Bumpy runway in Cape Canaveral

[Heady978]

As the topic says, the runway of standard KSC has several bumps or gaps so it is difficult to start or land. Got this problem on 3 different installations even when using vanilla Kerbal Space Program and just the default installation of Real Solar System in 1.2.2. I also found several posts with this problem - only a workaround is available: install a KSC mod an use any other runway.

[TwistedMinds2]

I uploaded some screenshots [here], if it's useful.
The gaps are always at the same distance, with the same 'shape' (first is a V, all others are straight).
The height is different on each reload / launch, and can sometimes reach half a kerbal in height.

Seems to be the same in stock, but greatly exaggerated in RSS, to the point of (sometimes) forcing half a dozen reload before getting a "usable" runway.

[PhineasFreak]

@Heady978 and @TwistedMinds2 Sorry to disappoint you lads but this is not going to be fixed for KSP 1.2.2. This bug is part of the KSP KSC runway model(s), inaccessible to anyone but Squad. It is worse under RSS because all point to point distances are multiplied by 10 (i.e. a gap 10 cm in length will be 1 m under RSS).

KSP 1.3.0 is supposed to include a fix for that (there is a devblog by @JPLRepo in the KSP forums that describes the steps that were required in order to fix these gaps) so be patient.

[TwistedMinds2]

Yeah, that's what I thought. Thanks for the explanation though :) Let's hope 1.3 truly fix the issue.
As always, the grass is straighter on the other side.

[JPLRepo]

Well definitely something I have already fixed in 1.2.9 for 1.3 so after that you should be fine.

[PhineasFreak]

Some of the bumps are still substantial sometimes (depends again on how lucky you are when the world loads but that is generally valid for RSS) but the situation has improved a lot (tested with a jet rover, final speed was ~200 m/s before it was destroyed by general instability due to the construction - not due to any bumps).

TL;DR works much better now. Thank you @JPLRepo for taking your time to fix the runway!

[JPLRepo]

That is great to hear. Looking forward to getting my RSS-RO install going under 1.3.

[laukejas]

@JPLRepo I am using KSP 1.3.1 with RSS-RO, and the runway is still bumpy as hell. Anything above 15 m/s crashes the plane. Is there any fix?

[raidernick]

use the kerbal constructs shuttle runway, the stock one will likely never be properly fixed as it's segmented.

[laukejas]

Okay, but what about landing on terrain for contracts and so on? They are still segmented...

[raidernick]

that's a stock ksp thing though, the segments in stock are 10x smaller so they aren't a problem, it has to do with the fact that rss scales the planet. I don't think it's fixable, at least not on our end.

[pack7df]

I would like to try some solution about runway bumping. But i think it is too straigforward. I think somebody would think about it; and it doesn't work. So i have to ask before put hand on it.

Raidernick said that the problem is because the scaling of the world and little diferences on stock ksp. What about if ithe KSC is moved (through KSC switch .cfg) to an area where for sure is plane?

Could this fix the problem?. Right know i'm parsing the whole terrain near the kcs and writing a heigh map in order to find a very plane area.

I will write any results about this try.

[rsparkyc]

I don’t think that will work, as I’m pretty sure ksp replaces the terrain around it. Try placing KSC in a bumpy area and see if it’s different. I think we have runway issues at all launch sites, correct?

On Wed, Dec 26, 2018 at 20:27 pack7df ***@***.***> wrote: I would like to try some solution about runway bumping. But i think it is too straigforward. I think somebody would think about it; and it doesn't work. So i have to ask before put hand on it. Raidernick said that the problem is because the scaling of the world and little diferences on stock ksp. What about if ithe KSC is moved (through KSC switch .cfg) to an area where for sure is plane? Could this fix the problem?. Right know i'm parsing the whole terrain near the kcs and writing a heigh map in order to find a very plane area. I will write any results about this try. — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#106 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AB8pssRiFUjcbX8GXMqdYf_ND8Qrfkiyks5u9CHtgaJpZM4Mj90Y> .

-- -Ryan

[pack7df]

Well, there are some places that are less bumby than others; even in long times statiscally speaking. I remember now (as you said KSC changes the terrain) that there is a place, almost inside hole in a mountain (a place that KSCSwitcher defines), near to NY. i'm using memory now. Wich is very stable related with bumpers, just it's to was to close to mountains and is hard to take off and land. I even remember that the Tacking Station was almost under the terrain visually. I will make some test in 2 diferentes positions. and let me see the diferences in mappings. Let me activate the plugins.

[pack7df]

Hi gain, I had made some research about runway bumping. I had realized the ship facing vectors at the start of the runway and at the end of the runway are not the same; expressed in earth rotating reference frame. In fact the angle diferece is the same as positions from the center of the earth.

It means (i guess) that the runway is trying to "touch" the earth curvature all the way but using planes. So, two planes have diferents slopes and the unions could be the bump; besides the scale done by RSS. I hade made some math. It gave me the diference of 1/4 m (for a very small angle inclination as the test i have done), a runway of 2500m and 2 planes. Is correct my guess? could it be the reason of this problem?

I had read also on https://github.com/NathanKell/RealSolarSystem/wiki/PQSCity-and-PQSMod_MapDecalTangent.
Would you please explain better what ReorientInitialUp is?

Thanks in advance.

[pap1723]

There is obviously very little we can do about this. The report is that it is better in 1.6.1, but it is a stock model issue, so it cannot really be addressed from our end.