• Welcome to PiBoSo Official Forum. Please login or sign up.
 

Alternative lean L/R behavior or gamepad?

Started by inTights, July 09, 2022, 08:40:37 PM

Previous topic - Next topic

inTights

Right now, you can lean (hang over the bike) and still turn the opposite way when moving the analog stick even just a little in counter direction. This is unnecessary because you wouldn't do that IRL. To turn the opposite way, you'd either get back central on the bike or hang over to the other side.

Instead the whole thing should sort of behave like direct lean at 0% (or maybe like 10%), and the whole range of the analog stick facilitate just altering the raceline/steering. It should also be a toggle because once you move into position, you stay there, so holding the button is redundant. You'd instead just tap it at braking zone and tap opposite at corner exit to get back on the bike, or tap 2x to hang over the other side. It would also free up the finger/thumb to do other things.

Conversely, when sitting central on the bike, movement should be much less erratic (already configured as low sensitivity as possible) to reduce twitchyness, as you wouldn't do these heavy leaning IRL either (but instead just move the bike under you in severe circumstance, which again could be triggered by tapping "lean"). Basically only heavily lean when L/R lean is "activated".

Thoughts?

doubledragoncc

The only problem I can see if Pib makes things work for gamepads only it will ruin it for others that use real controls. This is getting into the same way that Pilestone and others only make their games for gamepads. It is monopolization of the industry and no type of controller should be the only one the programmer makes the game or sim for. If you use a device that is limited in its function, that is your choice but dont expect others to suffer because of it. Piboso is making a simulator, not a game. If you have a console mentality then you will never get the full feeling of GPBikes. If you use a gamepad, you have to except its limitations, thats just a fact. If I could make systems so cheap for all I would, but it dang hard.

You can use programs likex360ce that allow you to change how your gamepad works, try that.

Just dont make itthat we all have to live with terrible gampad coding.

DD
GPBOC Live Streams: https://www.youtube.com/c/IASystemsComputerControls; i7 12700K 5.1GHz Z690 ASUS Strix Z690-A Mobo 32GB 3600MHz DDR4 RAM ASUS Strix RTX3080 OC 10GB DDR6X ASUS Ryujin 360 AOI Cooler ROG Thor 1200w PSU in ROG Helios Tower Case.

inTights

But it wouldnt affect your controller... it would be a translation or barrier for gamepad configuration. The physics is clearly box model based anyway.

Also, milestone doesnt make the games for gamepad either, or at least not properly. You're playing them with your controller as well

doubledragoncc

Actually you are a bit wrong on me using my system in milestone games. I have to use sowtware to trick their games into thinking my system is a gamepad and the steering in their games is hard coded with a big deadzone in the center as gamepad thumbstick movement is so small. I have to make an anti-deadzone and minimize the sensativity to use my system and it is not always perfect, plus the control amouts per view are off as they are made for gamepad too. It is a pain in the ass and thank godness for Piboso lol.

DD
GPBOC Live Streams: https://www.youtube.com/c/IASystemsComputerControls; i7 12700K 5.1GHz Z690 ASUS Strix Z690-A Mobo 32GB 3600MHz DDR4 RAM ASUS Strix RTX3080 OC 10GB DDR6X ASUS Ryujin 360 AOI Cooler ROG Thor 1200w PSU in ROG Helios Tower Case.

Vini

Doesn't make sense. And rider lean going to opposite direction of lean happens plenty irl. To a small degree on basically every corner exit and also in braking zones with a kink (Spielberg T2/3 or Mugello T1).

In any case, as DD has said there are tools that allow you to remap your controls in whatever strange way you want.
For example universal control remapper or even OpenTrack, which offers filters that act in a similar way to low direct lean values.

PiBoSo

Quote from: inTights on July 09, 2022, 08:40:37 PMConversely, when sitting central on the bike, movement should be much less erratic (already configured as low sensitivity as possible) to reduce twitchyness, as you wouldn't do these heavy leaning IRL either (but instead just move the bike under you in severe circumstance, which again could be triggered by tapping "lean"). Basically only heavily lean when L/R lean is "activated".

Thoughts?

An option to set the rider L/R lean to a value lower than 100% could be a good idea.

How did you configure the sensitivity as low as possible?
"La perfezione non è il nostro obiettivo, è la nostra tendenza".

doubledragoncc

I use my body motion unit so I have seperate rider to bike control and I use x360ce version 4.17 for games

DD
GPBOC Live Streams: https://www.youtube.com/c/IASystemsComputerControls; i7 12700K 5.1GHz Z690 ASUS Strix Z690-A Mobo 32GB 3600MHz DDR4 RAM ASUS Strix RTX3080 OC 10GB DDR6X ASUS Ryujin 360 AOI Cooler ROG Thor 1200w PSU in ROG Helios Tower Case.

inTights

July 13, 2022, 06:34:55 PM #7 Last Edit: July 13, 2022, 06:44:02 PM by inTights
Quote from: Vini on July 11, 2022, 04:42:09 PMDoesn't make sense. And rider lean going to opposite direction of lean happens plenty irl. To a small degree on basically every corner exit and also in braking zones with a kink

Small degree, exactly. And that is what im talking about.

The thumbstick is inherently a relative bad device, at least for this purpose, which is why even car games assist with steering. Console ego shooter players basically make patterns with them rather than being precise (at least not with both of them at the same time since we cant actually multi task contrary to popular believe), which is also why i think controlling lean with the other stick is futile as well.

Anyway maybe my idea isnt quite clear. Just setting sensitivity in a "strange way" isnt addressing the issue of subject matter which is 2 entities working together. It might work for car games, but simulating bikes is obviously much more complex. Im not sure to what extend this open tracker works, but unless it can listen to in-game events, im not sure how this would help

Think of it as toggled asymmetric sensitivity. When right lean is toggled (gamepad pleb user cornering mode, if you wanna call it that), 100% of the left range is as effective as maybe 10% (user configurable), and vice versa. This would allow to make those small degrees of adjustment for line correction etc. Its never going to be perfect, but i think this could help.

And In the same sense, i'd like the sensitivity on the straight to be less. Dont know how difficult its is to implement, but a curve editor to create a sensitivity like this would probably be great

https://cubic-bezier.com/#.03,.5,.99,-0.11

This way, around the center its rather smooth or slow with a strong output at the last 10% or so.

@piboso, linearity 0% and and gain 100%

PS: another option might be thumbstick speed sensitivity. basically moving the stick fast (for straight line mode) results in violent output and slow input in small movement. Like mouse cursor acceleration

Vini

Quote from: inTights on July 13, 2022, 06:34:55 PMAnd In the same sense, i'd like the sensitivity on the straight to be less. Dont know how difficult its is to implement, but a curve editor to create a sensitivity like this would probably be great

Yes, almost everything you want can already be accomplished through external tools like OpenTrack or Universal Input Remapper.
I suggest you look into them.
Alternatively, you could even write completely custom control code with relatively simple AutoHotKey scripts. I could point you in the right direction regarding this as well if you want.

inTights

Sorry for the bump, but this topic came to mind again and i feel like i'm not translating my idea properly. Mind you English is not my native language.

Quote from: Vini on July 11, 2022, 04:42:09 PMrider lean going to opposite direction of lean happens plenty irl.

yes indeed, and it is dictated by the physics model, and the interaction of the rider body and its mass/cog in conjunction with the bike. In the mode of the rider sitting [laterally] central on the bike. Moving the stick laterally at speed, the rider would lean with the bike (keeping inline), and at slow speed, lean the bike and move his weight opposite to compensate, or exiting a turn for that matter. Now to combine this with the traditional "direct input" method is what causes difficulty of controlling a computer model because obviously that feedback you have IRL doesn't exist in a simulation, and of course it's not even possible. A body tracker works to some degree, but thats another subject matter. The simulation must represent and allow for this behavior, which it doesn't right now.

Having the 2 modes of sitting central or laterally/hang off the bike, provides some of the feedback not just visually, but also in form of a representative input model depending on what the rider experiences when he sits central or hanging off. Again, external programs dont have access to the physics model, and neither would this affect body trackers. In fact, these 2 modes could be triggered by the body tracker, or contacts patches in the physical simulator.

Does this make more sense?