Marissa Farrell

Hi Phil! There are a few different reasons why we can't tell you guys sooner than we do, one VERY important one is that we really want to be sure that the feature will be complete and ready for the next release. Had we told you sooner before it went through the whole development process, including testing, and then realized it wasn't in good enough shape to ship, then we would be falsely advertising the feature and there would be a chance it wouldn't even make it into the release. Trust me when I say that @JimW, myself, and many other people we work with would love to be able to share the things we are working on throughout the whole cycle, but that could hurt our reputation in the long run when things don't go as planned. We don't want to hype you guys up for something that won't be ready. Although the engineering team is VERY talented here, we still have to take some precautions when tackling a difficult task, such as multiview, to make sure you guys will have a good experience with it.

@Pat Stanford & @Stephan Moenninghoff I've been through all of those functions and unfortunately I still haven't figured it out. The locations returned all appear to be on the layer plane instead of in their 3D location projected to the screen. I don't think the function Pat mentioned takes the Z value into account, and I've struggled to find anything that can help offset that (which is why I was playing with the Layer Ref functions)

I'd be more than happy to help you out with this project, as well as guide you with any other Marionette projects you may have. I'll take a look at your progress when I get back into the office on Monday.

Well I got close (but not close enough, no screen plane) I had to write the new nodes. I'll keep looking for a screen plane solution, though. just not sure one exists. MoveObjToPlane_v2017.vwx

I'm not certain of a current node that will give you the screen plane location of a 3D point. I do believe, however I haven't tested, that you can use the Move node to move the text to a 3D location. It would still exist on the layer plane, though, instead of the screen plane (if that even works) I can look into this more for you.

This is what I see in Top/Plan and Top when I run the network I shared to you. If you're not seeing the same, then maybe your version is missing some fixes. I'm not positive when the node was fixed, but it may have been as early as 2016.

Yep! So you're using an older Path Extrude node. Depending on what version you're running now, you could replace it with a new one. Otherwise, I replaced it in this file with the current one. I hope this helps! frame_3_MFarrell_v2017.vwx

@the frog Could you attach the file, or send it to me privately? You may be using an old node for the path extrude. There's something slightly special about EAP objects, but we solved this a while back. If I can look at your network, I can see if it's the same problem.

Fantastic! Glad I could help

This may be a limitation at the moment. I have yet to see any other (simplified enough for ease-of-use) web viewer capable of what Theta does. I've been through many different options to find the best way to share my 360 images, and all of them perform how we do with Vectorworks. I don't think these limitations are due to our lack of research into the topic.

I also prefer the second version The plus side of all of this is since you get that jpeg file, you can use whatever viewer you like, my example was as simple as importing it into a different viewer. I'm still 100% a fan of the feature overall. I really like that I don't have to do any extra work to share my pano-view after exporting it with Vectorworks, it's just nice to know there are other options.

Here are two different viewing modes using Ricoh Theta's app, which I use to share my 360 photos that I take... The first mode (Straight) is similar to how we display things, and the second mode (Mirror Ball) will round the outer edges of your current view to prevent some of the stretching you see with Straight. EDIT: Neither view will give a perfect representation of what you would see in real life, the mirror ball helps a little because it sort of emulates a fish-eye view which may feel more normal for someone to see.

(Just so everyone knows, this is probably my favorite new feature... I've been anxiously waiting for it to become known to you!)

Haha! I actually wrote that specifically for this. I have a bad habit of not saving my custom nodes. I rewrite them often.

In addition to Pat's script, if you wanted to opt for a Marionette solution, the attached file includes a wrapper which accepts user input for the range of length of the lines you want selected (these values can be changed in the OIP). My default example is lines greater than 1inch to lines less than 5inches. All objects on any layer that can be manipulated will be selected. (If you have Show/Snap/Modify others on, it will select lines on all layers that are not grayed or hidden.) If you were feeling risky, you could swap out the Select node for a Delete node, but I wouldn't advise that unless you were sure you weren't going to delete anything you didn't want to, which is why I assume Pat didn't include deleting in his script, either. DeleteLinesInLengthRange_MFarrell.vwx

Agreed! It's a huge project, but we're trying to figure something out.

I haven't done it, and I'm not sure if it's possible, but I won't say for sure that it's not. The problems I can see are that you would probably need to return more than one handle of objects to manipulate and the script won't export the relationship between the objects (in terms of positions) which would make the Marionette incorporation even more complicated. If you want to take the dive, I can try to help, but I don't think it's a worthwhile venture for Marionette. A traditional plug in object makes more sense, to me.

My understanding is the same, you are correct as far as I know. I would suggest using the 'contents' node rather than having to ungroup the control geometry to use it, though.

@Kevin McAllister Try playing with the contents node with your control geometry. I can get it to work (a little); still looking into it. EDIT: It doesn't seem to work with nested groups, but if get all your stuff into one overall group, I think it will mirror. (It doesn't like all shapes, though)

I would like to second @Patrick Winkler's opinion on this. I'd also like to say that Patrick's assumption on where the network flow begins is accurate to a degree, from what I've learned from smarter beings than myself. This could (and likely will) introduce problems when networks are non-linear and branch off. I also have very little understanding on where networks begin from when wrapped, so using wrapper nodes could also introduce problems.

Right now there is only one Mirror node, I'm not sure of the limitations on it. What kinds of 3D objects are you trying to mirror?

If it went to tech support, it would just end up on my desk anyhow. Which version of Vectorworks are you using, including Service Pack? I can take a look at that build locally and see if I see it there, otherwise I'll second Pat on it being a mystery.

@RickR I'm also unable to reproduce the problem you've stated. However, I'm glad you got it working pasting it into a new file!

I can second this, and he's not even my boss.

Also, just as an additional resource, the Vectorworks help files were updated for Marionette in 2017 to include two examples of network creation in Marionette. I designed the more advanced one which should help to show what you need to know. I believe this link should bring you to it. http://app-help.vectorworks.net/2017/eng/index.htm#t=VW2017_Guide%2FMarionette%2FMarionette_Tutorial_Creating_a_Simple_Cabinet.htm%23XREF_41117_Creating_a_Simple&rhsearch=marionette&rhsyns=