Fly knot?

[pduggan]

Thanks, Karim, but you gave me the key!

Thought I'd post my method here, but the .xar file's far too big and I'll have to settle for a GIF version (so a ridiculous 400K+ for a simple knot must be what you get for converting close-spaced brushes to shapes!)...

Shouldn't have needed the top layer, but had already converted the middle layer to shapes before spotting that last bit!

Making the knot was tedious rather than difficult, with finding the best split points for the original line crucial in keeping the rotations of the 'rope' brush aligned. But the simplified version posted first took minutes rather than hours because its brush stroke could be split, joined and overlapped at will.

Although the joins look passable at 100% (same size as last night's posting), they're not quite so neat at the 500-1000% I had to zoom in to for some subtracting and slicing. And I can't help feeling that it's a hell of a lot of effort for the end result, so I'm still dreaming of being able to draw a single line that can be directed to pass over or under itself as necessary!

Peter</p>

Peat Stack or Pete's Tack?</p>

[This message was edited by Peter Duggan on March 14, 2001 at 11:33 AM.]

[Egg Bramhill]

I did this one using Ivans technique of slicing a bitmap etc

[gwpriester]

Looks like a Figure 6 knot Rric [img]/infopop/emoticons/icon_rolleyes.gif[/img]

Gary

Gary Priester

Moderator Person

Be It Every So Humble...

[pduggan]

Well, I managed this one with just a brush, one foreground layer and no converting to anything! Nearly repeated the feat with my figure-of-eight on the bight, but landed up with a single unwanted join on top...

Peter</p>

Peat Stack or Pete's Tack?</p>

[This message was edited by Peter Duggan on March 15, 2001 at 05:36 PM.]

[pduggan]

And here's the .xar file in case anyone wants to disect it.



Peter</p>



Peat Stack or Pete's Tack?</p>

[Egg Bramhill]

Peter

It's interesting but having downloaded your excellentsheet bend I noticed something strange regarding the "rotation" feature of the brush tool. It doesn't appear to rotate around the centre.

I cant really explain what I'm trying to say but if you look at my attachment and your image you'll notice that the strands only render on one edge of the rope instead of both.

In other words the brush doesn't follow a sine wave pattern as you'd expect (bottom example) but the top type wave (whatever thats called) ??

Egg

[pduggan]

Thanks Egg:

I've been aware of this too (how could I not be with a pile of real rope sitting on my study floor?), but I've been through every tool in the 'edit brush' dialogue box and haven't come up with a way round it. And it looks fine in the attached screenshot (thanks, Klaus!) - replicating this in Xara at 500% shows the alignment of the circles in my original brush to be less than perfect but the rotation to be bang on, which doesn't explain what you're talking about!

(So can anyone explain this effect?) [img]/infopop/emoticons/icon_confused.gif[/img]

Another thing that's been bothering me is the lay within the strands, which should twist with the strand like the strand twists with the rope, but I think you'd need some kind of amazing stacking brush facility for that.

But, hey, it all started as a piece of fun - it's only now that we're starting to get frustrated at our inability to make it look like real rope!



Peter</p>



Peat Stack or Pete's Tack?</p>

[Karim]

I've got a feeling that you're looking at the same problem we had trying to get ropes to overlap. If you look at Egg's blow-ups, you realise that the top of the brushed rope looks correct because strands to the right are on top of those to the left. Look at the bottom of the brushed rope, and strands to the right are still on top of those to the left (because that's what the brush tool does), but in real rope they should be *below* strands to the left, with the topmost strands of course being in the middle. Have I lost you yet?? [img]/infopop/emoticons/icon_smile.gif[/img]

[pduggan]

<BLOCKQUOTE><font size="-1">quote:</font><HR>Have I lost you yet??<HR></BLOCKQUOTE>

Not sure, Karim, but, if my latest attachment helps anyone to explain things (all of these rotate about their centres but look awfully like Egg's top wave form), please post it here!



Peter</p>



Peat Stack or Pete's Tack?</p>

[mckinney]

The brushes that Peter Duggan and others have been using for the ropes *are* rotating about their centers as expected. You are seeing the "peaked wave" effect because the brush rotation and the motion of the brush are in the same plane. These brushes must be rotating counter-clockwise. To make the rope look symmetrical, the brush would have to rotate about the axis of the line drawn (i.e. into and out of the page). A rotation in that third dimension would cause an apparent sine wave motion on the drawing plane and create the more realistic rope patterns you are after.

A way to understand the effect of rotating the brush in the plane of motion is to visualize a dot on a circle. Imagine that the dot is moving countner-clockwise around the circle and that the circle is moving left to right at the same velocity that the dot is moving around the circle. When the dot is at the top of the circle (12:00), it will effectively not be moving at all. As the dot moves to the left of the top of the circle and the entire circle is moved to the right, the apparent motion of the dot is nearly straight down. At the 9:00 mark on the circle, the dot is moving down at the same rate the circle is moving to the right and the apparent motion is a 45 degree angle down to the right. When the dot nears the bottom (6:00) part of the circle it's apparent speed is twice that of the "speed" of the circle and it's trajectory is the same as that of the circle itself. The process mirrors itself as the dot climbs the the right side of the circle and comes to a stop at the top again.

The first rope in Peter Duggan's most recent post can be used to visualize the effect. It has two rotating dots - simply focus on one of them. You can see the effect of the dot "velocity" by looking at the dot spacing from one brush rotation to the next - i.e. the dots near the top peaks are mush closer together because the apparent speed of the dot is very slow. Near the botton of the rotations the dots ae spread apart because their apparent speed is much faster.

Egg's latest post shows clearly the path of the dot (the upper line with the peaks). To create a sine wave motion, the dot must move up and down only (i.e. use only the vertical displacement of a dot moving around a circle).

Hope that wasn't too confusing.... I'll draw an example when I've got a moment or two in the next couple of days...

- Pete