Inertial dampeners cant cope with multiple grids/subgrids.

Isaac Fry shared this bug 5 years ago
Solved

Any craft with the capacity to keep itself in place has the following issue:

Attaching a subgrid via rotor or piston, or any grid via landing gear or connector will cause the main craft to slowly sink or tilt until it falls from the sky.

I have linked a blueprint that shows this with a rotor at the bottom of the post. If you sit the camera very close to the arm on the rotor you can see it slowly dropping. Eventually the craft will tilt too much and fall. You can replace the rotor with a piston with a few blocks on or a landing gear with blocks on. The problem is the same. Attaching a landing gear to one side of the craft can sometimes cause it to even tilt in the wrong direction.

I have also observed the craft continue to tilt slowly even after deleting the rotor or piston. Which may be another issue entirely.


I used a lot of words but essentially: Any craft with rotors or pistons or docking craft cannot remain in a gravity field without eventually crashing.


https://steamcommunity.com/sharedfiles/filedetails/?id=1474941994

Replies (46)

photo
5

I would like to clarify that despite a note in the patch notes saying this was fixed, it is not as of 1.187.2

I can see no difference.

photo
3

I have the same. If I connect, clamp, or otherwise dock a ship into my landing bay, the entire ship grid will slowly list in that direction. I've also noticed that my ship grid, for some reason, keeps moving even when it should be stationary. I often get small grid tremors that end up randomly destroying the landing gear of smaller clamped ships.

photo
2

Small Hydrogen ship with enough thrust and full tanks - roughly 26t weight.

I attached two solar panels on rotors (thus about 1t more with 2x rotor head, few armor blocks and 2x solar panel)..


Ship with enabled inertial dampeners can't keep altitude by itself despite having full hydrogen tanks and 6 small hydrogen thrusters in right direction (that should be enough for almost 50t).

photo
2

It would be nice if they fixed this, Perhaps via adding a system that requires you to use your noggin to build ships, You have a thruster on the right of a mass centerpoint, Your ship will veer to the left, and vice versa, up down, left right, back forth.

That would add some much needed depth to this game, Note: rather then needing people to place thrusters perfectly, it would simply throttle down a thuster that would otherwise cause the craft to veer, Meaning, Newbies could still build ships, and they would work (as long as they have at least two thrusters on all sides, and or perfectly balanced their ship) But, It would be more efficient for older players to build ships with mass and center points in mind.


As i said, This is an opportunity to add more depth to the game.

photo
1

Hello,

I wasn't able to reproduce it with attached blueprint. Could you provide video and confirm it still happens?

Thank you.

Kind RegardsKeen Software House: QA Departmen

photo
1

Getting video could be hard. I just checked again and its still happening.

Are you sure you got close enough to see it moving? It's very slow. Fly as close as you can, hold the camera still and you can see it slowly move down the screen. Add more heavy blocks to the end of the arm and it will speed up.

If you still can't see it I could try and get video.

photo
1

I checked it again and I can see small movement, so we created a ticket for it and some programmer will have to look at it. From what I heard it seems to be a problem with Havoc constrains and the fix probably won't be as easy as it seems it could be.

We will inform you here once we have more information.

photo
1

Thanks! That's all I could reasonably hope for.

photo
1

I also noticed that the rotation movement can be stopped by overriding gyroscope. Just changing "override gyroscope" checkbox effectively stopped the grid provided above. I know it is not definitive solution, but it could help with parking of this kind of vehicles in gravity until it will be solved.

photo
2

Still an issue in 1.188.1


Mass attached to a rotor/adv rotor is not compensated for and ship in atmosphere cannot maintain hover on any axis. :[

photo
2

Still an issue in 1.189.041

see linked workshop blueprint. with the 'guided bombs' hanging on rotors under the wings the damn thing cannot stay aloft. As soon as the bombs are released (detach rotor command set up in upper cockpit) it can hover just fine.


https://steamcommunity.com/sharedfiles/filedetails/?id=1679887251

photo
2

I have the same issue with a ship that has another ship connected via a connector

photo
1

Still an issue. I worked around it using the Gravity Aligner script, but it's very annoying.

photo
1

Same issue here :(

photo
1

This makes a Tug completely impossible since the rate of falling is based on the ratio of mass between the two grids.

photo
2

Why is this still an issue almost a year after being posted? Extremely aggravating.

photo
1

Have this bug as well. I added some engines to the second grid past the rotor and that helped some but not 100% fixed.

photo
1

I've noticed this happening whenever I add or remove a block from a grid. I built a miner awhile back and it hovered just fine but when I ground a block down and it disappeared, the back would tilt a bit, add a block and the same thing happens. And eventually the ship tilted too much and crashed.

photo
4

17 months, this bug is still occuring when using landing gears / connectors to lift and move any grid/item.

photo
4

I think this bug has been in the game since day 1 :P


It is caused by the dampeners only using the ship's mass for how much thrust to apply. So all the extra weight on a subgrid is not added into the math. I believe i have seen dampener scrips, those may work

photo
4

Yep. It seems like it would be a few lines of addition to the mass the inertial dampener uses in gravity, it's been a bug since the beginning, and it still needs to be fixed.

photo
1

Could be, we don't know what the code looks like. In the most simplistic way, yes, it would only be a simple vars that need to have their names changed..


The reserve issue of this is found when connected to a base. You see the ship's weight as the FULL base, and not just the ship. Same type of issue when it is doing the math.

photo
3

It's such a fantastic game but some of these bugs, specially regarding inertia and camera zoom are almost unbearable. I feel so spoiled from Factorio dev team and really hopes Space Engineers got a similar treatment.. Is there any bugfixing going on nowadays except from critical bugs? Some issues have been listed for many months...

photo
photo
2

Same issue here.

First I thought it was because of the mods, but without the mods the game is a bit "empty".

At a certain point I managed to get it hovering again, but it doesn't work anymore and I can't figure out the combo I did to get it working. I have a mining ship with 2 rotors with a drill on them. This was to mine down then flip them 90° and mine horizontal... For now I think I'll shelf it

photo
4

Any updates or plans to fix this? This issue completely breaks many designs.

photo
1

I am having this problem also! My two small ships docked by landing gears in by large ship that hovers in the atmosphere slowly sink to the ground because the Inertial Dampeners don't take into account small ships that are attached via landing gears.

I have found a workaround though is to turn off all power sources in the small ships after you land in your big ship however that solution is far from ideal and this bug still needs to be fixed. Please fix this Keen :)

photo
3

Is this ever going to be fixed?

photo
5

Yeah, in 2021 with the next DLC for 5 dollars they might fix this, but only for people who own the incoming DLC

photo
photo
2

A grid has a center of mass and knows how much thrust is has in each of the 6 directions. Keeping afloat is then a matter of activating thrusters on up to 3 of the 6 sides by different amounts depending on orientation of the ship. Simple.

Now when you have connected grids these conditions break:


  • Connections are always ever so slightly wobbly - not rigid. They are two separate "particles" now. If you turn down the nose of your ship and you have a sub-grid attached to the back of it, that sub-grid may be moving up at 100 m/s. As that is the speed limit, the engine will slow down the sub-grid and your main ship is going to sink.
  • There is currently "Clang" introduced that exerts strong rotational phantom forces on grids when connections are under tension. (The many videos of grids suddenly spinning out of control and exploding.) Ships may roll due to Clang or this bug, so both need to be fixed for this one to be considered solved.
  • The connected grid may or may not have active thrusters. If it does, they can point in other directions than the main grid's thrusters and also move around. E.g. thrusters on a VTOL ship. The thruster code is not prepared to combine multiple grids with various thrust directions not aligned to the main grid's 6 axis. A more complex algorithm needs to be developed.

Maybe if landing gears created a rigid connection akin to a merge block with a common center of mass, no physics interactions and only one grid having thrust I could see a fix for tug ships and motherships on the horizon.

So stop complaining and start becoming pessimistic! :D

photo
2

As of right now i am still having an issue with this

The circumstances in which i was experiencing it:

I have a small grid ship

I attach a large grid module by connector and then with landing gear

bug continues with only the connector locked or with only the landing gear locked

photo
7

22 Months! Come on! This is annoying and prohibits making a lot of construction ideas. Do something KEEN.

photo
1

I have the same bug with a combo of: station>advanced rotor>small rotor head>thrusters


It works but stops working for reasons that I couldn't identify, i think it is when i add blocks or when i remove/add thrusters

photo
1

I think it's clang, but maybe with the new collision shapes update it is solved?

I haven't tried since then.

photo
1

i was about to mention that Whiplash141's excellent subgrid thruster script would be an interim solution (just put thrusters on the subgrid and use the script and dampeners should work) - but it does look like that the recent update may have fixed this issue

photo
1

UPDATE: I was wrong! Just hadn't tested with a big enough ship. (lost my old blueprint where this was happening)

So yes - big ship, hovering nicely, add some pistons and it still starts to sink as before.

However - Whip's script will fix it - but obviously it requires one or more thrusters on one or more subgrids (but since the dampener drift is usually small will likely not need much additional thrust)

photo
photo
2

This is still not fixed as of update 1.196.

photo
2

Same.

Still happening.

This happens even on small ships. I have a little Grinder/Welder ship with rotating arms to swap tools. And even though the majority of its thrusters are on its arms, it still "slides". bigger ships with less thrust have now problem.

photo
3

Still an issue. Wouldn't simplest solution be to add mass of any object attached by landing gear to overall mass of ship? Or at least a good temporary workaround perhaps.

Using merge blocks works for certain aspects (large block to large block ships) but doesn't help small blocks to large block or piston / rotor sub-grid to main grid.

photo
1

Yes, this is annoying. Cannot create any in atmosphere ship with VTOL thrusters - when is this going to be fixed?

photo
2

for the moment i suggest using this script

https://steamcommunity.com/sharedfiles/filedetails/?id=757123653

it works

photo
photo
2

I've just encountered this problem in 1.196.019 and been exploring different solutions with scripts and such, I have enough understanding now I think and I have a suggestion for keen if they see this, basically the inertial dampening calculation of the main grid cannot take the weight (or mass) of sub grids into account because they are not inherently linked variable-wise, this means in natural gravity fields or artificial ones, the connected sub grid will always sit on top of the main grid without the inertial dampeners of the main grid taking the extra mass into the calculation and supporting the weight regardless of the amount of thrusters attached to the main grid which adjust to match that already calculated ratio for just the main grid, for gameplay reasons it would make sense that non attached grids will weigh another down, I would suggest that the keen programmers simply implement a small script that registers a main grid and a sub grid and simply adds the additional weight into the final dampening calculation, this means in gameplay for instance if you have a large grid carrier and a small grid fighter, in order for the carrier to hold the fighter and allow this type of script to work, it must be connected to the carrier through a connector, but cannot be simply landed in a bay with landing gear, there is that landing gear limitation, but I think its a viable solution to an otherwise annoying bug, I'm not quite apt enough at writing code to do it myself but I think one might be able to do this with an artificial inertial dampening script with that simple added sub grid weight to the calculation and use that instead of the in game inertial dampeners, and I know that Whip's sub grid manager does a similar thing and I've been experimenting with it but instead of allowing the main grid to support the full weight, instead its more of a work around using the onboard propulsion of the sub grid to support its own weight, but this can create a lot of bugs an tends to be finnicky with thruster controls and whatnot, but hopefully this helps keen get an idea for a simple patch fix, if not it would be cool if someone figured out how to code the calculation into a script, if it still isn't fixed and I get good in the future and figure it out myself ill put something in the workshop.

photo
1

Thanks for investigating.

photo
photo
1

Hello, Engineers!

Thanks a lot for sharing with us so much information and keeping this thread going. First of all, I need to say sorry that it took so long before we were able to get to it. On the other hand, this issue was successfully reproduced from another submitted thread. See here:

https://support.keenswh.com/spaceengineers/pc/topic/grids-with-attached-subgrids-sinking-while-in-natural-gravity

I´m pretty confident it is about the same issue.

As I already reproduced the issue from the other thread, I need to close this one as duplicate, as there is nothing more we can do about it. It was already put into our internal system.

You can check for status and updates in thread shared above.

Hope you will understand.

Please, if you come across anything else, don´t hesitate to write to us again.

Kind Regards

Keen Software House: QA Department

photo
3

Hello, Engineers!

I´m happy to inform you that this issue will be fixed in upcoming game update v200.

Kind Regards

Keen Software House: QA Department

photo
2

Is this going to be fixed on something v2XX? or in v200?

Cause i am still the issue on every grid with subgrids.

photo
2

Interesting, i just experienced exactly this bug with v1.202 ...

Am i doing something wrong or is the problem still present in this version?

photo
1

No, John, the problem is still present in the current version.

I made a new thread. They will most likely file that as a duplicate report.

This thread is filed as 'fixed', which it isn't, and hasn't been for 4 years, so to hell with that corporate simile.

QA Department should demand a video of the fixed behavior, not take their third word for it.

photo
photo
3

I am still experiencing it - my welder ship with welders/grinders on advanced rotors does not stay in the air hovering even so there is enough thrust and simply sinks in the same way as reported years ago.

photo
2

Problem still exists with the current version despite the multiple responses from Keen saying it is/will be fixed. Got a rotor/hinge construction ship that sinks altitude regardless of inertial tensor and rotor/hinge locks.

photo
4

This is still a problem. Literally every grid with a subgrid or attached grid tilts and sinks in gravity. Not sure why it says it's fixed.

photo
3

THIS IS STILL A PROBLEM. Unmark it as solved, this is very much STILL, years later, not fixed.

photo
1

FIX THIS DAMN THING IT'S CRIPPLING FOR ANY INTERESTING SHIP DESIGN!!!!!!!

photo
2

Confirmed, still an issue.

photo
1

This might be related to an issue with AI task block I experience and described here https://support.keenswh.com/spaceengineers/pc/topic/41608-ai-recorder-ai-move-not-able-to-find-first-waypoint-when-rotor-attached

In short: Sub grids break the way point AI.

photo
2

Unbelievable. i just made an Account to confirm that this bug is still present... and this Thread is 5 Years old. WTH?

photo
2

Not big soup rice.

photo
1

Confirmed still an issue on dedicated servers. I don't think I've ever seen devs ignore an issue quite like this. It'd be nice if they tested fixes before deploying them to production and claiming they work.

photo
Leave a Comment
 
Attach a file