Distance constraint parameters

[canerozdemir]

Hello there,

So I am making progress with my project by creating rope from the code between two points and I inserted some components with the guide from the official website. My only problem is the distance constraint within the Obi Solver. Rope is not stable enough for me to work on so I need it to be stable enough so it will look like a straight line but with a little bounce and movement when I move the points, that holds the rope together, to another position.

I tried incresing the distance iteration and it worked really well when the iteration is around like 1500-2000 and the relaxation was 1.3. However, this caused a major FPS drop after 10-15 seconds. So, how can I achieve that desired effect within the limitations of the constraints? 

By the way, I am pleased with the capabilities of the Obi Rope and would like to give a thank you for providing such capable asset to work with.

[josemendez]

Hello there,

So I am making progress with my project by creating rope from the code between two points and I inserted some components with the guide from the official website. My only problem is the distance constraint within the Obi Solver. Rope is not stable enough for me to work on so I need it to be stable enough so it will look like a straight line but with a little bounce and movement when I move the points, that holds the rope together, to another position.

I tried incresing the distance iteration and it worked really well when the iteration is around like 1500-2000 and the relaxation was 1.3. However, this caused a major FPS drop after 10-15 seconds. So, how can I achieve that desired effect within the limitations of the constraints? 

By the way, I am pleased with the capabilities of the Obi Rope and would like to give a thank you for providing such capable asset to work with.

Hi there!

Thanks for your kind words

2000 iterations is a lot. Typically, you don't go over 50 iterations even on extreme cases. Furthermore, substeps are more effective than iterations in increasing simulation quality. So you will most certainly benefit from keeping the amount of iterations low (2-3), and increasing the amount of susbsteps (10 should be more than enough for most purposes). For an in-depth explanation of how iterations/substeps work, see:

http://obi.virtualmethodstudio.com/tutor...gence.html

Also keep in mind that rope length (or rather, amount of particles in it) plays an important role on the cost, so by reducing the blueprint resolution and/or its length you will also improve convergence, at the same substep/iteration count.

[canerozdemir]

Hi there!

Thanks for your kind words

2000 iterations is a lot. Typically, you don't go over 50 iterations even on extreme cases. Furthermore, substeps are more effective than iterations in increasing simulation quality. So you will most certainly benefit from keeping the amount of iterations low (2-3), and increasing the amount of susbsteps (10 should be more than enough for most purposes). For an in-depth explanation of how iterations/substeps work, see:

http://obi.virtualmethodstudio.com/tutor...gence.html

Also keep in mind that rope length (or rather, amount of particles in it) plays an important role on the cost, so by reducing the blueprint resolution and/or its length you will also improve convergence, at the same substep/iteration count.

Hmm, alright I will give it a go and comment the result later on the day. Thank you for your support