Posts: 15
Threads: 2
Joined: Nov 2018
Reputation:
0
16-11-2018, 02:16 PM
(This post was last modified: 16-11-2018, 02:38 PM by imtehQ.)
(16-11-2018, 02:02 PM)josemendez Wrote: Did a quick test with a 1 meter rope, 12 particles (2000 kg each), and several rigid bodies, ranging in weight from 6000 kg to 80000 kg. All tests were perfectly stable (if ever so springy), so clearly something else is wrong with your setup. Can you describe it in more detail?
Here's a video of my tests:
I will do some tests outside the current scene and i will get back to you on this, it might be indeed something with the setup.
Yea ... i think that there is something not right with the setup or someone from the team change the code somewhere.
With a default rope solver and particles at 2000 kg, and the block at 60,000 there will be some spring like behavior happening.
https://i.imgur.com/QWTDXZl.gifv
Posts: 6,348
Threads: 24
Joined: Jun 2017
Reputation:
400
Obi Owner:
16-11-2018, 02:49 PM
(This post was last modified: 16-11-2018, 02:50 PM by josemendez.)
(16-11-2018, 02:16 PM)imtehQ Wrote: I will do some tests outside the current scene and i will get back to you on this, it might be indeed something with the setup.
Yea ... i think that there is something not right with the setup or someone from the team change the code somewhere.
With a default rope solver and particles at 2000 kg, and the block at 60,000 there will be some spring like behavior happening.
https://i.imgur.com/QWTDXZl.gifv
Crank up the amount of distance constraint iterations to around 20. That will reduce the springiness to an acceptable level. Keep in mind that the more particles there are in the rope, the higher you will need to crank the amount of iterations. It is often much better to just reduce the timestep a bit.
Posts: 15
Threads: 2
Joined: Nov 2018
Reputation:
0
(16-11-2018, 02:49 PM)josemendez Wrote: Crank up the amount of distance constraint iterations to around 20. That will reduce the springiness to an acceptable level. Keep in mind that the more particles there are in the rope, the higher you will need to crank the amount of iterations. It is often much better to just reduce the timestep a bit.
So i reinported the asset to make sure everything was as default and increesed the distance constraint iterations to 20, yet no change.
Even put it to 50000 to only see a fps drop (to be expected)
Yet no change is how it behaved
https://i.imgur.com/PnEPz9p.gifv
Posts: 6,348
Threads: 24
Joined: Jun 2017
Reputation:
400
Obi Owner:
16-11-2018, 04:46 PM
(This post was last modified: 16-11-2018, 04:48 PM by josemendez.)
(16-11-2018, 04:41 PM)imtehQ Wrote: So i reinported the asset to make sure everything was as default and increesed the distance constraint iterations to 20, yet no change.
Even put it to 50000 to only see a fps drop (to be expected)
Yet no change is how it behaved
https://i.imgur.com/PnEPz9p.gifv
What's the mass of the particles? With 50000 iterations the rope should be as rigid as a steel rod...
Are you using pin constraints to attach the body to the rope?
Posts: 15
Threads: 2
Joined: Nov 2018
Reputation:
0
16-11-2018, 04:50 PM
(This post was last modified: 16-11-2018, 04:51 PM by imtehQ.)
(16-11-2018, 04:46 PM)josemendez Wrote: What's the mass of the particles? With 50000 iterations the rope should be as rigid as a steel rod...
Are you using pin constraints to attach the body to the rope?
Here is me going the setup on the new rope, maybe im just doing it wrong:
https://i.imgur.com/ssSEIqE.gifv
Posts: 6,348
Threads: 24
Joined: Jun 2017
Reputation:
400
Obi Owner:
16-11-2018, 04:54 PM
(This post was last modified: 16-11-2018, 04:56 PM by josemendez.)
(16-11-2018, 04:50 PM)imtehQ Wrote: Here is me going the setup on the new rope, maybe im just doing it wrong:
https://i.imgur.com/ssSEIqE.gifv
Ok...I don't see the part where you set the amount of iterations from the default (3) to 20...
Also, you're making the rope several times heavier that the load (2000 vs 200 kg).
Posts: 15
Threads: 2
Joined: Nov 2018
Reputation:
0
(16-11-2018, 04:54 PM)josemendez Wrote: Ok...I don't see the part where you set the amount of iterations from the default (3) to 20...
Also, you're making the rope several times heavier that the load (2000 vs 200 kg).
ah forgot that part,
https://i.imgur.com/6nbIprj.gifv
Posts: 6,348
Threads: 24
Joined: Jun 2017
Reputation:
400
Obi Owner:
16-11-2018, 05:09 PM
(This post was last modified: 16-11-2018, 05:13 PM by josemendez.)
(16-11-2018, 04:57 PM)imtehQ Wrote: ah forgot that part,
https://i.imgur.com/6nbIprj.gifv
My result is vastly different from yours.
- What is the amount of particles in the rope? mine is 11.
- Mass of the particles? mine is 2000.
- Mass of the load? mine is 60000.
- Amount of distance iterations? mine is 20 (sequential mode, parallel mode converges much more slowly).
- Your timestep? mine is 0.02
- Also, what Unity version are you using?
Posts: 15
Threads: 2
Joined: Nov 2018
Reputation:
0
(16-11-2018, 05:09 PM)josemendez Wrote: My result is vastly different from yours.
- What is the amount of particles in the rope? mine is 11.
- Mass of the particles? mine is 2000.
- Mass of the load? mine is 60000.
- Amount of distance iterations? mine is 20 (sequential mode, parallel mode converges much more slowly).
- Your timestep? mine is 0.02
- Also, what Unity version are you using?
- What is the amount of particles in the rope? 21
- Mass of the particles? 2000
- Mass of the load? 60000
- Amount of distance iterations? 100 sequential mode
- Your timestep? mine is 0.01
unity 2018.2.7f1
Posts: 6,348
Threads: 24
Joined: Jun 2017
Reputation:
400
Obi Owner:
16-11-2018, 05:28 PM
(This post was last modified: 16-11-2018, 05:28 PM by josemendez.)
(16-11-2018, 05:19 PM)imtehQ Wrote: - What is the amount of particles in the rope? 21
- Mass of the particles? 2000
- Mass of the load? 60000
- Amount of distance iterations? 100 sequential mode
- Your timestep? mine is 0.01
unity 2018.2.7f1
I'm using the same Unity version, but still my result is completely different from yours. Even using just 20 iterations and 0.02 as timestep, the rope is much, much stiffer than yours.
Can you send a sample scene? support(at)virtualmethodstudio.com
|
