• Welcome to PiBoSo Official Forum. Please login or sign up.
 
March 28, 2024, 03:14:09 PM

News:

World Racing Series beta14 available! :)


Project Realistic Rig

Started by speedfr, December 08, 2017, 02:10:31 PM

Previous topic - Next topic

speedfr

April 16, 2018, 08:46:08 PM #90 Last Edit: April 16, 2018, 08:48:26 PM by speedfr
Coming back on my "feelings" and after retrying, i noticed that FFB stayed off on GP Bikes and i had what i wanted : the fact that the wheel constantly, not too strong but always there, try to go back to center and at the same time maintain the pressure on both sides give back this perfect feeling (at least to me).

So, as i was thinking before, FFB is not the thing to search for.
"Just" need to get a back to zero, just as springs does, but having a mechanical maintain of the handlebar, both ways, fighting against you, is perfect.
So, yes, a DC motor with a stereo potentiometers on the lean axis, one for the reading for the Leo Bodnar then the game, the other one to catch the angle and the movement. And in between some Arduino stuff that could sniff the exhaust of the potentiometer to tell the Dc motor what to do.
Need to find good reduction gear to fight a handlebar which is big, at least the double of a steering wheel.
That's the plan Stan !  ;D

And if possible, the Graal would be making this movement harder when braking from front, as it is in real life. Actually, the lever we use are regular ones, designed to accept a braking switch to turn the stop lights so, that will be the ignition that the Arduino will have to read in order to give the Dc motor the instructions to be stronger while braking. That way, the gyroscopic effect making the bike go strait instead of turning will be able to be in the pilot feelings.
Sure, the ones that rides knows that and how different the situation is when we have to "naturally" fight against that effect.

After that, i think, the main objective will be reached.
But with a system like that, my first post with my first draw becomes possible. All in all, trying to get to the same point by different way.

So, that's not finished   :)

Missing Gp500 (Microprose)  Testing EDTracker Pro on YT   R7-3800X/32Go/RX6800/1440p/Full WC

doubledragoncc

Not sure what you mean about centering. My systems auto center so you can let go of the bars and they are dead straight?

A lot of feeling comes rom how you sit. If the bars are just in front of you it wont work. You have to be leaning down on them. Everyone gets it wrong and has the bars too high. Use them like on a real bike not like a car system.

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.

speedfr

April 16, 2018, 09:12:12 PM #92 Last Edit: April 16, 2018, 09:13:45 PM by speedfr
You'll see on the video, I'm on my arms and wrists. My basin is at the same level as the handlebar almost.

But i know i don't explain well because of English. Will be a lot easier in French... But i will try to explain that to you through the video.
I don't care actually that it centers back, it just that by doing that, when you push to the left, it forces to the right, giving me something to countersteer, meaning putting a force on the opposite side on my turn.
As on every bike.
And springs are good in princip but not in reality (to me). They are not linear in the way they fight you.
And retrying tonight with the FFB Wheel made the point. Its hard enough and smooth at the same time but it doesn't shake (what i called wiggle earlier but that might not be the word) and so the bike is going nicely round on the turn, like it supposed to be and exactly the way it is in real life.

And Allan, you know it, when you start a turn and brake at the same time, its harder to turn the bike, in your mind nothing to think about, you automatically adapt to that turn and know inside you that the supplementary effort is due to turn+braking. That's magic and the best braking i ever did on bike, on Ledenon for example (one more time, in real, with my 350RDLC, many many years ago) on a triple left, most on the brake fighting to turn and keep the trajectoire, that's what gives you the most thrill, specially when that means keeping your place or trying to catch one on this special curves.

So, that's what i called Graal to me. Having this feeling in GP Bikes whitout risk anymore since anyway, too old and too parents to have time/money and capacities to do it for real. And tired of racing "by procuration" with MM, VR and Zarco !!!
Missing Gp500 (Microprose)  Testing EDTracker Pro on YT   R7-3800X/32Go/RX6800/1440p/Full WC

poumpouny

FFB in GP Bikes is far more  than "centering the handlebar" and I'm not even talking about Direct Steer FFB ;D. i'm sure you let the "autocenter" option of the wheel on when you tryed. Turn that off (i don't even know who need this centering thing but they keep it by default". All you're looking for (steering get hard when braking, woobling handle bar on hard exit/hard braking ...) is already handled by GPbikes FFB. I Think you're best solution is to fund a used G25 wheel and then just link your steer axis to the wheel (easier and cheaper that recreating a decent FFB from scratch). I'm doing (still in a long WIP) exactly the opposite of what youre doing, i'm making a handle bar that i could just "Attach" To my G29 Wheel ( a detacheable thing with what i will be swap between car and bike games).

speedfr

Hello Poumpouny,

i just read your post and there's a way that i'm gonna explore and that what i described earlier and what you say in here, i'm first start from a old FFB wheel and attach it to the existing axis of my rig. Then see what happen.
But as far as i tried FFB on the wheel playing GP Bikes (full FFB not just the centering) i didn't like the all-shake stuff, probably need to be tune but it's not what i'm searching for the most.
The centering, one more time, i don't care, but the fact that it forces against you when leaning, that thing i want. And i'm gonna do.
But after that, yes, feeling harder when braking to lean, that will be an outcome since it's the main feeling to turn on a bike while braking.

And starting from a wheel that i "operate" to extract what i need is a good and easy idea, but in middle terms, i'd like to reproduce the rig for my youngest son ( they both are motorcycle riders too, like Daddy !!), who badly wants to ride too on GP Bikes with other thing than just the pad. So, it would be nice to understand and be able to reproduce. And if it's not that complex, give it to DD so he can add it to his HSx system.

At the same time, i hope you gonna reach your point and get what you want, but it's been 25 yrs almost that i'm playing Motorcycle game (GP500 mostly) on a Pc so i decided that at 50 yrs old, i could make my own complete stuff, i keep my old wheel safe and can use it if necessary but i don't want a "middle" solution. But i understand your choice and whish its gonna work nicely as you want it.

i found old wheel for a few € (20) and sure it's cheap. But an Arduino like plus a Mabushi 540 and the gearing won't cost much, the potentiometer need to be a stereo so that cost nothing too, at the end, i don't think it will cost a lot more than a used wheel and the main work, the piceless work will be the correct programming of the Arduino to monitor and pilot the Dc Motor.

Thank you for your post and have a good day !! (or night...  ;D)


Missing Gp500 (Microprose)  Testing EDTracker Pro on YT   R7-3800X/32Go/RX6800/1440p/Full WC

doubledragoncc

I have been toying with a unit to mount on car systems too. Here is my original setup for development from 2010 with G27

https://www.youtube.com/v/JDhvXkO_vZA

If I remember right I had it set to 40 degrees rotation not sure.

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.

speedfr

i've seen that for a long time, that's the first "contact" with IASystems.
I did the same with a big wood stick to give it a try, that was... special !!

This week-end, i'll go pick up the used wheel i found and i start the work. I'll tell you after.  :)


Missing Gp500 (Microprose)  Testing EDTracker Pro on YT   R7-3800X/32Go/RX6800/1440p/Full WC

poumpouny

The problem is that in normal mode, the FFB is attached to the lean input/output and every bike movement is sent to the wheel, that's why it is unnaturally shaky, in order to get the right feeling on the handle bar (like in real life) it have to be attached to the handlebar steering, and the only way for you to do such is to use Direct Steer with Angle (DSA). in DSA mode the handle bar only "shake" when you handlebar shake or wooble like in real life, and you have the handlebar countering you when you lean, and you really have to counter steer on each corner. But riding in DSA/DST mode is another challenge ......

Vini

April 17, 2018, 01:51:02 PM #98 Last Edit: April 17, 2018, 01:55:22 PM by Vini
while it may work ok with a wheel, DSA mode wouldn't feel right with a (swinging) handlebar controller because the handlebar will no longer represent the bike lean angle, ruining the immersion (it would remain at 0 lean most of the time).
as i said from the start, if you want to do it properly you need two axis, one for lean angle (output only) and one in DSA mode with GPB's FFB.
so you would need two wheel motors.

racer73

April 17, 2018, 02:55:00 PM #99 Last Edit: April 17, 2018, 07:48:21 PM by racer73
Hi all,
  I've been lurking here for a couple of months and been impressed by the incredible talents on display here. I'm currently waiting on a rig from dd and am very excited to try it out. I decided to chime in here because I may have a perspective that is useful to the current discussion. I have had no experience whatsoever with racing games or simulators. I am a motorcycle racer and due to our long winters here in Canada I've been interested in putting together a motorcycle simulator that will provide me some practice and diversion during the non-racing seasons. I'm sure there are plenty of racers on the forum who have different takes on this than me. I'm just throwing in my 2 cents.

In response to speedfr's discussion relating to "try to go back to center and at the same time maintain the pressure on both sides give back this perfect feeling"

I'll use resistance to counter-steer (rcs) to describe that feeling that opposes counter-steer or the tendency for the bike to stay moving in the direction it is currently going. That feeling is strongest when you're initiating a change to the direction of the bike. The greater the change needed the more input force needs to be applied.  As you lean the bike over and reach the desired lean angle that push-back feel is greatly diminished. So much so that once the desired lean angle is reached the amount of rider input into the handlebars should be 0. The bike will now follow the new line that was set. The bike needs no additional input whatsoever to keep the lean and therefore the rcs feel should be 0. When riders describe the need to keep pressure on the handlebars when leaned over it's often the case that they are unintentionally loading the other side of the handlebar. Then when exiting the turn and standing the bike back up the rcs felt on the "outside" end of the handlebar is strong and then diminishes as the bike returns to an upright position.
rcs will be strongest when the desired lean angle (dla) is the greatest distance from the current lean angle (cla).  rcs = abs(cla-dla)  The rcs should be felt on the side of the handlebar that is initiating the counter-steer (i.e. the pushing hand)

To me this is VERY different than a steering wheel.  When initiating a turn with a steering wheel you generally feel the same pressure on the wheel and once you're in the turn you constantly feel the same pressure so when you release the wheel or ease up on your grip the wheel wants to return to the center position.

To simulate rcs you would need some kind of mechanical means to have the pressure be greatest at initiation and then diminish as the chosen lean angle is approached. I'm sure dd or speedfr could come up with something cool.

Ok... back to my lurker position. ;)


Vini

You need an electric motor that is controlled by GPB' force-feedback in DSA mode for realistic behaviour. A purely mechanical system would not work because it cannot take into account the grip levels of the tyres (e.g. the rear sliding and introducing additional forces on the steering).

speedfr

April 17, 2018, 09:20:41 PM #101 Last Edit: April 17, 2018, 09:41:24 PM by speedfr
DSA, DST, i tried them, couldn't do more than 5m without falling, nightmare, but simple : i have no idea of what it means in terms of interaction between rider and the bike.
But i understand that's by using the "real" torque on the axle of the bike.
And if it's that a copy of reality, my first draw should correspond but as i said, it's "another world", i wanted to have a handlebar that works, not as real, but that could give me the possibility to ride GP bikes and have fun. i'm almost there.
Language doesn't help and to be more politically correct, i would say that if i was better in English i surely understand more than what i can sneak here and there.
Language about all this thing, specially the DSA-DST is obviously technic and my English is "everyday" not technic. and worse, sites like Google Translate are super bad in technical language so ...
it's "my" problem so i'll finally find out one day, i use my mind to glue the puzzle of what i understand mostly...
i was maried 15y to a girl from Mill Valley (near SFO) so i can speak and read. But i never had discussion about epicycloidal train or the setup of a distribution belt.... I'm sure you all get it  ;D i didn't learn technical words but she gaves me two beautiful kids that rides like Daddy !  :)

In long terms, at least on the princip side, i will work on my first idea after, specially practicing GP Bikes with another system ie my bad copy of DD's invention.  ;)

So yes, Vini, two motors for the both axles, and one that blocks the vertical one around 35-40km/h, only there to block the steering and shake as a wooble when Melandri hits the 5 and 6 gear (still did that in Aragon, incredible how he set his bike up, dont get it... ).
And the other one to send you the gyroscopic effect plus gravity on leaning, and all the effect that a bike gives you.

It's not in the idea to change anything about GP Bike, it's to try the best to get the closest possible to what a bike gives you when you ride it, so i could focus on trajectoire and braking, i mean on what is interesting.
When you forget about UI and controller, that mean they are adapted to what you do. That's my not-secret-at-all goal.

If FFB is able to send that back to the rider then ok. But if FFB is like they are in car games mostly shaking to gives you sensations without any correspondance to the reality, i'll forget. But Poumpouny seems to say it's really that... we'll see maybe one day.

Quote from: racer73 on April 17, 2018, 02:55:00 PM
Hi all,
  ... I'm currently waiting on a rig from dd and am very excited to try it out.
Great idea !!

Quote from: racer73 on April 17, 2018, 02:55:00 PM
I'm just throwing in my 2 cents.
This could be a fortune actually, i always thought that more than one brain on a problem is not a solution but THE best solution. So, stick your ass in here !!

Quote from: racer73 on April 17, 2018, 02:55:00 PM
So much so that once the desired lean angle is reached the amount of rider input into the handlebars should be 0. .... The bike will now follow the new line that was set. The bike needs no additional input whatsoever to keep the lean and therefore the rcs feel should be 0
i cannot be more ok with you. Perfect and so real.

Quote from: racer73 on April 17, 2018, 02:55:00 PM
To simulate rcs you would need some kind of mechanical means to have the pressure be greatest at initiation and then diminish as the chosen lean angle is approached. I'm sure dd or speedfr could come up with something cool.
God, i whish...

Hello racer73,

Welcome aboard, happy that your first message falls in here.
Always happy to see a Canuck (!!!) around, one day maybe, i'll tell you about the French Canadian Beaver Licker team and the Beaver's Motorcycle Club, if my pal from Gore Bay shows up...

Most of the crazy guys here are riding IRL and for some, pads are ok, but for the older like me, we need more. If you bought an HS3 from DD, you'll soon know what i mean. But even if DD's work is great and unique, i'm sure you're gonna miss some sensation.
Anyway, even with the best simulator ever, you'll never have gravity, inertia, gyroscopic and others even if they are exact IN the software, for us it's not "the same" and it won't ever be.
it's a Sisyphe story, trying to reach, to move towards an ideal horizon...

Hope to see you on track online soon.  8)


Missing Gp500 (Microprose)  Testing EDTracker Pro on YT   R7-3800X/32Go/RX6800/1440p/Full WC

racer73

Thanks speedfr for the kind welcome.  We Canadians are certainly fond of our beaver ;) so I hope to hear about "the French Canadian Beaver Licker team and the Beaver's Motorcycle Club" sometime.

As for the simulator... I don't have any unrealistic expectations.  Things like gravity, inertia, and gyroscopic effects are extremely difficult or impossible to duplicate (at least within our budgets).  Winters here are too long and the riding season too short so having a nice system that gives me a general realistic feel where I can practice things like hitting my reference markers as well as some fun racing will fulfill my needs quite nicely. 

With regard to that rcs feel... I was thinking about it a bit more.  There used to be fitness machines called Nautilus that had a gear system whereby at the start and end of the movement (where presumably the user was weakest) the resistance was the least.  And as they reached the center of the movement the load would become the greatest.  I'm wondering if something like that could be used in this application.  However we would want the greatest resistance at the start of the counter-steer.

My general feeling is that when counter-steering I want to feel that rcs the most when I'm pushing the hardest and as I decrease my input the rcs should diminish proportionately.  Almost like it's reflecting my input pressure back at me. (But a bit less so that the bars actually move. ;) )

And as you mention, much of this seems like a Sisyphean task, but every once and a while we get a boulder to the top.   :)


speedfr

April 29, 2018, 08:32:05 PM #103 Last Edit: April 29, 2018, 08:39:42 PM by speedfr
Hello riders !

Here is a little update.

Adding the FFB parts from the wheel i disassembly, the "rig" is almost finished but the main is done.
I changed the seat for a real bike seat, pictures coming, as the other one was stopping blood or nerves on my both legs and it started to be dangerous.

Yesterday night, trying things, i went to the "Onboard" server of Vini (Vin97) and thought that the MotoGP 2017 were going to be unridable, absolutly too powerfull, cuz trying the MMX500 or the Moto2 was super hard.
I did a lot of Moto3, that was fun an easy but not fast enough as you all know.
So i went there without any special hopes, finding the riding easier with the handlebar but still hard to place my bike wherever i wanted too.

And then the light went on me.
This light is named Vin97, he gave me good tips about setting the view, which i already try to work on but without great success.
But now, it's completly different.
This Vin97 changed my GP Bikes experience by his know-how and i had a super evening riding MotoGP on Brno, that was a blast.
I've reached the GP500 feeling that i had 20 yrs ago.
As he says (and i was desperate to reach), now it's "easy" to catch "Lorenzo lines" and lick the corners. Great, great, great.

@Vini ==> Danke von ganzem Herzen.

I wanted to thank him again so much, and for that i taped on OBS some of the runs.
But OBS did an update just before and maybe my bad, but all my settings went away...
So the video is in very poor quality, 25fps on 1080p. I tried all afternoon to cut this video for the best part and "glue" it together, VLC doesn't work well enough (delay in the sound on the output video), tried several "free" software, that mostly doesn't work, even with a flv converted to mp4, so i finally dropped and decided to put the raw video like it is.
For the record.
And at the end, there's a special Easter Egg...

Next video will be cleaner, as the other ones i've done on my YT channel, and it will come with full explanation.

@DD : i ride without any springs now, FFB is off but the steering system constantly fight me except when i have a line of angle, so its a lot more "realistic" than springs.
I'll show you on the coming video.
Now, it will be nice to find out how to hide this system under the steering head but that's to have something to do at night otherwise, i just sleep, pffff....  ;D

So far, i noticed this 3 points :

_Why the MMX500 have trouble gearing down, it looks like a "gear saver" that refuse to gear down too fast...  :o

_Why the clutch on Moto 3 is asking you to constantly accelerate otherwise you stall...  :-X

_And the last but not least, why can't we split ambiant sound (other bike) from our bike ? Very disturbing, and that's the next thing with 64bits managment that i'm gonna stick in the whishlist.
(and the first one who answer me "because its like that in real life" i will tell him that in real life you feel your RPM by the vibration of the bike, here it doesn't vibrate so it's kind of "guessing" situation).

Next will be to ask a guy who know how to design so i could order a Bee frame (Black and Yellow) with number 40 on top and the brand of my bike : Sapetoku
(i'll explain later why  ;D but for the French guy who are old enough, just remenber Gaston Lagaffe)


Journey at Vin97's Academy !

https://www.youtube.com/v/HqwTXQCnE3M
Missing Gp500 (Microprose)  Testing EDTracker Pro on YT   R7-3800X/32Go/RX6800/1440p/Full WC

speedfr

@DD : After what you wrote in the GP Bikes video topic, here is what i had to do because regular FFB is made for a wheel and our handlebars are a lot wider so it forces like hell and is not strong enough. In addition, the DC motor gets super mega hot and i'm afraid everything is going to smoke dark...

So, since we have 3D Printers, i went to change the gearing for the DC Motor to the main axe and try to give it more forces to fight the handlebar torque.

1. The white is the original, Module 0.75, one 22, one 100 that gives torque to a end one, module 0.75 in 22 teeth.



2. The one i did are module 0.8 one 38 et one 70, and the center one is still 0.75 module in 22 teeth also.



3. The original



4. How it looks without the FFB gearing



5. After mounting my own production   :)



Ans so at the end, the FFB is stronger and you feel it a lot more. I set the dynamic down  and the static more agressiv, and that gives good sensations, specially when braking while entering a curve.
I'll do the video, but not until minimum one week cuz i'm waiting for some parts first specially Yellow 3D filament, but with this kind of "Gruyere" week we have, it takes longer.
Missing Gp500 (Microprose)  Testing EDTracker Pro on YT   R7-3800X/32Go/RX6800/1440p/Full WC