Well just one example for Rift + LFS + motion chair.

Let's say you are driving up a 20 degree slope at constant speed. Your chair, and steering wheel, should lean back 20 degrees.

Now, if you look "forward" with your head relative to the car body (as if you were a stiff crash test dummy) your head will be looking UP 20 degrees relative to the real world. So the Rift camera, regardless of whether it is mounted on the seat on on a nearby tripod, will think you are looking UP 20 degrees. So now, the way LFS + Rift works at the moment, your view will be pitched up 20 degrees relative to the car. And that will be UP 40 degrees relative to the in-game world, which is wrong. So although your head doesn't move relative to the steering wheel, you would now have to look down 20 degrees (i.e. move your head to be level relative to the real world) to see it as it is supposed to be.

EDIT: It seems to me the Rift camera should be mounted on the same platform as the seat and wheel so that it can detect your x, y, z movements correctly, but I imagine vibrations affecting the camera. Anyway, if the vibrations aren't a problem, can't LFS solve this problem by showing the view relative to the horizontal plane? So in the example above, the in-game viewpoint would use the YAW of the car but not PITCH and ROLL, in a special "motion simulator type 1" mode? Adding the pitch and roll of the actual headset, so your view is once again aligned with the car body? Though this will only work if the motion seat pitch and roll matches the in-game car pitch and roll exactly (and doesn't allow for extra pitch and roll often added by motion simulators in an attempt to recreate acceleration forces).

Or will there be a problem as the in-game car tilts to go up the hill but the motion chair has lag in its response...? Or are there still more problems, as when you tilt back the camera goes higher than you, so the Rift drivers think you have got lower in your seat...?

Either way, these problems could be solved by the Rift drivers themselves, if indeed they know the pitch and roll of the motion chair, by use of an additional hardware device mounted on the chair and doing a subtraction as mentioned by sinbad:
https://www.lfsforum.net/showt ... php?p=1869220#post1869220

Thanks for that Scawen, that does now make sense. It's more complicated than I first thought. I will do some testing with my rig because I'm not sure mine tilts back if you go up a slope. I think it only takes movement from the car which I know is a contradiction but I believe it moves relevant to the cars cornering, braking, acceleration, engine vibration and any road bumps (chair up & down) only. They are the only options for turning forces up or down.

I'm going to try keeping a constant speed and picking a track which has a nice upslope (south city upramp?) and have a friend watch the chair & see if it leans back. I guess if it doesn't lean back then it will match the rift but wouldn't be as realistic and maybe cause motion sickness? If it does lean back then I must be counteracting the pitch discrepancy myself which as you say is wrong.

I am interested in what mounting the rift camera with the chair would do... Although I think people have already had problems with the camera if it is banged about from their steering wheel on the same desk which makes it lose tracking.

Unless motion is turned off at low speed, you can simply park up on the slope (constant speed can be zero mph), try parking with the car pointing in different directions on the steepest part of the Kyoto Oval.

Quote from Scawen :

Well just one example for Rift + LFS + motion chair.

Let's say you are driving up a 20 degree slope at constant speed. Your chair, and steering wheel, should lean back 20 degrees.

Now, if you look "forward" with your head relative to the car body (as if you were a stiff crash test dummy) your head will be looking UP 20 degrees relative to the real world. So the Rift camera, regardless of whether it is mounted on the seat on on a nearby tripod, will think you are looking UP 20 degrees. So now, the way LFS + Rift works at the moment, your view will be pitched up 20 degrees relative to the car. And that will be UP 40 degrees relative to the in-game world, which is wrong. So although your head doesn't move relative to the steering wheel, you would now have to look down 20 degrees (i.e. move your head to be level relative to the real world) to see it as it is supposed to be.

My highlight, and is that sentence really correct?

If the camera is mounted on the chair, it will also be tilted 20 degrees and see you as looking straight ahead. If it is mounted on a tripod it will see you as tilted 20 degrees.

It's a matter of references, really. with the camera on a tripod the real gravity will match ingame gravity in your case, and simulating forces by tilting would be very difficult. With the camera on the chair you could ONLY simulate accelerations, as camera "down" direction would alway be the virtual world "down" direction.

This stuff is hard to explain by typing.... :-P

No, because I think you missed one point. The rift actually senses the direction of gravity.

The down vector, as seen by the rift software, is not down relative to the camera's own vertical axis, but the real world's down direction according to the direction of gravity, detected by an accelerometer.

So, for example if you attach a rift and a rift camera rigidly to a board, then tilt that board, something DOES happen.

Ah, that changes stuff!

After driving a bit more today, it would be really nice to have steering animations going up to 1080 degrees on all street cars. I just can't get used to driving with 720 anymore, and using anything higher feels way too weird in VR as the wheel rotation wouldn't match after that.

Assetto Corsa way of doing steering animations would be the best I think, give users an option to set the range of the animation to match their real wheel, anything from 180 to 1080 degrees.

You have a T500 RS? Its the only steering wheel which supports 1080.

I think it's pretty lorry driving alike, but whatever floats your boat

Yes. Frex also does 1080 and I'm pretty sure the powerful direct drive servo wheels (Bodnar/SimSteering) also support it, or more.

Majority of real cars have steering between 950-1200 degrees, old BMW's even have up to 1440. It's nothing special in real life, if anything, 720 degrees of LFS street cars is unrealistic.

Matrixi, I know real cars have a large degree of turning but if your racing round a track they reduce the faster you go (don't ask me how it works but they do). So at 100mph going round a corner you only need to turn the wheel a small amount, yet at 2mph you need to turn the wheel 2 and a half times to park.

I found if you set a racing sim to 1080 for example you just end up having to turn the wheel like a madman at 100mph which isn't realistic!

Quote from Matrixi :

old BMW's even have up to 1440. It's nothing special in real life, if anything, 720 degrees of LFS street cars is unrealistic.

Those numbers looks like screen resolutions, we need a 4K wheel

re. dk2 with motion

I'm using a motion setup with the dk2 with extremely good results
the camera is in a fixed position on a wall

my platform pitches / rear traction loss
+seat motion on top of the platform

home setups such as mine do not make high movements! just enough to feel the onset of gforces + position indication

the dk2 movement actually adds to the motion effect
-chair moving slightly to the left to simulate ones body being moved to the side in relation to the wheel shows in the dk2 & gives good additional visual clues


something like the redbull simulator would obviously need a VR track cam mounted to the actual platform

http://youtu.be/rE-Fge3gN9w

Quote from Alric :

Matrixi, I know real cars have a large degree of turning but if your racing round a track they reduce the faster you go (don't ask me how it works but they do). So at 100mph going round a corner you only need to turn the wheel a small amount, yet at 2mph you need to turn the wheel 2 and a half times to park.

Unless you drive something like a new Lexus or BMW with optional dynamic power steering, that isn't the case at all. Standard rack and pinion or recirculating ball power steering systems are linear, only the steering resistance might change as you drive faster. I've driven a 5-series E60 BMW with one of those dynamic power steering things and it was horrible, I had no idea what direction the front wheels were pointing at.

Quote from Alric :

I found if you set a racing sim to 1080 for example you just end up having to turn the wheel like a madman at 100mph which isn't realistic!

Only games where I've noticed this to happen, are arcade stuff like Dirts and Grids. If a sim has been programmed properly, the steering will be linear at all times and you don't need to turn your wheel at higher speeds more than you would in real life.

Anyway, this is heading offtopic. It was just a simple suggestion I wanted to make for those who want as much realism as possible when they drive in LFS.

I suppose one easy way to support steering animations over 720 degrees, would be to enable 1080 degrees of animation with the wheel visible but driver hidden. Not ideal for VR as it would feel like being a ghost driver, but no other way to solve it unless Eric is willing to kick up completely new steering animations.

Quote from Alric :

Matrixi, I know real cars have a large degree of turning but if your racing round a track they reduce the faster you go (don't ask me how it works but they do). So at 100mph going round a corner you only need to turn the wheel a small amount, yet at 2mph you need to turn the wheel 2 and a half times to park.

I found if you set a racing sim to 1080 for example you just end up having to turn the wheel like a madman at 100mph which isn't realistic!

Speed sensitive steering ratio? (Not to be confused with speed sensitive steering/power steering.) I think you're slightly off the mark with that one.

I think if such a feature did exist (doubtful that it does with mechanical steering- it might, but I'd expect it's more easily feasible with steer-by-wire systems) it would actually make the steering less sensitive as speed increased, rather than more.

The reason real road cars do not have quick steering racks is that most people find high speed lane control easier with less sensitive steering, and overly sensitive steering gives an impression of instability to people.

edit: Not aware of those systems Matrixi refers to. Edit again, yes I was - it's the Active Steering thing, I didn't realise it had that feature though.

Quote from adrianstealth :

re. dk2 with motion

I'm using a motion setup with the dk2 with extremely good results
the camera is in a fixed position on a wall

my platform pitches / rear traction loss
+seat motion on top of the platform

home setups such as mine do not make high movements! just enough to feel the onset of gforces + position indication

the dk2 movement actually adds to the motion effect
-chair moving slightly to the left to simulate ones body being moved to the side in relation to the wheel shows in the dk2 & gives good additional visual clues


something like the redbull simulator would obviously need a VR track cam mounted to the actual platform

http://youtu.be/rE-Fge3gN9w

Interesting. I have been on the edge to start investing into a such setup, but one reason why I haven't has been the tracking issue. Which system are you using?

Do you feel that you get less nauseous when using the moving setup? I feel really really bad hit of nausea and head dizziness when stopping the car and in slow speeds generally.

hi

it's a custom setup ( professionally welded & powder coated )
-tried to post image but can't seem to do it on these forums

it gives all the movement I feel is needed (whilst being able to fit it into a house)

I feel no nausea at all , not tested without motion, I'm surprised you get nausea whilst sim racing, make sure your fps is high with no vsync - you need very low latency, turn down graphic details / no AA etc go for fps of above 150fps

Quote from adrianstealth :

hi

it's a custom setup ( professionally welded & powder coated )
-tried to post image but can't seem to do it on these forums

it gives all the movement I feel is needed (whilst being able to fit it into a house)

I feel no nausea at all , not tested without motion, I'm surprised you get nausea whilst sim racing, make sure your fps is high with no vsync - you need very low latency, turn down graphic details / no AA etc go for fps of above 150fps

Would you mind testing without motion one day?

My PC runs LFS very well (4.5ghz i7, gtx770), so that's not the problem. The problem when stopping the car is that I anticipate the deceleration ending and the nudge forwards that it causes. You know when you stop a real car all the passengers nudge forward. And when that doesn't happen my eyes go blurry, head feels dizzy and an instant hit of nausea to my stomach. I have thought that a motion platform with even the slightest amount of front & back tilt could fix the issue and make it comfortable. At high speeds I have no problems what so ever, but the slower I go the more pronounced the effects are and stopping the car is the most extreme case.

What I have read about the issue it's because of the vestibular system and visual system "fusion" conflict. There's plenty of research made for military sims that are available online and the conclusion seems to be just that. An other conclusion was also that the less real experience, the less simulation sickness and vice versa.

yes I'll give it a go without & report back.

I'm surprised your having such a result, I thought as long as everything in sync with your head movement you'll be fine

remember whilst in motion my head is moving around (vs track cam which is fixed on wall) but the visuals are in sync same as if no motion

the only thing I can think of is there any head wobble effect in LFS which should be off? thus the visuals are adding movement not corresponding to your real head

your system should eat LFS , I use no vsync no fps cap

Quote from Alric :

Matrixi, I know real cars have a large degree of turning but if your racing round a track they reduce the faster you go (don't ask me how it works but they do). So at 100mph going round a corner you only need to turn the wheel a small amount, yet at 2mph you need to turn the wheel 2 and a half times to park.

I find that highly unlikely. More likely is that a bend that seems tight to you at 100mph is very much not tight at 10mph, that is all.

I found one of example of BMW having a variable steering ratio, but this is designed to do the opposite of what you say (it reduces steering lock when parking) and has an manual override to fixed ratio for track use.

Quote from adrianstealth :

yes I'll give it a go without & report back.

I'm surprised your having such a result, I thought as long as everything in sync with your head movement you'll be fine

remember whilst in motion my head is moving around (vs track cam which is fixed on wall) but the visuals are in sync same as if no motion

the only thing I can think of is there any head wobble effect in LFS which should be off? thus the visuals are adding movement not corresponding to your real head

your system should eat LFS , I use no vsync no fps cap

Thanks, will be interesting to hear how you feel.

No the problem isn't with head movements, but lack of g-forces. When you try without the moving platform, try slowly stopping the car while looking out of the side window. That is the absolutely worst for me.

Quote from PeterN :

I find that highly unlikely. More likely is that a bend that seems tight to you at 100mph is very much not tight at 10mph, that is all.

I found one of example of BMW having a variable steering ratio, but this is designed to do the opposite of what you say (it reduces steering lock when parking) and has an manual override to fixed ratio for track use.

I didn't explain myself very well, what I was trying to say is if I set my wheel to 1080 degrees in a sim then when I go round a corner which in real life would require 90 degrees of turning I seem to have to put on about 270 degrees of turning in game. I maybe have a setting wrong or something but that's just what I've personally found. It's probably me doing something wrong but I have to use 540 degrees or less to get anything like a realistic sim racing experience.

I spent a couple of hours testing last night with my motion rig and 100% it doesn't pitch from what the in game slope is. (My camera is on a wall facing me) As I go on the banking it stays totally flat and level, only the accelerometer effects make the chair move which is specifically, acceleration, braking, cornering, engine vibration and road texture up & down. So that is why I don't get the issue with the camera thinking I'm pitched higher than I am.

I know my setup isn't correct but it works really well in lfs. It gives very quick movement & works hand in hand with the rift. It has 4 powerful actuators with about 65mm of travel and in the MRT5 round South City really pushes my head to one side which you can see in the rift and gives the impression of some g-force. I've tried with and without motion, without motion I can last fairly long until I feel a bit ill but get ill instantly when coming to a complete stop. (I now close my eyes until I've stopped). With the motion I can last a bit longer still and when coming to a stop I don't get ill so it is doing something to trick my brain or inner ear. The only problem with the motion is it seems to vibrate the rift a little on my nose (even with the strap tight) which gives a jittery image over high frequency bumps if the motion is turned right up. Maybe less weight of the consumer rift will help with this small issue.

I wish I could let everyone have a go, someone with more brains than me would get better results out of my setup I'm sure. I've been racing cars for 7 years now and so far motion + dk2 + lfs is the most realistic racing experience I've ever had. Feel very lucky lfs has supported the rift so well.

Quote from Scawen :

So the Rift camera, regardless of whether it is mounted on the seat on on a nearby tripod, will think you are looking UP 20 degrees. So now, the way LFS + Rift works at the moment, your view will be pitched up 20 degrees relative to the car.

...

Either way, these problems could be solved by the Rift drivers themselves, if indeed they know the pitch and roll of the motion chair...

Yes. As someone who's somewhat involved in this kind of thing, yeah, you need to know, at some level, the position of the motion platform, so you can subtract back. We're looking at a few different ways of doing that; the easy way is for the Rift hardware to use a camera you can mount on whatever hardware (vibration probably isn't a huge issue, since we're not talking about super high frequency stuff getting sent through and it gets damped anyway; further, big transients will result in your head moving around anyway so little problems won't be too noticeable), but that's not the only method the Rift is using - or, for DK1, not the method it's using at all. So the Rift needs to know the position of the platform. Luckily this is fairly easy to accomplish; it's just a matter of picking the best way to do it, which is a problem we're working on - it's more a matter of deciding on the best policy vis how things will work going forward for Rift development, making sure things aren't too kludgy or shutting doors for other features, etc.

But that's definitely happening.

Interestingly, you can get away with quite a bit of motion - at least, non-yaw motion! - even without any compensation; the more you turn it up, you start to feel like you're sitting in a hammock inside the car tub and swinging around inside it, though, which is a decidedly bizarre sensation when coupled with the eerie realism of the Rift and the 'I'm really on a road' physical sensation of the motion system.

My best guess is that the games won't be responsible for any of this stuff; it makes much more sense to handle it as far from there as possible and as close to the hardware as possible. So for your purposes, Scawen, you probably don't have to really worry about it.

Also, Scawen - PM me if you get a chance; I'm working on something you might be interested in but the public forum isn't the best place to discuss it...

Quote from perisoft :

Luckily this is fairly easy to accomplish; it's just a matter of picking the best way to do it, which is a problem we're working on - it's more a matter of deciding on the best policy vis how things will work going forward for Rift development, making sure things aren't too kludgy or shutting doors for other features, etc.

Are you working with the Rift developers on this?

Trouble is, even though LFS is constantly in the top 3 games on their site, they treat me like any other customer, and that's unfortunately not nice - any time I ask them anything, it's the same as talking to a brick wall. I've tried their forum and Twitter, but it's the same. I am totally ignored by them, although LFS is obviously responsible for many Rift sales.

Makes me quite angry really with the amount of work I put into this thing, they can't afford even 1 minute a week to reply to a simple question.

Quote from perisoft :

So for your purposes, Scawen, you probably don't have to really worry about it.

Good, thanks.

I'll keep working on Rift stuff for now because it's good for LFS, even if Oculus people think they are better than everyone else and can ignore developers asking simple questions about their inadequately documented SDK.

Quote from Scawen :

Makes me quite angry really with the amount of work I put into this thing, they can't afford even 1 minute a week to reply to a simple question.

That's pretty naff but somehow not surprising.

I'm borrowing a DK2 again this weekend, looking forward to trying out the updates.