jmcewen

I have similar but different problems. Perhaps we could share a solution. My workspace is different from typical because of how my screens are arranged and to minimize distance I travel when moving from other screens into the OIP. OIP is on the left and navigation is underneath it because Navigation is most useful to me when it is the same width I want to keep the OIP. They hold position and width as I expect, but they are also bot at the default 1/2 screen height. If I were to change their heights, I might find that they revert spontaneously to a default 1/2 height. My toolsets are on the right. Basic in the top right, the assorted toolboxes immediately below and Attributes underneath, with the Attributes height reduced to the minimum that it will allow, Basic stretched until I can see all the tools, and the assorted toolboxes set between with all the remaining space to minimize need for scrolling within any given toolbox. Regardless of how many times I save this position, the Object Attributes reverts to being 1/2 of the height available every time i reopen Vectorworks, squashing the other toolboxes to where I need to scroll to find anything. This leaves dead space at the bottom of the screen because there is only so much info that the Object Attributes can display. The state it reverts to necessitates scrollbars that just take up more real estate, and it leaves 3 inches of unused space at the bottom of Object Attributes. How it usually looks vs how I want it to look: Also my resource manager is pinned under the ribbon. Upon restart it collapses the folder navigation part of the screen on the left until it virtually disappears, and squeezes the open folder to a narrow sliver. I end up with a giant space for a little thumbnail of my resources and I need to rearrange the screen to find any other resource I am looking for. Similar problem with the text size drop down. I usually stretch it to where I don't need to scroll anymore at the beginning of a session, but it seems to revert to showing 4 rows when I restart.

I may be missing something, but in the 3rd picture you appear to be set to wireframe rendering. If you set it to shaded do you get different results? What are your view options for Layer and Class set as? Is the fence's layer or class set to greyed in either? Do you have class or layer visibility set to grey others? There may also be some other complicating factors here. Is the fence a group or symbol? Class of the group may be a different class form the components of the fence, so all of those visibilities need to eb set to visible. A screen shot that includes your object info palette and your navigation palette and might help us clear some of this up,

If your class list is short enough, you can drag the bottom of the dropdown menu to show a longer list. If the whole list is visible it will not jump. I have been experiencing this as well and that is the best workaround I have found so far. it is funny how it delays just long enough for you to think it will not do it htis time, then the jump triggers the moment you click.

Right now, the engineer is regenerating lines himself. I am happy to let him do so. It is not my bill to pay, and I still have a lot more of this project to model, plate, and document. It sounds like he is using RISA,which is another Nemetschek product. I would expect the two should have a pretty direct line of communication.

You could even make an array of symbols so you only need to model one or 2 repeats.

I have only barely dipped my toe in Marionette, so this is probably not a good project for me to learn on... Drawing those nurbs curves does not seem too ridiculous. I am just not sure that it gives me what he needs. I have occasionally drawn things that got sent to engineers in the past-- did they just regenerate my geometry each time to satisfy their software? or is my current engineer just using different software than what i have encountered in the past? I am hoping some of you who deal with engineers more often than i do might have some insight about workflows. He had me export as Revit and DWG Is there a good way to go about generating geometry that satisfies these needs in the first place in the future? Is there a workflow that makes sense for the engineers needs but still gives me what I need for making construction drawings? For example, if I use the split tool to miter a corner on a structural member, I end up with 2 solid sections, and I would guess that eliminates the centerline that the engineer is looking for. However if i just make but joints or overlap the 2 members that doesn't look correct in my construction drawings. I also have a lor of roll-bent structure in my drawing. If I bend the resultants form the structural members tool with the deform tool, i will end up with a generic solid, and i mam certain that would eliminate the centerlines (if they exist in the structural members tool-- i have been assuming they do up to this point.) As a workflow for my breakdown models, it generally seems fastest to me to frame out the face of something in 2d, and extrude it to the appropriate thickness, either as a series of extrudes representing each member, or to draw a polyline and clip surface for the voids within the frame, then use the split tool to break it down to the constituent members. Regardless-- usually my extrusions are not usually built with the extrusion direction along the long axis because of how i generate them.

I have a model that needs to go to an engineer for calculations. When the engineer imports my drawing into his software he gets 3d polys, which is not what he needs. He wants me to eliminate all my fascia and details to just have the framing weldment, but he still wasn't satisfied and he still got just 3d polys. Now he is asking if I can generate centerlines of all my framing members. I didn't use the structural members tool because I couldn't get to small enough stock for what I needed to frame. I think using this would have gotten what he wanted but I am not sure. What I currently have is extruded rectangles representing my framing members. Is there any way to generate centerlines that does not involve me completely rebuilding my model?

Thank you for the stab. Each boltset is a single object made with the 3d fasteners tool. it only has one class assigned to it-- hardware. And the holes are just holes, no faces in the holes. It is not only happening in this class change either. one of my standard procedures is to set up a sheet with all my ortho view of an object and then duplicate it thorough the process. So at first i have an overview of the complete object on the first plate. I duplicate that plate and its constituent viewports and turn off all visibilities except the framing. Then I duplicate and add fascia. Then I duplicate and add trim and details. the same thing happens throughout the process regardless of what class I am looking at. This has always worked for me in the past. Not sure what has changed. it is as though it is leaving a picture of the old viewport behind until I update again. I need to get out my other computer and see if it is doing the same thing. Side note-- is Gadzooks a Mojo Nixon reference for you or just a general everyday sort of gadzooks?

I realize I may not have described the issue well. Here are some screenshots. hardware visible: updated, and the hardware remains as a lighter green ghost image behind the holes. update again for it to completely disappear without changing anything.

It isn't the same problem but it has very similar symptoms. I don't have any image effects activated. It almost seems as though it updates the viewport, but an image of the shaded view is left behind with the old and the new update occupying the same space, maybe?

Recently I have found myself needing to add and change class visibilities during revisions on viewports quite a bit. I find that these changes do not always get executed on the first update, but often do work after making no further changes and re-updating. It is most noticeable when turning off a class visibility. For example, imagine if I had a hardware class activated on a viewport, but then I realize I need the hardware to be invisible so I can dimension the hole patterns. I have a standard array of ortho views set up already, so I highlight all of them, I turn off the hardware class, and I update. The image of the hardware remains, though it often appears in a lighter color or just peaking out from behind the updated viewport. Then I update again immediately and it usually (but not always) comes out as expected with the hardware now invisible. If not, I update again and it probably works. This happens very commonly if i am updating more than one viewport at a time. It seems perhaps a little less frequent if I update a single viewport, though it is still happening most of the time. My current file is sitting at 155 viewports so far (I realize this is not a huge project for some of you but it is a lot of viewports in a single file for my typical workflow) and I don't relish updating them individually twice or needing to go back for an extra proofing session to find the ones that did not update properly. Is anyone else seeing this or is it just me? Any thoughts on remedies?

I sometimes have this happen with containers where the container is one class, but items within the container are not the same class. Typically I assume if any class in the of the object or the container that holds the object are set to invisible that the item will be invisible, and that both the contained object and the container must be visible for the item to be seen. Sometimes it feels like it doesn't work this way though and items show that should not show. This can get pretty complicated if you have objects inside of groups inside of symbols. So 2 takeaways-- are there any containers involved in your disappearing content that may have been assigned to the class you have turned off even though the contained objects are properly classed (possibly helpful) sometimes things are screwy (not helpful at all)

Also you don't need to leave Vectorworks to fix your image. You can do a relatively simple edit like this in the bump edit menus by just moving the sliders to change exposure, contrast, highlights, shadows, etc.

Sorry if i am late to the party, but even within the symbol I would still use the classes, I think. For example if you had a car symbol, it could be modeled with doors that were closed, and those doors could occupy a class called "doors-closed" You could have identical doors but in the open position in the same symbol, and they could occupy a class called "doors-open." This gives you the potential for many different looks for the doors: Turn on "doors-open" and turn off "doors-closed" and the doors will appear open Reverse the states of the 2 classes and the doors will appear closed. turn off the class "doors" and neither state will appear. Your car will not have doors at all set one class to greyed and the other to be visible and you can see the car with a door and have the alternate state be viewed but with less clarity. Unless you are just talking about drawing a car once and then filling a parking lot with it. In that case, a symbol is exactly what you are looking for. Draw it once, convert it to a symbol, and the data will only take up the space of the first model and every instance beyond the symbol definition will only be coordinates for the location of your symbol insertion point rather than the entire geometry of the model for each instance. Here is a nice intro to symbols. The Vectorworks University classes are pretty good and are broken into small enough chunks that you don't need to wade through a huge lecture to find the snippet of info you are looking for. https://university.vectorworks.net/mod/overview/view.php?id=2200 As a side note, I took longer to realize that container objects like groups could have classes of their own, so if you are using your classes for other things like individual materials or systems you can also group any item and give it the opportunity to have another class assigned to it. If either class visibility is turned off the object will not be visible.

Thank you! I was hoping it was something basic like this. The relative view was activated. I just don't understand how it seems to randomly trigger, though, or how it triggers so often. Is this setting tied to a hotkey somewhere? I don't spend much time with my cursor inn that area-- almost all of my common views are triggered by hotkeys on my mouse, and I don't use saved views very often (though that might save me from some of my rotated plans.) I can see accidentally hitting a bad keystroke though if it is tied to a hotkey. I need to start looking for how I am triggering this, but at least I know how to fix it properly now.