Hi. Is there a short cut to convert the stepy-edge's to a circular shape in edit poly mod ? i started with a cylinder and then i focused on a quarter of it. modeled the first quarter and then using symmetric modifier i got it to this level. since i have to continue the inside part and its flat, i have to convert the last edge to this round shape. is there a short-cut to make it soft ? i think im also able to use "shapes" to make the insite part. (yes) but couldn't be better to model this all in one part ? also let me ask, is there an easier approach to the out part ?
I had to cheat a bit with an Edit Poly, to get a top bevel, separate the drain area, and fix a bad vertex. I hate using Edit Poly because it breaks parametric editing... Edit Poly is vertex-count-dependent so if you edit stuff below it you're more likely to break things.
Anyhow, you can go through the modifiers one-by-one to see what I did. Just turn off "Show End Result" to see the steps.
thank you so much for the info and the trying it ! i got it. i used an eclipse, duplicating and scaling it to match the final shape, then i created a surface, then i customized only a part and using symmetry it just fit really nice together. love this section. very informative post. well done everyone.
Hi. I'm new to the whole world, i just started studying it, but i was wondering if you could give me some advise. I was trying to model this pillar complex in maya for a game environmental asset, based on the reference. (pic 1)
I modelled one thicker, and one thinner pillar with the decorative top part as extrusions, and i lined them up as they should be, then booleaned them to make a union. As you can see on the picture it is a pain to clean up the topology at the joints of the decorative top part. (pic 2)
I tried a different approach, which has a nicer topo, but it just doesn't look as nice as the original. (pic 3)
Is there any way to model this correctly? Or should i just import it into ZBrush and Dynamesh them to weld them together instead of doing a boolean and than re-topologies it after and decimate it back?
You might get some "fitting" radiuses for two different sized circles with different amount of vertices repeated in this circular way. But as soon the radius changes because of the profile you have to be very lucky to get some fitting vertices... The green marked may fit; but the red ones don't and also may produce triangles (or even n-gons..)
It's the same with this 32 and 16 vertices cylinders with a 1.5 bigger top:
The side edges just cross over in different angles all over the place.. so this kind of topology is a really nasty..
Hi, thanks for the answer. Yes, i know, tried to clean it up by hand adding new vertices and supporting edges, then target weld the miss-matching vertices, and it kind of worked but this is why i ask if there is any way to approach this correctly, or re-topologies the final result somehow. I tried welding the pillars together in ZBrush and dynamesh it to a higher polycount to keep the details and then subdivide it with zremesher, but it just became way to high poly for game asset. Any advise?
Hi, thanks for the answer. Yes, i know, tried to clean it up by hand adding new vertices and supporting edges, then target weld the miss-matching vertices, and it kind of worked but this is why i ask if there is any way to approach this correctly, or re-topologies the final result somehow. I tried welding the pillars together in ZBrush and dynamesh it to a higher polycount to keep the details and then subdivide it with zremesher, but it just became way to high poly for game asset. Any advise?
You would usually bake this detailed mesh down to a normalmap to be used on a low poly mesh
The only trick I have for models like that is use so much geo that the width of the bevel at the intersections is the same as the spacing of the edges around the pillars. It's never going to be perfectly clean, and you could delete some of the edge loops after modeling out the intersection. If I just needed a clean high poly and simple low poly I'd just use sculpting software. Only for subdiv or nanite would I worry about modeling it super cleanly, but even with nanite you can get aware with using sculpting geo.
I'm not a 3D modelling expert by any means, but I believe the key here is to model the whole thing with a not so dense mesh. I've spent a little time with it and started with a 12-sided cylinder. That way, you have better control over the intersecting parts. I recorded a time-lapse, so maybe it'll help you in some way. It's not perfect or finished at all—there are some triangles here and there, and overall it would need more time (which I don't have), but hopefully the point is clear:
That is amazing, thank you for the video, it gave me a great approach. I will get to it taking all your answers guys into consideration and create the low poly mash just as you did in the video and will bake the higher details. Thank you guys.
If something like that is available in Maya, you could just stick them together and use a rounded edge map / shader to get a small bevel transition between them and bake that to your lowpoly.
Slightly OT but I think worthwhile mentioning at least - if you've thoughts of expanding upon this? possibly further practice or when time allows of course, I might suggest also modeling in additional detail these colonnade shapes are renown for. Namely ornamental capitals usually carved as decoration, now although generated in Blender fair while ago there's no additional operations involved just manually editing it's cage which should be straightforward enough translating across too MAYA.
Anyway if you choose to do so, have fun with it....I did :thumbs up:
@sacboi Thank you so much, great video, it gave me a great idea. I was thinking of doing something like that, but i thought that i will model and sculpt the achantus leaf pattern onto a trim sheet that i was planning to make for my scene and will bake it into my mesh and use it as for my texture, because the pillars ornament is in 6m heigh and won’t be seen that close up, so i can get away with not having that much geometry on those details, so i can make one detailed leaf and make it tileable to wrap it around the ornament. Or is it a lazy approach in your opinion? 😅
No, efficient is how I'd describe your planned course of action. It's only that this thread is primarily a teaching/learning resource for those interested in sub division modeling, hence reasoning behind my post but all good sounds like you're on top of things and btw if interested, may I suggest opening a WIP in 3D Art sub-forum? results of which could also help others, when facing similar potential issues.
Thanks, good idea, i will do that. I only posted here as initially i was interested in how i can approach modelling the intersections of those pillars as i was struggling with that and you guys gave me great ideas and way to start on. Thank you for that.
I'm not sure why you have to subdiv this.. anyway i started with a simple grid; made all the tips; experiemented a bit and added a bevel; made this "arrayable"; then made my way through this to "quadrify it".. except two small triangle at the tips, which could be quaded too.. but this may not even be needed ?
I'm not sure why you have to subdiv this.. anyway i started with a simple grid; made all the tips; experiemented a bit and added a bevel; made this "arrayable"; then made my way through this to "quadrify it".. except two small triangle at the tips, which could be quaded too.. but this may not even be needed ?
Here the blend file and also the result as obj..
Thanks for the answer
I'm doing a high poly of a Colt 1911, you did it in a flat area, but in my case there is a curvature
Ohh.. this of course is some interesting context.. ( now i also understand the last image ) but then some more subdivisions along the bending axis and maybe some inset on the marked faces to break the smooth shading like so (totally not the curvation of the model just as an example):
Ohh.. this of course is some interesting context.. ( now i also understand the last image ) but then some more subdivisions along the bending axis and maybe some inset on the marked faces to break the smooth shading like so (totally not the curvation of the model just as an example):
Thanks again
Your application seems to be working very well
I tried to replicate it here, but the artifact still appears. I'm probably doing something very wrong.
Hello everyone, I would like to know how can I create this purple shape. Please if anyone can help me it would be great. I tried making it for sub div but couldn't get it right.
@jhonerick This type of pinching artifact is generally caused by excess smoothing stress generated where edges cross over the face of a curved surface to connect the corner vertex to the support loop. In this case the effect is exacerbated by the edges of the support loops themselves because they disrupt the segment spacing of the curve and are generating some unintended surface deformation before the smoothing is even applied.
While it is possible to brute force a solution by arbitrarily increasing the geometry density of the curve, it's also possible to simplify the mesh and solve each issue individually at a higher level. Keeping the shapes relatively simple and using the existing edges in the curve as part of the support loops for the intersecting shapes is generally considered best practice when working with subdivision and allowing the subdivision do a lot more of the smoothing work will help make things easier.
Below is an example of what this could look like: Start by blocking out the basic shapes until there's enough room between the segments of the curve to accommodate the intersecting shape, plus the width of the support loop for that same shape. Work on solving the basic topology flow paths around the shape and across the curve. Use at least one of the edges in the curve to act as an intermediate support loop across the flat interior surfaces of the intersecting shapes. Route the primary loop flow paths around the intersecting shape first, avoiding unintentional deformation of the curved surface, then add the final support loops to sharpen the edges of the shapes.
The basic topology flow can be routed directly around the shape intersection with manual loop placement
operations and the tighter support loops, used to define the sharp edges
of the shapes, can be generated with vertext group or edge weighted beveled
/ chamfer modifiers. Using the existing edges of the curve as support
and keeping the tighter edge loops within the existing segment spacing
will help reduce undesired pinching and surface deformation.
Here's what the final base mesh looks like before and after the edge sharpening support loops + subdivision are applied.
The same basic topology routing strategy works for most types of serrations. In this example there's no space between the edges of the serration and the resulting triangle on the curve is constrained by the adjacent loops and doesn't cause any noticeable smoothing artifacts so it's acceptable.
A lot of A1's and similar variants have fairly shallow serration patterns and when dealing with details like this it's generally best to keep things as simple as possible. Let the subdivision do the smoothing. Here the same segment spacing from the previous example works with the shallower serrations.
Subdivision smoothing is an approximate process: so there's a trade off between mesh density, editability, and shape accuracy. The topology routing in this example does produce some very minor surface imperfections, however they're only visible at extreme glancing angles and when viewed up close.
There's diminishing returns on the amount of time and effort spent
improving these sorts of results and if the high poly model is going to be baked down to a low
or viewed from first person then this is generally going to be more than
acceptable, given how small these details are and how subtle the minor
smoothing discontinuities are. Especially once normal and roughness details are added.
Close up comparison between reflective high gloss material and soft highlight material with smooth roll off. Subtle surface quality issues like this aren't visible under all conditions but could be resolved using the same topology routing strategies and increased segment density along the curved surface.
With subdivision: whether or not all that extra effort makes sense depends entirely on the use case, view distance, and material reflectivity.
Depending on what the project goals are: it may also be worth looking at alternate poly modeling and re-meshing workflows (some of which are native to Blender) or maybe even a parametric modeling workflow like Fusion or Plasticity.
The example below shows how those subtle surface quality issues are largely unnoticeable at first person view distances.
Also, tighter support loops around the edges might look great up close but the
edge highlights around the shapes will tend to disappear when viewed
from further away. Over sharpened high poly models can also cause baking
issues. Which is why it's generally better to have slightly softer edge
highlights so the shapes remain readable at a distance or when baking
down from a high poly to a low poly.
Recap:
-Block out the shapes and solve the larger topology flow issues first.
-Use the existing geometry of the curve to support the transitions between intersecting shapes.
-Avoid over sharpening the high poly as this can make the shapes difficult to read and might cause baking issues.
Links to additional write ups that cover similar shapes and smoothing artifacts:
@naman When modeling formed textile products it's often a lot easier to develop the basic shapes first then collapse the subdivision so there's enough starting geometry to place the seams, folds, darts, channels, tufts, etc. then use those edges as a starting point for modeling the rest of the details.
Here's an example of what this process could look like: Start by modeling the basic shape with subdivision then apply that modifier and connect the adjacent vertices to create new edges that match the channel pattern in the reference. Bevel / chamfer that set of edges to create the channel then shrink the inner edge loop down to give it some depth. Merge down the vertices in the top corners and adjust the rest of the shapes as necessary.
If the mesh needs to be all quads
then it's just some light clean up work to either merge down some of
the verts or add some edge loops. (For
items with more complex patterns it might be easier to use a shrink
wrapping or procedural / generative geometry modifier based approach to adding the details.)
Links to some other write-ups on subdivision modeling soft goods:
@jhonerick This type of pinching artifact is generally caused by excess smoothing stress generated where edges cross over the face of a curved surface to connect the corner vertex to the support loop. In this case the effect is exacerbated by the edges of the support loops themselves because they disrupt the segment spacing of the curve and are generating some unintended surface deformation before the smoothing is even applied.
While it is possible to brute force a solution by arbitrarily increasing the geometry density of the curve, it's also possible to simplify the mesh and solve each issue individually at a higher level. Keeping the shapes relatively simple and using the existing edges in the curve as part of the support loops for the intersecting shapes is generally considered best practice when working with subdivision and allowing the subdivision do a lot more of the smoothing work will help make things easier.
Below is an example of what this could look like: Start by blocking out the basic shapes until there's enough room between the segments of the curve to accommodate the intersecting shape, plus the width of the support loop for that same shape. Work on solving the basic topology flow paths around the shape and across the curve. Use at least one of the edges in the curve to act as an intermediate support loop across the flat interior surfaces of the intersecting shapes. Route the primary loop flow paths around the intersecting shape first, avoiding unintentional deformation of the curved surface, then add the final support loops to sharpen the edges of the shapes.
The basic topology flow can be routed directly around the shape intersection with manual loop placement
operations and the tighter support loops, used to define the sharp edges
of the shapes, can be generated with vertext group or edge weighted beveled
/ chamfer modifiers. Using the existing edges of the curve as support
and keeping the tighter edge loops within the existing segment spacing
will help reduce undesired pinching and surface deformation.
Here's what the final base mesh looks like before and after the edge sharpening support loops + subdivision are applied.
The same basic topology routing strategy works for most types of serrations. In this example there's no space between the edges of the serration and the resulting triangle on the curve is constrained by the adjacent loops and doesn't cause any noticeable smoothing artifacts so it's acceptable.
A lot of A1's and similar variants have fairly shallow serration patterns and when dealing with details like this it's generally best to keep things as simple as possible. Let the subdivision do the smoothing. Here the same segment spacing from the previous example works with the shallower serrations.
Subdivision smoothing is an approximate process: so there's a trade off between mesh density, editability, and shape accuracy. The topology routing in this example does produce some very minor surface imperfections, however they're only visible at extreme glancing angles and when viewed up close.
There's diminishing returns on the amount of time and effort spent
improving these sorts of results and if the high poly model is going to be baked down to a low
or viewed from first person then this is generally going to be more than
acceptable, given how small these details are and how subtle the minor
smoothing discontinuities are. Especially once normal and roughness details are added.
Close up comparison between reflective high gloss material and soft highlight material with smooth roll off. Subtle surface quality issues like this aren't visible under all conditions but could be resolved using the same topology routing strategies and increased segment density along the curved surface.
With subdivision: whether or not all that extra effort makes sense depends entirely on the use case, view distance, and material reflectivity.
Depending on what the project goals are: it may also be worth looking at alternate poly modeling and re-meshing workflows (some of which are native to Blender) or maybe even a parametric modeling workflow like Fusion or Plasticity.
The example below shows how those subtle surface quality issues are largely unnoticeable at first person view distances.
Also, tighter support loops around the edges might look great up close but the
edge highlights around the shapes will tend to disappear when viewed
from further away. Over sharpened high poly models can also cause baking
issues. Which is why it's generally better to have slightly softer edge
highlights so the shapes remain readable at a distance or when baking
down from a high poly to a low poly.
Recap:
-Block out the shapes and solve the larger topology flow issues first.
-Use the existing geometry of the curve to support the transitions between intersecting shapes.
-Avoid over sharpening the high poly as this can make the shapes difficult to read and might cause baking issues.
Links to additional write ups that cover similar shapes and smoothing artifacts:
Hello everyone, i am new here so i am not totally used to how this works. i have some questions and i hope you can help while modelling in 3DMAX. trying to model this sofa but i usually get lost in the process, unfortunately i don't have blue prints so only from these reference images.
these are the Refs:
and here is my progress, i mean i am there but not there at the same time. i donno how to finish it. also if you have any technical tips or even how to mentally think or approach this would be very helpful.
Hello, trying to get this to not widen up at middle. Any tips on correcting?
You'll either have to use ngons from the crease loops ending within the circle hole (and really play with the positioning of the verts), or use more geo.
Hello, trying to get this to not widen up at middle. Any tips on correcting?
To achieve that, the support loops need to go into the circular cutout, which would destroy the curvature of your cutout since it's not dense enough. For a clean subd result you'd need to use a cylinder with a lot more sides. Something like this:
Depending on how much pinching and inaccuracy of the circle shape you are willing to accept, you can go lower. Hope that helps!
@naman You're welcome. Glad to hear that the topology routing strategy worked.
@jhonerick Not a problem, thank you for the kudos.
@Ekladiuos How much detail needs to be fully modeled Vs added with textures or displacement and cloth sim depends on the art style and what the model will be used for. Other than that, it's generally considered best practice to block out all of the shapes by modeling each real life component as an individual mesh object, to establish both the basic proportions and topology flow, before jumping into adding subdivision. A couple of posts up there's a brief example of what a basic soft-goods modeling workflow could look like, along with some links to additional write-ups on that topic.
@Kidveggito Just to what ZacD and kosh3d have already covered:
If shape accuracy is the primary concern then increase the number of segments in the intersecting cylinder so there's enough edges there to carry the support loops across the intersecting shape.
If mesh simplicity is more important than shape accuracy then try terminating the support loops coming off the chamfer on the larger shape with a diamond quad that connects those support loops to the center edges of the intersecting cylinder. Like @ZacD mentioned, smoothing stress will deform the shape slightly but the vertices can be pulled out a bit to counteract the smoothing and sharpen the corners, while still allowing the intersecting shape to remain visually accurate.
Below is a quick example of what that could look like.
Here's a few links to write-ups that cover this type of shape intersection and some different topology routing strategies:
I’m working with a quad sphere and extruded a section inward on one of the faces. After that, I started trying to clean up the geometry for subdivision, but the more I tweak things, the worse the shading gets.
I’m running into artifacts, especially once I add subdivision — the surface just doesn’t look clean.
I’m trying to fix the topology so it holds up under subdivision and looks smooth without any weird shading issues.
Any tips on how to approach this kind of geometry cleanup would be much appreciated!
@SoldatenkoDmitry Most smoothing artifacts on curved surfaces can generally be traced back to adding or removing edges in a way that disrupts the segment spacing of the curve or unintentionally deforming the underlying curvature of the surface when editing the mesh.
This simplified approach to subdivision modeling, combined with blocking out the shapes and establishing the basic topology flow before adding lots of smaller support loops, keeps things a lot simpler and helps prevent smoothing artifacts caused by unintentional deformation, while also allowing any differences in the shapes to be taken up by or constrained to the space between the inner and outer support loops created by the edges in the intersecting shape and the existing geometry in the curved surface.
If the intersecting shape or cut out roughly follows the topology of the sphere then it's a fairly straightforward process of establishing the minimum amount of geometry required to outline the basic shape then insetting the surface to create the basic loop path around the shape intersection.
Afterwords the interior section of the cut out can be extruded inwards and flattened to create the required depth and a support loop can be added close to the surface of the shape intersection. Since subdivision smoothing tends to push outwards when loops are closer together along the edge of a curved surface, it's often necessary to push the defining edge of the cutout inwards a bit to compensate for the smoothing stress.
The interior edges of the cut out can be further sharpened by selecting the edges and running a bevel / chamfer operation to add some support loops along the inside of the intersecting shape. These support loops can be joined to the corner vertices on the surface of the sphere using a diamond quad. This allows the interior edge loops to terminate as all quads and constrains any smoothing artifacts to the space between the inner and outer support loops of the shape intersection.
It may also be necessary to manually push or pull some of the corner verts to compensate for the smoothing stress coming off the diamond quads.
Below is an example of what this modeling process could look like.
Subdivision modeling is an approximate process and a lot of it boils
down to making sensible tradeoffs between the accuracy of the smoothed
shapes and the editability of the mesh. Context is important: when
deciding how much geometry to use or whether or not a minor smoothing
artifact needs to be resolved, try to think about how the mesh will be
viewed and consider what's actually visible at that distance. That will
make it easier to figure out what's worth the effort and what's going to
contribute the most the overall visual quality of the asset.
If the sphere requires a higher degree of shape accuracy or surface quality then the same approach can be used on a denser mesh. The trade off here is that a denser mesh is typically harder to edit and it's still necessary to slide some of the existing edges into the corners to sharpen them, which could lead to some minor deformation issues or smoothing artifacts where the segment spacing is disrupted.
How sharp the edges
need to be really depends on the visual style and what the model will be
used for. The example directly above and below are more than passable for most
assets that will be viewed at this size on screen.
This strategy of using the existing edges as support loops can be applied to more precise shape intersections.
The example below shows a boolean subtraction sliced out of the shape then cleaned up using a join through operation and support loops generated by a bevel / chamfer operation. As a final step to cleaning up the shape intersection, the upper most loop between the intersecting shape and curved surface of the sphere is dissolved to minimize the disruption of the existing edge segments that make up the sphere.
Since the intersecting shape doesn't align perfectly with the edges in the sphere there's a few spots where the edge highlight along the inside of the shape intersection varies in width but this is generally more desirable than potential pinching artifacts generated by too many support loops bunching up on the outer curved surface of the sphere.
Sometimes a sharper edge highlight and more consistent surface is required but it's unreasonable to use additional geometry. In these cases it can make sense to shrink wrap the outer segment of the unsubdivided edge loop onto a quad sphere of similar density to the final subdivision surface.
Whether the shrink wrap is before or after the subdivision really depends on which order produces the best results.
The same approach of using the existing segments also works on a sphere with a denser mesh but in this example the spacing between the existing edges is close enough that it's possible to retain the outer support loop around the shape intersection, without disrupting the smoothing of the curve. This makes for a sharper edge highlight and more accurate shapes but the mesh is a lot harder to edit at this density.
Which modeling approach and mesh density makes the most sense depends
entirely on the desired sharpness, surface quality, and shape accuracy.
Efficient subdivision modeling is all about solving topology issues at
the highest level possible and making trade offs between accuracy
and editability that best fit the use case for the model.
Recap:
Block out the basic shapes and solve the basic topology flow issues before adding a lot of support loops that only sharpen the edges of shapes and shape intersections. Try to use the existing edges of the curved surfaces as support loops and avoid deforming or disrupting the spacing of curved surfaces whenever possible. Use the minimum amount of geometry required to hold the shapes while still meeting the desired surface quality level for a given view distance and use case.
Some links to other write-ups that cover similar dome and sphere shapes:
Hey, I'm having a hard time making this four sided nut. I feel like I'm almost there but I can't quite get the corners to match the reference. I'm still very new to SubD stuff.
Hey, I'm having a hard time making this four sided nut. I feel like I'm almost there but I can't quite get the corners to match the reference. I'm still very new to SubD stuff.
Your corner curvature does not quite match the reference, you can see that the sides of the nut have slightly curved edges which give the corners their rounded shape. Here is how you could approach this:
1. Make the side profile and the top
2. Duplicate the side profile around the nut
3. Since everything is symmetrical we only need to work on 1/4 of the nut. Connect the top and side profiles together by bridging the edges
4. Place support loops
5. Use symmetry to complete the object
Here is how it looks when subdivided:
Depending on how realistic and accurate it needs to be, you can add the threading by using a helix (which might not be needed since it won't be visible if the nut is on a screw or something)
Also, since you wrote that you're new to subd, I highly recommend scrolling through this thread. It is a goldmine of information, especially posts by FrankPolygon are excellent and you can learn a ton here. Hope this helps!
Hey, I'm having a hard time making this four sided nut. I feel like I'm almost there but I can't quite get the corners to match the reference. I'm still very new to SubD stuff.
Your corner curvature does not quite match the reference, you can see that the sides of the nut have slightly curved edges which give the corners their rounded shape. Here is how you could approach this:
1. Make the side profile and the top
2. Duplicate the side profile around the nut
3. Since everything is symmetrical we only need to work on 1/4 of the nut. Connect the top and side profiles together by bridging the edges
4. Place support loops
5. Use symmetry to complete the object
Here is how it looks when subdivided:
Depending on how realistic and accurate it needs to be, you can add the threading by using a helix (which might not be needed since it won't be visible if the nut is on a screw or something)
Also, since you wrote that you're new to subd, I highly recommend scrolling through this thread. It is a goldmine of information, especially posts by FrankPolygon are excellent and you can learn a ton here. Hope this helps!
Appreciate the help. Even with the guide you gave this is as far as I was able to get. Still not 100% perfect but it is what it is. I'll be sure to check out Franks stuff as well.
The radial turning features could be generated from a simple edge profile that's spun into a solid mesh using a screw modifier and boolean subtracted from the basic cube. This would make it a bit easier to visualize the profile and make adjustments to the shapes without directly editing the mesh.
Once the shapes are correct it's a fairly straightforward process of cutting in the edge loops across the shape and using an inset operation to add support loops around the inside of any flat surfaces. The top shoulder profile on the corner can be softened by sliding that edge down slightly.
Below is an example of what this process could look like.
Additional details like internal threading can also be added using similar spin and boolean operations.
Recap: Keep things relatively simple and focus on blocking out all of the primary forms before adding support loops. Accuracy is more or less locked in by the shapes in the base mesh so time spent improving the accuracy of the block out is usually paid back during the rest of the modeling process.
The radial turning features could be generated from a simple edge profile that's spun into a solid mesh using a screw modifier and boolean subtracted from the basic cube. This would make it a bit easier to visualize the profile and make adjustments to the shapes without directly editing the mesh.
Once the shapes are correct it's a fairly straightforward process of cutting in the edge loops across the shape and using an inset operation to add support loops around the inside of any flat surfaces. The top shoulder profile on the corner can be softened by sliding that edge down slightly.
Below is an example of what this process could look like.
Additional details like internal threading can also be added using similar spin and boolean operations.
Recap: Keep things relatively simple and focus on blocking out all of the primary forms before adding support loops. Accuracy is more or less locked in by the shapes in the base mesh so time spent improving the accuracy of the block out is usually paid back during the rest of the modeling process.
Hey, I appreciate the help and yeah it seems like a really good non destructive way to make screws and stuff. That's all I'm looking for right now, just basic, non destructive ways to make interesting shapes. I got a hard time making anything even remotely curved or rounded so small objects like these are pretty challenging but guess I gotta keep trying and see what happens. Will check out the intro thread when I can. Cheers.
Having trouble getting the indentures on this shape accurate as well as having it turn back into a normal cylinder without pinching. Any help on getting this shape?
I take it you are trying to isolate and analyze this first, which isn't a bad idea, but since this is such a "simple" form, you might just as well begin with the real thing right away and nail the shapes and proportions from the start. If you have a hard time seeing what's going on from your reference, the advanced search filters of Google allow you to set a minimum resolution, so here are a couple more images:
They are not all of the same model, and in some, the shapes seem more soft, which might be the result of camera settings/properties, lighting (reflections), upscaling, or worn down molds later in the production run (as opposed to product images taken right at its start) and the white version is a different product line and might be a 3D render (but shows the shape well). So you'll have to pick and choose regarding some aspects, but I hope it helps anyway.
Basically, you can imagine the valleys as being subtracted from the main shape (and maybe even use a Boolean as a starting reference like e.g. kosh3d has shown in one of his threads). You could also model this flat and then bend it into a ring later or go with radial symmetry. You might have done that already, but especially without a cage/basemesh it's hard to tell what you tried so far and what would need fixing in your model specifically.
Here's a suggestion with radial symmetry, I chose to make the most protruding edges a bit more rounded, like after some use (granted, I winged it a bit, too, and looking at the references again, I don't think the subtracted shape tapers that much, or, perhaps, at all).
I am sure similar questions have been asked before, but i cant happen to find a fitting post...
So: I am currently trying to model a rifle and almost made it to the baking step. I only have two issues left where the model doesnt properly subdivide for the highpoly.
How would I change the topology, so the normals flow as expected? Neither Bevels, Creases, additional loops (usually breaking topo somewhere else) or anything seems to do the trick, so the only thing i can think of is changing loopflow.
I am thankfull for any recommendations about how it should be done, so that i can go and adjust from there. Much love.
@gojushin The very short answer is: the curved areas in the base mesh need enough geometry to carry the edges of any intersecting shapes and their support loops.
If the overall shape of the existing stock model is acceptable
then one of the easier solutions would be to apply 1-2x subdivision, cut out any slots or other surface features, clean up the topology
around the shape intersections, and add another subdivision modifier to
finish smoothing the final high poly surface.
(How much geometry is required really depends on the shapes and how sharp they need to be. A few posts up there's a couple of conversations about routing support loops for shape intersections on loops on curved surfaces and spheres. Each of those has a few additional links back to other posts that cover similar topology routing solutions.)
Keeping things relatively simple and solving the topology flow issues around the shapes as part of the block out process will tend to make things a lot easier. It's also generally considered best practice to block out the
larger forms first, adding mesh density when and where it's needed
to support smaller details, so the existing
edges in the curves can be matched to or used as support loops around
the intersecting shapes. Below is an example of what this process could look like.
Start by developing the larger forms first. In this example only basic subdivision and manual modeling operations are used to create the basic shapes, though more complex non-destructive modifier based approaches could also be used.
Once the basic shapes are accurately modeled, move on to adding any intersecting shapes or cut outs and adjust the density of the base mesh so the existing edges can be used as support loops around the shape intersections.
Join in or cut out any intersecting shapes and clean up the mesh around the intersections to solve any major topology flow issues. Use the existing edges as support loops by landing the intersecting shapes between the segments of the curves and try to avoid causing any unintended mesh deformation when connecting edges and support loops to the existing geometry on the curved surfaces.
In the example below there's a few spots where some extraneous geometry is left in during the modeling process. Mainly to prevent other modeling operations from deforming the surface. This can be cleaned up, after the support loops are placed around the shapes, by merging down or dissolving any stray edges that would otherwise disrupt the segment spacing along the existing curves.
With these kinds of shape intersections on compound, complex curves it's important to: avoid unintentionally deforming the geometry in the existing curved surface, avoid bunching up support loops in a way that disrupts the segment spacing of the underlying curves, use the space between the support loops to make up for any differences between the intersecting shapes and the underlying curves, and try to keep the corner vertices / poles as close to the support loops as possible.
Here's what the final base mesh looks like, without and with subdivision smoothing enabled.
Though there are some exceptions that do have sharper corners, most inletted load bearing features on stocks tend to have softer or gradual shape transitions with radiused / fillited / rounded corners to prevent stress concentration. These types of softer corners are usually easier to model since they can typically be created using less geometry than sharper angled corners.
The same basic block out and topology routing strategies shown previously can also be used for modeling sharp or square corners on curved surfaces. It's just the base mesh may require more geometry to accurately hold both the curvature and sharper edges around the intersecting shapes.
Links to additional write-ups on using an iterative block out process:
I take it you are trying to isolate and analyze this first, which isn't a bad idea, but since this is such a "simple" form, you might just as well begin with the real thing right away and nail the shapes and proportions from the start. If you have a hard time seeing what's going on from your reference, the advanced search filters of Google allow you to set a minimum resolution, so here are a couple more images:
They are not all of the same model, and in some, the shapes seem more soft, which might be the result of camera settings/properties, lighting (reflections), upscaling, or worn down molds later in the production run (as opposed to product images taken right at its start) and the white version is a different product line and might be a 3D render (but shows the shape well). So you'll have to pick and choose regarding some aspects, but I hope it helps anyway.
Basically, you can imagine the valleys as being subtracted from the main shape (and maybe even use a Boolean as a starting reference like e.g. kosh3d has shown in one of his threads). You could also model this flat and then bend it into a ring later or go with radial symmetry. You might have done that already, but especially without a cage/basemesh it's hard to tell what you tried so far and what would need fixing in your model specifically.
Here's a suggestion with radial symmetry, I chose to make the most protruding edges a bit more rounded, like after some use (granted, I winged it a bit, too, and looking at the references again, I don't think the subtracted shape tapers that much, or, perhaps, at all).
Modeling flat then doing a radial array worked perfectly, thanks for the demonstration. I didn't think to do it.
Replies
Is there a short cut to convert the stepy-edge's to a circular shape in edit poly mod ?
i started with a cylinder and then i focused on a quarter of it. modeled the first quarter and then using symmetric modifier i got it to this level.
since i have to continue the inside part and its flat, i have to convert the last edge to this round shape. is there a short-cut to make it soft ?
i think im also able to use "shapes" to make the insite part. (yes) but couldn't be better to model this all in one part ?
also let me ask, is there an easier approach to the out part ?
There’s also a tool in the Graphite ribbon, to make the edges evenly-spaced, maybe under Loop tools?
is there an easier approach ? like making using world path modifier ?
I had to cheat a bit with an Edit Poly, to get a top bevel, separate the drain area, and fix a bad vertex. I hate using Edit Poly because it breaks parametric editing... Edit Poly is vertex-count-dependent so if you edit stuff below it you're more likely to break things.
Anyhow, you can go through the modifiers one-by-one to see what I did. Just turn off "Show End Result" to see the steps.
i got it.
i used an eclipse, duplicating and scaling it to match the final shape, then i created a surface, then i customized only a part and using symmetry it just fit really nice together. love this section. very informative post. well done everyone.
I'm new to the whole world, i just started studying it, but i was wondering if you could give me some advise.
I was trying to model this pillar complex in maya for a game environmental asset, based on the reference. (pic 1)
I modelled one thicker, and one thinner pillar with the decorative top part as extrusions, and i lined them up as they should be, then booleaned them to make a union. As you can see on the picture it is a pain to clean up the topology at the joints of the decorative top part. (pic 2)
I tried a different approach, which has a nicer topo, but it just doesn't look as nice as the original. (pic 3)
Is there any way to model this correctly? Or should i just import it into ZBrush and Dynamesh them to weld them together instead of doing a boolean and than re-topologies it after and decimate it back?
Thank you for your answers.
It's the same with this 32 and 16 vertices cylinders with a 1.5 bigger top:
The side edges just cross over in different angles all over the place.. so this kind of topology is a really nasty..
Thank you so much, great video, it gave me a great idea. I was thinking of doing something like that, but i thought that i will model and sculpt the achantus leaf pattern onto a trim sheet that i was planning to make for my scene and will bake it into my mesh and use it as for my texture, because the pillars ornament is in 6m heigh and won’t be seen that close up, so i can get away with not having that much geometry on those details, so i can make one detailed leaf and make it tileable to wrap it around the ornament. Or is it a lazy approach in your opinion? 😅
Here the blend file and also the result as obj..
I would like to know how can I create this purple shape. Please if anyone can help me it would be great. I tried making it for sub div but couldn't get it right.
Please help me.
The example below shows how those subtle surface quality issues are largely unnoticeable at first person view distances.
https://polycount.com/discussion/comment/2771257/#Comment_2771257
https://polycount.com/discussion/comment/2742955/#Comment_2742955
@FrankPolygon
trying to model this sofa but i usually get lost in the process, unfortunately i don't have blue prints so only from these reference images.
these are the Refs:
and here is my progress, i mean i am there but not there at the same time. i donno how to finish it. also if you have any technical tips or even how to mentally think or approach this would be very helpful.
thanks a lot!
Hello, trying to get this to not widen up at middle. Any tips on correcting?
You'll either have to use ngons from the crease loops ending within the circle hole (and really play with the positioning of the verts), or use more geo.
@jhonerick Not a problem, thank you for the kudos.
https://polycount.com/discussion/comment/2770261/#Comment_2770261
https://polycount.com/discussion/comment/2775032/#Comment_2775032
I’m running into artifacts, especially once I add subdivision — the surface just doesn’t look clean.
I’m trying to fix the topology so it holds up under subdivision and looks smooth without any weird shading issues.
Any tips on how to approach this kind of geometry cleanup would be much appreciated!
This strategy of using the existing edges as support loops can be applied to more precise shape intersections.
https://polycount.com/discussion/comment/2733818/#Comment_2733818
Briefly adding to what @kosh3d said about modeling: an alternate approach would be to block out the primary forms using solid shapes.
The radial turning features could be generated from a simple edge profile that's spun into a solid mesh using a screw modifier and boolean subtracted from the basic cube. This would make it a bit easier to visualize the profile and make adjustments to the shapes without directly editing the mesh.
Below is an example of what this process could look like.
Recap: Keep things relatively simple and focus on blocking out all of the primary forms before adding support loops. Accuracy is more or less locked in by the shapes in the base mesh so time spent improving the accuracy of the block out is usually paid back during the rest of the modeling process.
If you have a hard time seeing what's going on from your reference, the advanced search filters of Google allow you to set a minimum resolution, so here are a couple more images:
They are not all of the same model, and in some, the shapes seem more soft, which might be the result of camera settings/properties, lighting (reflections), upscaling, or worn down molds later in the production run (as opposed to product images taken right at its start) and the white version is a different product line and might be a 3D render (but shows the shape well). So you'll have to pick and choose regarding some aspects, but I hope it helps anyway.
Basically, you can imagine the valleys as being subtracted from the main shape (and maybe even use a Boolean as a starting reference like e.g. kosh3d has shown in one of his threads).
You could also model this flat and then bend it into a ring later or go with radial symmetry. You might have done that already, but especially without a cage/basemesh it's hard to tell what you tried so far and what would need fixing in your model specifically.
Here's a suggestion with radial symmetry, I chose to make the most protruding edges a bit more rounded, like after some use (granted, I winged it a bit, too, and looking at the references again, I don't think the subtracted shape tapers that much, or, perhaps, at all).
So: I am currently trying to model a rifle and almost made it to the baking step. I only have two issues left where the model doesnt properly subdivide for the highpoly.
How would I change the topology, so the normals flow as expected? Neither Bevels, Creases, additional loops (usually breaking topo somewhere else) or anything seems to do the trick, so the only thing i can think of is changing loopflow.
I am thankfull for any recommendations about how it should be done, so that i can go and adjust from there.
Much love.
If the overall shape of the existing stock model is acceptable then one of the easier solutions would be to apply 1-2x subdivision, cut out any slots or other surface features, clean up the topology around the shape intersections, and add another subdivision modifier to finish smoothing the final high poly surface.
(How much geometry is required really depends on the shapes and how sharp they need to be. A few posts up there's a couple of conversations about routing support loops for shape intersections on loops on curved surfaces and spheres. Each of those has a few additional links back to other posts that cover similar topology routing solutions.)
Below is an example of what this process could look like.
Though there are some exceptions that do have sharper corners, most inletted load bearing features on stocks tend to have softer or gradual shape transitions with radiused / fillited / rounded corners to prevent stress concentration. These types of softer corners are usually easier to model since they can typically be created using less geometry than sharper angled corners.
Links to additional write-ups on using an iterative block out process:
https://polycount.com/discussion/comment/2751240/#Comment_2751240
https://polycount.com/discussion/comment/2751340/#Comment_2751340