Negative Extrusion E3d
Libby Cowen
I am having a hard time getting a feature class to consistently extrude downwards in ArcPro 2.2 in a local scene. I was able to successfully extrude downwards with a GCS NAD 27 coordinate system, but I cannot seem to consistently replicate the extrusion regardless of the coordinate system (scene and all layers have the same coordinate system).
What are the current limitations with downward extrusion? It would be very useful if this functionality was more robust. It looks like downward extrusion is also not a supported functionality for a feature layer in a web scene (positive, upward extrusion works fine).
Thanks for the reply. I was using the same process using base height and an expression with navigate underground enabled. I ended up doing the extrusion work I needed in a global scene vs. a local scene, and it appears to be working fine so far. I have then been reprojecting the exported multipatch to ingest it into a local web scene.
After checking again, I opened the old Pro project I was working from, and I did not have 'Navigate Underground' selected for the elevation surface in the local scene I was working in.... Glad it's almost Friday
I had trouble getting the text to look the way I wanted it to, so I actually created it in Illustrator, where I then converted them to paths which I imported into Rhino as curves. The string of textcurves basically arc around the top of the medal.
I moved the curves above the model, and then used the Project command to project the curves onto the surface, as you can see here:
I want to extrude the text inwards, but the bottom of the inside (if that makes sense) should be at the same angle as the surface is on. If I extrude the letters straight down, one part is much deeper than the other.
Hi Rick- if you offset that beveled surface inward, (OffsetSrf Solid=No) to the depth of the letters, you can then extrude the letters through this surfface and then BooleanDifference the offset from the lower ends of the extruded (solid) letters. Make sure the normal direction of the offset (Dir command) is upward. Then your extrusions will all have the correct bottom shape and can be used to BooleanDifference from the original bevel. Shout if you want an example file.
Where is the setting for negative extrusion in Blender. In case I have the name wrong, this extrudes non-existence, ie you extrude back through an object to make a hole. I can make a pit (hole partway into material), but if I go through the other side it just starts to positive extrude from the back. I can't find the option to remove mesh and make holes.
Is there a preferred/most efficient way to perform an negative extrusion (?intrusion? LOL) involving polys on the edge of an object? See image 1 and 2 attached for before and after for an example geometry change (changed polys marked in red).
My current workflow involves making the extrusion, then deleting the double faces along the edge, then bridging the gap left by the lower portion of the exterior face. Seems like a lot of work but maybe that's just how it's done?
This seems like a modeling issue where pre-planning a topology would be more beneficial, but it's understandable sometimes even planned geo needs random changes, on demand. The operation you're trying to accomplish is something very common in solid-modeling (CAD), but requires a different strategy in surface modeling.
Personally, I'd probably delete the existing faces, and use Shift+translate to pull out new polys from the border edges where I need them, then collapse/weld the verts at the junctions. Not sure it's any more efficient than how you're doing it, but ensures a clean topology, and no worries about overlapping faces, etc. However, there's other techniques that would depend on if I knew ahead of time the profile of the shape. Also, there's some maxscripts that can cleverly create topology in this way, but that's off topic.
I totally agree with the sentiment about pre-planning, and I try to do that as much as possible. I generally start with the smallest common dimensions in XYZ and work outwards. But sometimes the geometry just gets complicated, or more commonly I just work myself stupidly into a corner! Most of what I'm doing is real-world architectual visualization, and the the stuff tends to not be simple box shapes.
Now that I'm grasping the basic modeling tools, I'm finding that it's hard to know the best philosophy for models regarding making a model out of just one starting primitive vs attaching geometry to the base model in elements vs just having a grouping of separate objects. If anyone knows of a good discussion of the pros and cons I would LOVE a link to that!
If you have a couple hours to spare sometime, you might try a few of my tutorials for 3dsmax (most of them are less than 30 minutes, but they all have different techniques overall). While they deal more with complex mathematical surfaces, there's a lot of techniques throughout some of them that I think you'd find very handy for this kind of thing, including strategic meathods to achieving certain problem surfaces. Let me know if you'd like to see some specific video examples of certain objects, and I'll try to find time to create them for you, with the best possible solution I can think of, using the standard tools in 3dsmax.
I'm all self taught, mostly from lessons on Digital Tutors (now Pluralsight). Some really great lessons there though I find that they tend to do the same thing almost all tutorials do: work on a project that gives a general but somewhat shallow coverage of many different tools. I think the Max community would really benefit from 2 categories of education:
I've been focusing on modeling philosophy in many of my tutorials, dealing with mathematical objects for design and printing. I'll be making a series of 3dsmax 101 tutorials very soon, which will examine all of the Graphite modeling tools, and when best to utilize them in a general workflow. There's tons of tutorials focused on the modifiers lists, but not many for the Graphite tools. Unbelievably, many modelers ignore these invaluable tools, because they simply don't like the ribbon UI, or feel some of the tools are redundant in the Editable Polygon command panel. I believe that's a very self-limiting outlook, as the Graphite ribbon contains tools that are extremely powerful, if used properly, and assigned to hotkeys.
I'm interested to see what you come up with. I think there are a lot of tutorials in the "101" category, and also lots of resources for more expert users (? graduate students to carry the metaphor). But what seems to be really limited is the "201" level tutorials. That's where I'm finding myself at this stage. I have the basics down pretty well but it is a real stretch to learn more just because the topics I'm looking for now are mostly aimed at experts or industry professionals.
FYI regarding the model from the original question: I decided to scrap and start from scratch with more/better planning. Maybe it's just me, but sometimes I get to a stage in a model where it becomes apparent that it will take more work to keep grinding versus starting over. I'm sure nobody else here has that problem though
Not turning on your heated bed can also help, because heat makes filament softer / saggy and therefore the radiation from the bed can also negatively influence overhangse which are relatively low on your print.
That was long ago on a umo right? I mean, umo has a larger hotzone and a overpowered feeder, that probably allowed you to push that filament with less heat. Have you tried to repeat that ever with a um2? Or maybe with the new flame hotends?, I forgot the name, the ones that has the flame nozzle symbol from 3dsolex. Maybe with that and the um2+ feeder + gudo fan system.
Btw @prevostc that's a good trick. Print stuff with big infills, like 50% or more, so each layer has more time to cold-down. Also, since the normal print order it's outside-inside, all that extra time 'inside' will help push air on the print, so everything get's to the 'ok' temp to avoid overhangs.
So my tips. Fans, ok, as much as posible after layer 1. Then print more than one at the same time, or print each part with larger infills. Ofc, you can print slower. It just that I prefer to print fast and use more powerful fans so I don't waste time or filament.
I agree but also disagree. The viscosity of the material used changes that 'pull' effect. Faber for example it's less viscous than colorfabb that it's more prone to get dragged. So the brand/material used changes the approach for each print.
Also I think that it's important to cool the material as fast as possible and because the hotend it's always moving (unless you print small objects) you need not only air on the nozzle but around it. That's why I use 13cfm fans and they actually push much more air than um2 fans. Ofc they add more vibrations at high speeds. So there's always a downside on most sollutions.
Another factor in play could also be material. For example ABS hides irregularities better than PLA. Does anyone have a preferred PLA brand for smoothest surfaces? (could also be just color related..)
As neotko suggested - I printed two UMO robots at the same time. This gave one time to cool down while printing the other. Both robots came out just as good. Mainly I needed to do that for the 2 antennas on top of the head.
An extrusion is a solid volume in 3D space, defined by a "base" face (Face) and an extrusion vector. This solid is defined by the prism generated by extruding the face in the direction and magnitude described by the extrusion vector.
A positive extrusion vector produces a solid whose orientation is front facing out, with all surface normals facing out. A negative extrusion vector extrudes in the direction away from the surface normal of the base face, and produces a solid whose orientation is front facing in, with all surface normals facing in.
The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.