, from the side comes another car at a moderate speed that hits your car. Physics does its magic, calculates all the forces, and sends the FFB to a wheel. The resolution of that FFB signal has nothing to do with the game controller, it's inherent to LFS physics. Now, what the FFB wheel will do with that signal is another thing and it is very important for a final FFB feel. In the same way, you would hear a better sound with better and more powerful speakers. The song is the same, but what you hear in the end depends on the device you listen to.
Now, let's say FFB is set to 0. You drive a car around, there the resolution of the steering axis dictates what will be the minimal input you can give that the game can still sense and act upon. LFS can still send the full res of FFB even if the steering axis is very low res. But we talk here about how sensitive the car is to the steering input. It may be ok for a GTi car to have 2000 steps, but something more serious like a BF1 or similar will benefit from utilizing the full available res of the controller (16bit usually for wheel, 8 or 10bit for gamepads).
FFB consists of many effects (forces), some are dependent on the steering input, some not. For example, track shape and all imperfections get picked up by a tire and suspension system, and that causes some forces to act on a steering column and a steering wheel. Whether you move it or not, they're gonna be there as long as the car is moving.
FFB resolution is fine, Scawen increased it to a full range a while ago.
It's about controller input resolution, it's fixed in LFS at 2000.
It's about controller input resolution, it's fixed in LFS at 2000.
One note, once you set keyboard stabilized controls, the in-car steering wheel can actually do these sub-pixel rotations, so the issue is only for game controllers like FFB wheels and some high-end joysticks.
No, unfortunately it's not. You can have a look in axis calibration range in lfs. It is +-1000, giving only 2000 for all axis. This is fine for pedal axis, but for steering axis is a major no-no. I didn't escalate this earlier, because it is not so urgent with with curent physics. But once we get new phys, I will insist on it until I see it implemented.
If you pay attention to the incar steering wheel, you will notice quite a jump when you move wheel with as tiny steps as possible. Compare it with for example any other sim where you will see tiny sub pixel angular movements that correspond to full 16bit steering axis if the wheel supports it (logi, thrustmuster do, some DD's have even more than 16bits).
Some math for minimal angular step: 900deg/2000=0.45deg, while 900/65536 is less than 0.014deg.
If you pay attention to the incar steering wheel, you will notice quite a jump when you move wheel with as tiny steps as possible. Compare it with for example any other sim where you will see tiny sub pixel angular movements that correspond to full 16bit steering axis if the wheel supports it (logi, thrustmuster do, some DD's have even more than 16bits).
Some math for minimal angular step: 900deg/2000=0.45deg, while 900/65536 is less than 0.014deg.
FGED GREDG RDFGDR GSFDG