After working with DC a while I've found these rules of thumb. You have to arrest me if I am wrong. One sibling cannot make instantly use of a custom attribute value from the others, a redraw is needed One sibling cannot make use of a custom attribute value within itself more than in one attribute at a time. If you make a custom attribute in the parent refer to a siblings value you can use it simultaneously in all parent/sibling attributes. So the safe thing to do is to route a value from one sibling to another via a custom att. in the parent. Even so if you want to use it locally in more than one attribute inside one sibling.

Has anyone figured this out, by chance? I know it's been a while, but I have an application for it, and I'm racking my brain trying to figure out how to make it work. For mine, it's more of a "snake" component, where the user begins with one part, then selects 1 of 3 interactable "buttons" that creates a copy in one of the directions, turning the copy to fit. The link below should make sense of what I'm saying. Thanks in advance! BT http://forums.sketchup.com/t/help-making-copied-arrays-with-distinct-attributes-for-each-copy/31487a @chris fullmer said: Ahh, yes I was thinking about this Jim, and I thought it was simple. But I started playing with it and in about 5 minutes I saw just how not simple it really is. The random width means that the amount of copies will vary depending on the width of the books. That is tricky! I had a thought on placing them though. Each X should be the X of the previous (called "LastX") + the LenX of the previous (called LastLenX"). So what about each component passes its X and LenX out to a custom attribute in the parent component. So the next copy created would look to "parent!lastX" and "parent!LastLenX" to figure out its new X. Then after it sets its X and LenX, it writes its current state to the parent, overwriting the previous states and the cycle continues. So it runs like this Copy0 X=parent!LastX + Parent!LastLenX (These values will be zero since they have not been filled yet) LenX=3 (randomly chosen) <<<Sends X to parent!LastX>>> (which is 0) <<<Sends LenX to Parent!LastLenX>>> (which is 3) Copy1 X=parent!LastX + Parent!LastLenX (Now this reduces to 0 + 3) LenX=2 <<<Sends X to parent!LastX>>> (which is now 3) <<<Sends LenX to Parent!LastLenX>>> (which is now 2) Copy2 X=parent!LastX + Parent!LastLenX (Now this reduces to 3 + 2) LenX=4 <<<Sends X to parent!LastX>>> (which is now 5) <<<Sends LenX to Parent!LastLenX>>> (which is 4) Anyhow, its all theoretical since I am not on an SU7 enlightened computer at the moment. I hope that makes sense. In my mind it does, but I'm sure it will probably fall apart due to something I've forgotten. Anyhow, think about it Chris

Dhruv was responded to here: http://www.sketchucation.com/forums/scf/viewtopic.php?f=289&t=14771&p=112038#p112038 Please respond there if you have additional comments on his post so we can keep the forum tidy. Thanks!

Wouldn't let my upload my mp4 file showing my onClick rotations so here's a screenshot of the tap at least. Be nice if you got a prompt or message that it was the wrong file type or something rather than just not working...does it have to be a gif or something? Doh! [image: BIrj_PuretecTriplaT3_screenshot230322.jpg]

The intention of this exercise was to move one component and have it immediately affect to color of another by reinterpreting its X value. This result is a failure. It requires an intermediate set of actions. The user must select the interact tool and click on the component. Perhaps someone with more knowledge and better skills (which would be just about anyone here) will see something I missed. The three sliders, SliderR, SliderG and SliderB's X attribute is evaluated as: =NEAREST(CURRENT("X"),0,0.015748,0.031496,{...},3.984252,4.000000) ColorCube Material=CONCATENATE(INT(ABS((SliderR!X/IncAmt))),",",INT(ABS((SliderG!X/IncAmt))),",",INT(ABS((SliderB!X/IncAmt)))) ColorCube onClick=REDRAW() Slider DC v1.skp

Hi Chris. I like your model. It's a nice example of the DC potential. After all, the main question is how DC can improve our productivity in the daily work. The DC feature is meaningless if there is no improvement, but a waste of time. It's a fancy thing to open/close door, but it should serve a purpose. In my work as a constructor, I always have looked for solution to automate repetetive tasks. I have tried many different CAD solutions, and many are quite good in doing autotasks . But I have never trusted their final output when it comes to details, mainly because other people have set the parameters/conditions. In SU I have full control of the output and details, because I am making the parameters fit my way of thinking/solving problems. How I see it, the DC- feature sets me in a position to save a lot of time with deploying/shaping components, gives me full control over the parameters and at the same time gives me the good "look what I have done" feeling when things work properly. So hopefully there will be a lot of useful DC out there, which can be shared. One last thing. I miss an option to lock access to attributes and even to components, preventing people from altering code. Keep up with the good work Chris! Dagfinn

and much more beauty than Fredericks ) though Fredericks logo is unbeatable

A while ago I wrote the Plugin SCP. There you only would need to add your chat URL to the favorites list. azuby

I've run into the same problem Chris has when I've been trying to help others get their models straightened out. I agree that forcing geometry to reside on Layer 0 makes much sense. pbacot, when I am teaching new SketchUp users about components and groups, I use an analogy to try to explain how they keep geometry separated. Think of a sandwich. A couple of pieces of bread and some stuff in between. Lay a great big dill pickle on top and the juice soaks into the bread changing the sandwich. Wrap the sandwich in cling wrap and the pickle juice can't soak into the bread. Assigning a layer to a component or group effectively turns that cling wrap into some of that new "cloaking device" fabric. Unchecking the visibility box turns on the cloaking device. The geometry remains where it is though on Layer 0. It's kind of a lame but it seems to get the point across.

Chris, Long time no see 'eh! I love the images, very cute and fitting for the subject. The popping primarys and the whimsical outlines are very effective. Not sure I would say the same on a different project with less flora and different subject matter, but it all seems very original to me. Very cool man.

Hi Chris, hi folks. O.3" is 7.5mm (75mm is 3 inches). Other than that, this is an excellent tutorial. How about softening the upper edges of the stone tiles ? Just ideas.

Chris, how very nice...and useful, thank you!