Agreed. Rampa's lock is kind of a 3D version of my loop lock. I'll have to download his sample project to see how he gets the sphere ends to stay rigid. Even my loop loses its shape a bit when Soft Clothed.
Maybe the cube inside the locking makes the sphere ends retain their shape. I'll see when I study the sample project.



"We often compare ourselves to the U.S. and often they come out the best, but they only have the right to bear arms while we have the right to bare breasts"
Bowser and Blue, Busting The Breast

Now I have run into two issues with the next phase: the unwind.

1. The smaller issue is that during the unwind the rope on the winch slips and thus does not unwind as neatly as it winds. It also seem to generate odd jagged breaks. I think, however, if I play with the rope settings I may be able to minimize this problem.
2. I have created a tray under the winch (this tray is shown in white for now but will eventually be invisible). The tray is just made of boxes with Kinematic physics settings (although I also tried Static). The idea is to prevent the unwinding rope from going through the base of the Trebuchet and to gather at the back instead. However, no matter what I have tried the rope seems to go right through the tray. The soft cloth rope works with the winch (i.e. does not go through it) but refuses to work with the tray. The physics settings are Rigid, Kinematic (although I did try Static), Box, Collision Margin 0.8 (although I did try 0.1).


 

"We often compare ourselves to the U.S. and often they come out the best, but they only have the right to bear arms while we have the right to bare breasts"
Bowser and Blue, Busting The Breast

I would delete the tray and insert a physics floor from the create menu, just to double-check myself. Size and place it correctly to stop the rope from falling.

The cubes do lock the rotation of the rope ends in all dimensions. So you can play with different shapes to completely lock one, or all, dimensions. A long capsule would lock only the length, for instance. The rope could still twist around the capsule. The cubes need to be a bit bigger than the spheres, so adjust to taste.

Thanks. That did solve the issue of the collision (not sure what the difference between a physics enabled box and a physics plane is). It is also oddly called an infinite plane but looks finite to me. But I think I need to re-evaluate my approach because even with the collision plane being there, the rope unwinding doesn't look natural.

I was thinking of using your lock rope idea with extra connection points along the way to allow the rope to excess to be guided better...but eventually I would want the guides to disappear so that the rope slack can bend on it own. The issue is I am not sure how to do that. I am thinking that if the guides move a large distance between two consecutive frames, hopefully the physics engine will not try to follow. I'll do some testing and see. Thanks for the sample project.



"We often compare ourselves to the U.S. and often they come out the best, but they only have the right to bear arms while we have the right to bare breasts"
Bowser and Blue, Busting The Breast

Interesting...Rampa's cable does not have ballooned ends like I expected. It seems that the introduction of the driving boxes is what inflates the tube. Nice and elegant. Nice work.

I just did some tests and if the position of the guiding boxes is changed dramatically over two consecutive frames then the guide disengages. That makes it even better because it means you can guide part of the way and then disengage the guide to allow the rest of the motion to be by pure physics.

"We often compare ourselves to the U.S. and often they come out the best, but they only have the right to bear arms while we have the right to bare breasts"
Bowser and Blue, Busting The Breast

Found a solution to the unwinding problem which actually solved both problems. The two issues where that the rope was not unwinding in the direction that I wanted and it was slipping on the axle when unwinding. It turns out all I needed was a physics object that would push the slack rope in the direction that I wanted. Not only did that make the rope unwind towards the position that I wanted but also the pushing object kept the rope tight minimizing the slipping effect.

"We often compare ourselves to the U.S. and often they come out the best, but they only have the right to bear arms while we have the right to bare breasts"
Bowser and Blue, Busting The Breast

Rampa/4u2ges are you selling your lock based tube/rope/cable in your stores? A long time ago I was working on a flexible cable solution that would allow attaching a cable to two or more objects. At the time I constructed it using a bunch of small segments which were joined together by constraints. The segments where small enough so that they would not crash into each other but then they each had a larger non-physics shell which gave the illusion of a continuous tube. The solution worked but was very finicky. If the rope was stretched too long or moved too quickly the rope would break apart. However, on the off chance that someone wanted it, I put it into my Marketplace store. I didn't implement it as any actual prop, just a flexible cable. To my surprise it is one of the most purchased item in my store.

The one advantage of the cable that I originally created was that it could be attached to any number of objects anywhere along the length of the tube/rope. I think Rampa's locking mechanism is ingenious and so simple but it has the disadvantage that the guide (cube) causes the tube/rope to inflate. This is not a problem at the ends because opacity can be used to hide that but it becomes a problem if you want to place a guide (cube) somewhere along the way. For such cases I see a couple suggestions:

1. The guide along the way could be just a physics object which is not actually connected to the tube/rope but just pushes the tube/rope as desired. This is the technique that I used to address my trebuchet issue.
2. The tube/rope could be constructed with bubbles, along the length, attached to the length of the tube/rope. The guide (cubes) could then be placed in those bubbles instead of inside the main length of the rope so that the main length of the tube/rope is not inflated. The UV for the bubbles could be isolated so as to easily allow them to be hidden with opacity.

    
As I indicated in my previous post, a guide can be disconnected in real time by moving the guide a large distance between two consecutive frames. This would mean that any number of guides could be used to setup an initial shape but then the guides could be disconnected to allow the shape to flow naturally from that point.

"We often compare ourselves to the U.S. and often they come out the best, but they only have the right to bear arms while we have the right to bare breasts"
Bowser and Blue, Busting The Breast

I had pretty much forgotten I had a store, from ages ago. I have no plans to put it in my store. You can take the idea and run with it!

@Lord Ashes
Quite frankly the idea of creating pockets might not be such a good for certain scenarios. There are many cases where the cable would need to be laid on the rigid ground.
In this case pockets might be counter-productive (even if not used) further destabilizing the cable/rope while coming in contact with the floor. Guides should be user-controlled IMO.
A simple capsule inside the rope with the diameter equals to the diameter of the cable (maybe a slightly larger), would not inflate the cable and at the same time could be an excellent helper guide for precision.

Here's something I am working with and I use the capsule to keep the near end high and parallel to the floor as I needed (this is simulated already).
It does hold well (for what I doing at least) with fairly low Drag number.


Without capsule guide



With capsule guide




The goal was to secure the cable inside the socket centered and prevent movements.
There should also be external guides to keep the cable from moving in certain directions.
So all this could be achieved on a user side with proper instructions.