1.1 wheels:

[Levelord]

Just now, NikkyD said:

why don't they just hire someone who has at least some clue about cars, suspension and the like

Because most of them (like us) are rocket nerds. Our first instinct is to put boosters onto everything

 

It's hilarious to watch them figure out wheels though

[NikkyD]

I'm a car guy, for me it's like watching a toddler try to build a tower out of blocks...

[smjjames]

1 minute ago, NikkyD said:

I'm a car guy, for me it's like watching a toddler try to build a tower out of blocks...

A toddler totally could, just trial and error though.

Maybe all of the car guys (or perhaps those with knowledge of how wheels work) could offer SQUAD some help? *shrug*

[NikkyD]

1) stiffen the suspension, when you drive into a turn and the vehicle can dive in on the outside it gets extra acceleration around the anchorpoint of the wheel and you flip

2) include something like a differential logic, the outter wheel has to turn faster than the inner wheel in a turn, this should reduce the massive oversteer i'd say

3) limit the max steering degrees in respect to Vmax of the vehicle but don't stop there as a landing plane can have way more speed but with 1° you'd be much safer

4) the dampeners need to be harder as well, my 2t Rover can almost hit bottom driving around the KSC (with brown baloon wheels) If someone builds a 10t vehicle and uses those wheel... OMG

5) breakaway torque  !!!! A standing plane on gravel on even ground can simply not move by millimeters per second, not with inflated rubber wheels!

just a "on top of my head" list

edit:

What modern cars can and should do (in a game you can do this as well, you have all the "sensor data" https://www.youtube.com/watch?v=gWL-r72tGOE&t=59

 

Edited April 28, 2016 by NikkyD

[Renegrade]

1 hour ago, Alshain said:

My plane is having a tendency to veer off the runway now though.

I noticed that during braking, my test plane tended to veer hard to the left.. oh well, here's to 1.1.2? 

[Alshain]

1 hour ago, Renegrade said:

I noticed that during braking, my test plane tended to veer hard to the left.. oh well, here's to 1.1.2? 

Oh, well I meant while taking off.

[Drahkon]

On 4/25/2016 at 5:53 PM, Geschosskopf said:

Now calm down and let's try to discuss things in proper sentences with punctuation, capital letters, and other such refinements.

Am I the only person bothered by someone going after a guy for punctuation then using more ridiculous and misplaced commas than William Shatner?

 

And yes, the wheels definitely are a bit more unpredictable, but I personally don't mind the changes.  Space planes remain one of the more difficult frontiers in this game.   

Edited April 29, 2016 by Drahkon

[Arsonide]

6 hours ago, Alshain said:

Steering is still enabled at all speeds.

Stress tolerance on the LY-05 remains unchanged.

Stress tolerance wasn't changed, but deflection magnitude was, which means the wheel takes half the stress from deflection (weight and impact). Slip magnitude was not changed, so if you are power sliding on take off, that might account for the issue.

Can you elaborate on steering being enabled at all speeds? Why would steering disable at a particular speed?

[NikkyD]

5 hours ago, Arsonide said:

Can you elaborate on steering being enabled at all speeds? Why would steering disable at a particular speed?

Sorry to tell this to a SQUAD staffer, but did you read the thread or not ?

[eddiew]

11 hours ago, regex said:

...It's like someone is trying to compensate for ridiculous vertical speed on landing, like they're used to slamming planes around all day and can't land softly for some reason...

I thought they were already set to 6m/s crash tolerance, which seemed generous. Dropping a plane straight down from 2 floors up could be fully expected to blow the bloody wheels off... and the doors. And the wings.

Admittedly there are other issues, but vertical impact tolerance is probably not one of them, imho. A longer runway on the other hand would really help with getting a soft landing

[Ted]

12 hours ago, NikkyD said:

why don't they just hire someone who has at least some clue about cars, suspension and the like

 

11 hours ago, NikkyD said:

1) stiffen the suspension, when you drive into a turn and the vehicle can dive in on the outside it gets extra acceleration around the anchorpoint of the wheel and you flip

2) include something like a differential logic, the outter wheel has to turn faster than the inner wheel in a turn, this should reduce the massive oversteer i'd say

3) limit the max steering degrees in respect to Vmax of the vehicle but don't stop there as a landing plane can have way more speed but with 1° you'd be much safer

4) the dampeners need to be harder as well, my 2t Rover can almost hit bottom driving around the KSC (with brown baloon wheels) If someone builds a 10t vehicle and uses those wheel... OMG

5) breakaway torque  !!!! A standing plane on gravel on even ground can simply not move by millimeters per second, not with inflated rubber wheels!

just a "on top of my head" list

edit:

What modern cars can and should do (in a game you can do this as well, you have all the "sensor data" https://www.youtube.com/watch?v=gWL-r72tGOE&t=59

 

 

1 hour ago, NikkyD said:

Sorry to tell this to a SQUAD staffer, but did you read the thread or not ?

 

The issue here is not that we don't have the necessary wheels (or "cars, suspension and the like") knowledge to approximate wheel physics, but that the middleware (both Unity and VPP) are making implementing this in KSP cumbersome and unfortunately challenging to balance currently. We've looked into PID systems for the wheels, and those can work for wheels for sure, but KSP is a game where you can make little to no assumptions about how the player has designed their vehicles - something that car manufacturers can take with certainty and something we'd need for a PID system to be fully capable. Thus, as I'm sure you can see, it's not really car expert that's needed. Going forward, we have solutions that can mitigate this and plans to solve the problems we're seeing, but this takes time and a lot more hard work from a team that has been pushing to the limit on that. I completely understand that it must be frustrating to see this as someone who is passionate about wheels and cars. And I welcome your suggestions, they make a lot of sense if you're building a system from the ground up, but at this point we're unable to implement those in the short-term.
 

Hopefully that has resolved some of your concerns, if you have any further ones, I'm always available via PM (though sometimes can take a few days to reply as things get busy!).

[Blaf]

1 hour ago, Ted said:

 

 

 

The issue here is not that we don't have the necessary wheels (or "cars, suspension and the like") knowledge to approximate wheel physics, but that the middleware (both Unity and VPP) are making implementing this in KSP cumbersome and unfortunately challenging to balance currently. We've looked into PID systems for the wheels, and those can work for wheels for sure, but KSP is a game where you can make little to no assumptions about how the player has designed their vehicles - something that car manufacturers can take with certainty and something we'd need for a PID system to be fully capable. Thus, as I'm sure you can see, it's not really car expert that's needed. Going forward, we have solutions that can mitigate this and plans to solve the problems we're seeing, but this takes time and a lot more hard work from a team that has been pushing to the limit on that. I completely understand that it must be frustrating to see this as someone who is passionate about wheels and cars. And I welcome your suggestions, they make a lot of sense if you're building a system from the ground up, but at this point we're unable to implement those in the short-term.
 

Hopefully that has resolved some of your concerns, if you have any further ones, I'm always available via PM (though sometimes can take a few days to reply as things get busy!).

Would increasing number of VPP simulation substeps help?

I mean, in such case number of substeps could be added as an option to game settings. (if it's possible, ofc)

[Alshain]

8 hours ago, Arsonide said:

Can you elaborate on steering being enabled at all speeds? Why would steering disable at a particular speed?

Because it was in 1.0.5 and it's not really usable beyond a certain point anyway once yaw control takes over.  I tested it, in 1.0.5 steering control disabled somewhere around 30 to 40 m/s, in 1.1 it never is.

Edited April 29, 2016 by Alshain

[TMS]

Is there likely to be another patch that addresses the over sensitive suspension of plane landing gear?

At present, even the gentlest and slowest of touchdowns result in massive over reactions, sending the plane tumbling head over heel.  Comedic, for now; but it's effectively half the game rendered unusable.

[Bedwyr]

I'm cross-referencing this discussion here: 

Most of the issues GoSlash fixed are in corporated in 1.1.1 but my conversation with him should probably contribute to this conversation. 

And also re-linking the pastebin of my rudimentary career mode airplane with '01 and '05 gear: http://pastebin.com/RtGUtmFr

The behavior is the same before and after the patch for me though GoSlash said the file flew better post-patch for him. I'm afraid I still haven't gotten this ship above 40m/s without crashing. If there's a straight design flaw (not a "here's a workaround you're not using" comment please) such as total lift being insufficient or weight and drag forces overwhelming the gear naturally and inducing the oscillation moments, please let me know.

 

[Knaapie]

On 4/28/2016 at 5:23 AM, regex said:

This got me to the Island Airbase:

that made me make my first vtol ^^

 

[NikkyD]

@Ted The issue I, personally, have is:

You need rocket science knowledge to build proper rockets and get to muns and such. Now you know there are different biomes and you want to max science, so a science vehicle is the next logic step. But if i now need some very specific and odd knowledge of how to make something with 4 wheels drive properly, then the game has a problem.

Wheels are smart enough (or the system) to group them when i place them with radial mirror. So you have a somewhat perfectly aligned axle. Most players build known-vehicle styles, that is 4 wheelers. So you can predict that a lot of designs will be car-like and lay a focus there. No one will complain much if their super crazy 18 wheeler has issues driving, but if a regular 4 wheeler can't even drive around the KSC without flipping over or ice skating, that is reason to complain.

[Violent Jeb]

 

5 hours ago, Ted said:

but this takes time and a lot more hard work from a team that has been pushing to the limit on that.

Take your time guys, its a marathon not a sprint.I am sure you will get it sooner or later. :]

 

Anyway, here are my general observations from wheels in 1.1.1

•Wheels seem relatively stable under ~35 m/s (126 km/h).

•Above ~35 m/s threshold the LY05 is very prone to destruction from "overstress", even with minimal planes.

•Distributing weight is key for stability and wheel survival. I managed to get airborne last night on the lvl 1 runway after roughly an hour and I settled upon a 7-wheeled plane I would almost venture to say that the wheel tolerance for weight should be increased a tad. I will put up pics when i'm at my gaming pc but I left the runway on wheels moving 80 m/s.

•Suspension seems to help steering wheels "adjust" to ground imperfections while remaining straight above 35 m/s

•Even with proper weight distribution, crafts are extremely prone to drift and spinning at high speeds.

•At speeds above 25 m/s (90 km/h), the steering is far too touchy. People have mentioned a damping mechanism to solve this. It is worth mentioning that the behavior almost feels like you are driving a car with extremely narrow tires.

In the meantime, I was able to maintain control by disabling steering and then turning only with control surfaces/sas while in "fine tune mode". 

I feel like with a big wide front tire and more weight tolerance I could get into the air on 3 wheels, so, it's getting there.

 

I suppose it is also worth mentioning that travelling at 126 km/h on grass with an open wheeled concept vehicle IRL is very likely to make you spin out and crash. so, if anything we are looking for "less realism" to make wheels fun at reckless, breakneck speeds.

Edited April 29, 2016 by Violent Jeb

[Bedwyr]

Ok, so talking with Alshain, he's got a video I'm linking here from the 1.1.2 thread:

I asked him to take a look at a craft file from the bugtracker linked here:

http://bugs.kerbalspaceprogram.com/issues/9498

 

It *looks* pretty close to the same issue that Alshain and I have had with the '01 gear. I decided to try the airplane and it flies off the runway just as well as you please. Alshain tried raising the gear mount location to get the CG, and therefore the gear moment of inertia, down. He reports back that it stops the oscillation but results in undercontrolling and veering (did I get that correct?).

I don't have time at the moment to tweak my own airplane but will report back when I follow up on it. If anyone else has the gumption to experiment, it would probably be useful to squad as an addition to that bug report or a new report with a link back to 9498.

I'd really like to trace exactly where on a normal 1.5m fuselage the oscillations on '01 gear when turning on a takeoff roll move from damped to undamped to unstable. Although I have to go to work, erm, "race ya!".

Edited April 29, 2016 by Bedwyr

[Alshain]

Yes, that is correct.  By raising the mounting point of the gear to half way up the fuselage my plane no longer oscillated on the suspension.  However the veering issue introduced in 1.1.1 was then apparent.  I couldn't take off because my tail would have hit, but even if it wouldn't have I couldn't get up to speed on the runway.   Good find @Bedwyr

Here is a new video demonstrating this.

 

Edited April 29, 2016 by Alshain

[Bedwyr]

For completeness sake, could you post a side pic of CM, lift, and thrust locations? I'll do the same when I can.

[Alshain]

[Bedwyr]

Cool thanks! I notice your nose gear is reversed. That might be unstable for steering even if the steering direction is properly reversed. Is it possible that could cause the veering?

For comparison's sake, I've got my CL closer to CM than yours and you've got a high-wing config with a level or slightly down angle of incidence. The Mk1 cockpit with two trailing fuel tanks is the same. Different engine. So I'd anticipate that the weights are similar. I'll go try a few things.

[Alshain]

21 minutes ago, Bedwyr said:

Cool thanks! I notice your nose gear is reversed. That might be unstable for steering even if the steering direction is properly reversed. Is it possible that could cause the veering?

For comparison's sake, I've got my CL closer to CM than yours and you've got a high-wing config with a level or slightly down angle of incidence. The Mk1 cockpit with two trailing fuel tanks is the same. Different engine. So I'd anticipate that the weights are similar. I'll go try a few things.

What gear is reversed?

[Bedwyr]

1 minute ago, Alshain said:

What gear is reversed?

Nose looks like. I could be wrong. The wheelpants fairing is supposed to trail behind like so:

 

 

Edited April 29, 2016 by Bedwyr