New Landing Gear issues, is it just me?

[g00bd0g]

There seems to be two new significant bugs affecting me. Buggy LT-1 and LT-2 Landing Legs. i.e. Gear just sink through surface and ships with LT-1 and LT-2 just sort of skitter around on the surface and never "settle down". Is this only affecting a few people? Otherwise I'm sorta shocked no one noticed before this was released. It's kinda a game-breaker when you can't have stable landed craft or bases.

Edit...

Adding a 3rd item. NOTHING will sit still on the surface, even a single cockpit will randomly slide around forever.

Added a bunch of example vids futher down.

 

 

Edited April 24, 2016 by g00bd0g

[Majorjim!]

It is not just you mate. I rarely use the aircraft wheels but I know they suffer the same as legs do and legs are now wheels. 

 I submitted a bug report about the skittering LT-2 landing legs here if you want to add anything to it feel free. I just hope they get fixed soon as I cannot play the way I want when they are this broken..

[Scarecrow]

Was just about to create a thread about this very issue and then saw this one.  Jittery landing legs that cause explosions before you take off are affecting my craft too.  I am also finding that even when the ships don't explode, previous craft that tracked straight and true during take off are now veering all over the place and are lucky if they can get in to the air before crashing.

Very disappointed so far with 1.1 because of these basic problems that really should have been spotted and corrected before release.

Edited April 24, 2016 by Scarecrow

[tallthomas07]

It's very annoying, pretty much puts a halt to anything I can do in ksp. Anyone hear anything about a patch for this or anything?

[Renegrade]

The "jitter" issue seems to relate to the flexibility of the craft.  If I make a sturdy craft, with gears/struts/wheels attached on non-wobbly parts, the issue doesn't crop up.  If the gear is installed on more flexible locations (like on a row of poorly supported modular girder segments for example), it hops and jitters around like mad.  Various wheel right-click settings do not help.  Strutting can make it better (or at least delay the onset and severity of the jitter)... I'd almost guess that the actual wheel physics are happening in the wrong order with craft physics, which could very well create a feedback loop..

I'm also rather alarmed by the sinking-into-the-ground issue as I'm afraid that when it's finally fixed, bases and landed craft will launch themselves into the air once loaded for the first time.

 

[bewing]

You all remember that 1.1 was originally scheduled for January? Do you understand that Squad has been working almost exclusively on landing legs and wheels for the last 3 months? They are a lot better than they were 3 weeks ago during beta. They will be improved again in the hotpatch. But an actual fix to get them finalized is probably targeted now at 1.1.1, 3 months from now or something.

Tweaking the settings on the wheels is usually enough to get you airborne and landed again. Use winglets instead of landing legs for now. It's a time for creative workarounds.

1.1 is a major rewrite of the game. It is not the same game as 1.0.5. Gotta adapt.

[Scarecrow]

Easiest way to adapt is to continue playing 1.0.5 until they sort the problem out.

[Majorjim!]

40 minutes ago, Renegrade said:

The "jitter" issue seems to relate to the flexibility of the craft. 

Nope, not the case at all, DL my test craft in the bug report, it is a single part with legs attached it vibrates across the ground. It is almost certainly connected the suspension model.

Edited April 24, 2016 by Majorjim

[tallthomas07]

Adapting is not an answer.  This is a major flaw that affects just about every part of the game that comes after orbiting.  Any lander based projects are unplayable which makes making any sort of progress impossible.  All hail the mighty squad for making this game and all but this is something we paid for, it should work.

[bewing]

The game works. Just don't use what are known as "landing legs" as landing legs. Use something else. That is adapting.

[N_Molson]

Agreed, you can definitively land anywhere without using landing legs. There is a ton of stock parts that can make makeshift legs. You can even land on your engine(s) if the surface is flat enough, just come down slowly enough. It might look a bit weird, but it doesn't prevents from playing and making progress.

[AbacusWizard]

40 minutes ago, bewing said:

The game works. Just don't use what are known as "landing legs" as landing legs. Use something else. That is adapting.

Then why even have them in the game?

[basic.syntax]

Adapting is what we have, until the next patch day comes around

In v1.0 we had to adapt to "new aero" and "rocket flipping." 

Use more legs, spread farther apart, descend more slowly.

[anarkhon]

3 hours ago, bewing said:

Do you understand that Squad has been working almost exclusively on landing legs and wheels for the last 3 months?

3 months for a landing legs and wheels? i dont buy that , a good modder can do it within 3 days. alone.

Edited April 24, 2016 by anarkhon

[N_Molson]

Quote

3 months for a landing legs and wheels? i dont buy that , a good modder can do it within 3 days. alone.

They completely rewrote the way wheels are simulated and that's not something you do in 3 days, wait for the patch and you'll see that the new system is much much more satisfying (and realistic).

[Majorjim!]

26 minutes ago, basic.syntax said:

 

Use more legs, spread farther apart, descend more slowly.

Another nope.

Even 8 or more legs cannot support anywhere near the weight they could before and break even landing at 1m/s. And on the groud they vibrate causing the craft to move across the surface like a vibrating phone on a table. They are broken, pure and simple. We just have to wait or a fix sadly. 

Edited April 24, 2016 by Majorjim

[DrMarlboro]

36 minutes ago, Majorjim said:

Another nope.

Even 8 or more legs cannot support anywhere near the weight they could before and break even landing at 1m/s. And on the groud they vibrate causing the craft to move across the surface like a vibrating phone on a table. They are broken, pure and simple. We just have to wait or a fix sadly. 

At least from my playing with the LY-01/LY-10 gear, it seems to be a problem with the traction control. If I have traction on, then it will skitter/skate across the ground, increasing in speed if I increase control, and will stop if I turn it off. The problem there is with it off you slide all over the place now. I'm wondering if legs (since they are wheels now) behave in the same way, and have this traction control coded into them. And I don't know a thing about reading code so until someone who can chimes in I'll have to keep guessing. Still no clue why wheels/legs clip through the ground sometimes though. 

Edited April 24, 2016 by DrMarlboro

[Renegrade]

2 hours ago, Majorjim said:

Nope, not the case at all, DL my test craft in the bug report, it is a single part with legs attached it vibrates across the ground. It is almost certainly connected the suspension model.

Even a two part ship has flex.  Parts are never perfectly rigidly attached. Also, the test craft 2 does not jitter in my test sandbox save on either the runway or launchpad unless I force physwarp.  The un-numbered one does explode and sometimes takes the launchpad with it.  My own tests show the more flexible a part is, the more the bounce happens.  Making long support shafts for wheels (or landing struts or landing gear) definitely makes it worse.

One other thing I've noticed though is that sometimes the wheels seem to be reacting immensely to what should be a tiny force, and there might be a fp underflow type issue going on.

NB: At no point am I denying that the problem exists.  Nor am I saying that the current code is acceptable. A positive feedback loop between a craft's suspension and flex shouldn't exist like that in the absence of any significant outside force.

[Majorjim!]

12 minutes ago, Renegrade said:

Even a two part ship has flex.  Parts are never perfectly rigidly attached. Also, the test craft 2 does not jitter in my test sandbox save on either the runway or launchpad unless I force physwarp.  

I have tested it on a fresh install and it indeed does jitter. I think you just can't see it. Look closely at the legs, they are vibrating very fast and over a tiny distance, this causes the craft to move across the surface, it has been seen and confirmed by a few others. Take another look. I will post a video tomorrow so it is clearer. 

[g00bd0g]

Here's a bunch of videos showing silly issues with bone stock KSP (not steam, from kerbal store) and only mechjeb installed to show surface speeds.

Basically, even when sitting on a perfectly flat piece of terrain, launchpad, runway, etc... ships never actually sit still, they are constantly sliding around. Even just a single cockpit sitting on the ground with no SAS still moves around on it's own.

My favorite is the super basic airplane just flopping over and spinning around in circles, with no SAS or anything. Seriously, how did this make it through QA?

I would have participated in beta but I do not own a steam release and I was not allowed to...

 

 

[basic.syntax]

@Majorjim I see the wheel jitter issue in your test craft. It slowly drifts toward the VAB, when the legs are pointing to it. (The rate of drift is 10x less, if the legs point toward the flagpole.)

In a more traditional rocket orientation, with legs around this same cylinder (vertically), stability is very good. 3 legs can support the tank.  "Jitter" shows up under high physics warp, but v1.05 wasn't kind to heavy craft on legs, either - sometimes falling over under increased stress brought by physics warp.

Majorjim's test craft with a center beam seems to suffer from two issues.  In v1.1 the craft seems to spawn farther above the pad than in v1.05, so it hits the pad harder in v1.1 causing pad explosion. 

If I lift a very similar craft in v1.05 a bit above the pad with launch clamps and drop it, the v1.05 pad will explode under the impact. (Whether an I-beam is included or not.) If I do not lift it above the pad with launch clamps, it spawns neatly and then slowly collapses [01, 02, 03]. The joint strength (v1.05) when attaching to the I-beam does not seem as strong as when attaching the leg directly to the tank.  Can't win either way, weaker joint strength with this design in v1.05, or explosions in v1.1.

I think the second issue revolves around angles and maybe legs too close to other legs.  Majorjim's test has opposite pairs of legs with their attachment plates angled far from perpendicular. That should increase torque applied at the attach point. I think legs in this configuration should handle more force, when the attachment plate is more perpendicular to gravity / force. Testing this, I rotated the legs inward one notch, and then two notches. While most of the time all were destroyed on physics load, in some instances some legs survived. Sometimes the tank was gone, and then the leg assembly would jitter and bounce around for a long time.

Wild jitter when only legs were left, got me thinking about legs using the same code base as wheels now, and how wheels are currently disabled ("blocked" message) when they are placed too close to other parts (or if suspension travel brings them too close) to prevent explosions... and Majorjim's setup has landing legs so close their attachment plates overlap. So I offset them, which did not fix legs exploding on physics load / drop onto the pad, but it did stop the leg assembly from wild jittering in cases where the tank exploded.

Then I went back to the simpler configuration of legs attached to sides of tank, and moved them away from each other one notch. Worse jitter. Angle legs inward one notch. Less jitter.  Angle legs inward one more notch, so that leg attachment plates look vertical: no visible jitter.  But moving the legs farther apart wasn't really necessary: reloading the original file, and rotating legs two notches so the attachment plates were essentially vertical, even if overlapping a bit, was all it took to stop the jitter.  Then I took it up ;)  Tank landing... Tank landed.

I have not yet come up with a work around, to allow soft landing for legs attached to I-beam on the center line. Or mirrored I-beams off the centerline.  "No beam" is essentially the work around ;).  Legs on the sides of a horizontal cylinder seem to offer some additional challenges compared to legs around a vertical cylinder (typical rockets.)

I hope we do see some tweaks in a future patch, but I'm learning some configurations to avoid, and ways to work around some of the current problems.

 

[Majorjim!]

1 hour ago, basic.syntax said:

Snip

 

Thanks for testing it mate. Legs in this configuration worked perfectly in 1.1 holding more than the weight of that fuel tank. It seems as if the legs have no suspension and either explode due to max pressure on the legs (No give, suspension) or vibrate very fast, still no give at all. I hope a dev notices my bug report..

[basic.syntax]

1 hour ago, Majorjim said:

Thanks for testing it mate. Legs in this configuration worked perfectly in 1.1 holding more than the weight of that fuel tank...

You wrote 1.1... think you mean 1.05? Which configuration? With what I put together, legs attached to I-beam in v1.05 = tank falls over, joints act like rubber under the strain. (Pictures labeled [01, 02, 03] above)

[Whackjob]

14 hours ago, bewing said:

The game works. Just don't use what are known as "landing legs" as landing legs. Use something else. That is adapting.

You can bear more weight, too.

#EDIT:  For rovering, too.

Edited April 25, 2016 by Whackjob

[tranenturm]

Even in 1.0.5 I used the diagonal wings as early lander legs. Ascent is a challenge, but the stiff wide base makes landings so much easier and crash tollerance is high.