Thanks to all, especially to @DaveR, for helping me resolve this issue. I modified the model to stay within the county limits and it eliminate the jagged lines in my original post.
Today however , working on a different model, I encountered the same issue. I was almost done with the initial model of a Thos Moser hall table. Then the same problem occurred in full after I copied and scaled the table top and legs up by 1000 and did a trim/keep. The model looked fine (see attached image 1) at the x1000 scale. Then when I deleted the 1000x top/legs copies and did a zoom extent to normal size model I observed the second image. Jagged lines are very severe. Any help would be appreciated.
Let me see if I can also upload the skp file.
I wonder if your scaling up and down has left something (a single bit of an edge, perhaps hidden, is enough) somewhere far from the origin. Also, you don’t show the model axes in your screenshot - is the whole model somewhere far from the origin? Also the distance to the SketchUp model origin point increases the model’s extents.
Adding the SU file now.
I checked as much as possible for stray bits but maybe another set of eyes can see better then me.
Sorry for delay, distracted watching ball games. Why does the stadium screen TV say “make noise” when only cardboard cutouts are in attendance ? But I digress…
You’re working with the Camera set to Parallel Projection and it’s a million miles from the model. Set the camera to Perspective and hit Zoom Extents. Save Parallel Projection for display. Don’t use it for modeling.
Thanks again Dave for your help. Changing the camera did the trick. Good to know. I may have been in parallel projection for most work and then did a perspective scene without changing camera.
@DaveR,
Surprised to hear that since for so many things like adding guidelines accurately I tend to switch to parallel projection. I use a keyboard shortcut to switch between parallel proj and perspective along with shortcuts for standard views. I do value your advice so I will see if I can break my habit.
SketchUp is designed to be used primarily in Perspective. There shouldn’t be any need to switch to Parallel Projection to place guides. There are a few things in which Parallel Projection is useful such as when trying to control selections of geometry but for the most part, modeling in Perspective is generally easier and it definitely creates fewer problems with camera position.
The problem with modeling in parallel projection is that the camera can end up far away and you can’t tell by looking at the view. This is inherent to the projection because the view rays are parallel - moving farther away has no effect on where they map the model to the view. Zooming changes the magnification of the model on the view, but not the camera position.
Once the camera gets far enough away, you start to run into issues with the way graphics systems handle distance from the camera to the model. They break the distance into “z-bins” and can’t distinguish between things that land in the same bin.
By the way, you might want to get in the habit of purging unused geometry and keeping raw geometry untagged. This reduces the file size about about 77%.
What do you find about parallel project that gives you more confidence in accurate placement of guides? I use guides all the time in my modeling (100% of which is in perspective view) and they are exactly where I want them with very little fuss. In particular, I often use SketchUp’s ability to enter exact numeric displacements when locating a guide. I’ll choose the Tape measure tool, click-and-release on an edge or existing guide line, then move the mouse in the direction I want, then let to of the mouse and simply type in the displacement (from the initial click position) that I want - say 2.35 - and press Return. The guide is created exactly 2.35 inches away from where I started.
I tend to be less accurate when adding guidelines in parallel mode, like tracing an image, you’re not sure if you are in the same plane unless viewed from aside in perspective. I do switch regularly for selecting purposes, though.
@DaveR
Another follow-up on the subject of model space organization to limit distortion when orbiting. In an earlier reply in Aug 20, you provided a sample of how you typical set up your model space using a furniture case piece.
Exploded assembly is located back and to the right. This set back is just enough so a right elevation scene can be made of the main model without the exploded assembly being included.
2 and 3 view elevations of case component parts are located above (and maybe slightly back) of the main model. Again, this location allows for orthographic scenes (top, front, right, left, back ) and perspective view of the main model without the camera viewing the other elements in model space.
I am working with a fellow woodworker who is designing a case piece. He has the same 3 elements in his SketchUp model space, plus an additional one showing an exploded view with the sub assemblies intact.
To help keep the model space compact (not “spread the model out across the county”), he has these elements stacked one above each other in model space, with the main model at the origin, the 2 exploded assemblies, and component 2,3 view elevations above.
He made components and defined tags for each of these main sections of the model. So the main model is a component and has its own tag, the same for the 2 exploded assys, and the entirety of the 2,3 view elevations are a component with tag. Of course each of the individual parts are components with tags. (similar to shown below using your case piece model) Its a very structured approach, albeit extra work to make these extra components and tags for the 4 sections of model space. It does however allow for compactness since for any scene, visibility of the other main sections of the model can be shut off.
Aside from the extra work in setting up the extra components and tags of the main areas of the model, do you see any other downside to this model structure ?
Not really. It’s just more tags to manage. I’m not spreading my components out across the county. They are still in a fairly compact space. I’ve seen “plans” done in SketchUp where copies of every single part of the model are strung out along the red axis and that gets excessive and really hard to manage when the author used groups instead of components.
Your friend can certainly use more tags if he wants and he can keep all of the objects in a big wad at the origin. As long as it works for him, that’s the main thing.
In my OP, it was my design that was clearly spread across the county (maybe even over the state lines). Based on your feedback and examples, I now have better model organization with a more compact approach, and no more distortion when orbiting. I do sometimes have to tweak the location of these main model areas when scenes are captured. That usually is enough so I dont need to define extra tags and components for each area. However, the approach I described above certainly could be used to isolate the model areas for scene capture. As I said, takes extra time, components and tags. Yes, one could take this to an extreme and create a messy pile with everything at the origin.
I guess I’ve got my “system” worked out so it’s efficient for me. That’s the thing everybody needs to do and it’ll surely be different for each.
One key thing for me is to avoid putting anything in the standard view sightlines to the assembled view of the model. I rarely ever need to adjust the positions of other instances of the components when I’m creating scenes.
Dave,
Great picture. You can really see how the sections are close but isolated from each other enough for a perspective or ortho scene. If you set the model in the right location, really no need to define extra components and tags for each model area.
At first, especially with different furniture types, sometimes it takes some time to know where the best locations are. Is there a big difference with modeling of a table, chair or bed.
Do you keep seperate SketchUp templates for each furniture types so you know where to place the model in space, and define pre-set scenes ? Similarly, do you set corresponding Layout templates for case, table, chair etc type furniture pieces with pages pre-defined ?
No. Not really. I guess a bed, being a larger piece of furniture requires a bit more space than that wall cabinet but it’s not a big deal.
Nope. I don’t have scenes or tags preset in my SketchUp template. While I’m modeling I’m thinking about the tags I’ll want and then the views. Some pieces of furniture require more detail views than others. When I create the various scenes it’s generally done in the same sort of order; 3/4 view, Front 2D, Right 2D, Exploded… It’s mostly done in the order I will need when I create the viewports in LO.
As for LayOut templates, I have a bunch of them but they are client or audience specific and not determined by furniture type. They contain things like title block info and page borders and numbers as well as text settings and some auto text fields.
