Making an Octopus Bracelet - 3D Print
polycounter lvl 10
It's not game art specifically, but it's relevant and I thought this would be of interest to some people here.
As a Christmas present, I wanted to make a bracelet to 3d print based on one I saw in a shop. I chose to this through Shapeways.com and their ever-expanding metal printing options. The original was silver, but with price and shine being what they are, I thought steel would do nicely. Also, this entire project doubled as an exercise for me to improve with zBrush.
I wanted to keep the general design as it, but there were some design changes, to come later. So here is the reference I gathered both online and in person:
And a sloppy blockout in Maya. Just trying to place the tentacles and be super sure that there are actually eight of them.
Once it resembled something I was trying to make, I dropped it into zBrush, hit Dynamesh, and smoothed it out.
Initial sculpting, still mostly concerned with proportion and volume.
After getting comfortable with my shapes and remeshing it, I started in with tests for the suckers. Here's where I got after a couple attemps. I did this with an alpha and just drag-placing them around. Camera work proved to be the hardest part with the tentacles.
Got tired of that boring matcap and wanted to see curvature better, as a lot of time was put into simply smoothing it right.
Finished adding the suckers and did a touch of head sculpting. I ended up exaggerating the proportions of the eyes because, well, I liked them. It was also at this point that I hollowed the model and sent it off to be printed in plastic as test for size and fit. More on that in a bit, as I continued the sculpt while waiting.
And here is where things started getting tedious. I wanted to cover a lot of the surface with the organic dotted pattern seen in the original. I tried a few different approaches to this, using an alpha, brush settings, procedural noise, etc, but nothing really seemed to beat the look I was getting by doing it manually. I decided to suck it up and do the whole surface manually because I felt it would yield the best results. I would take a section and mask a bunch of dots at varying scales and then sculpt down and smooth the areas in between. Here I was not concerned with the dots matching the placement of the original.
Hooray! I finally finished covering (mostly) all of it with stupid little dots and the sculpt was coming together. I would eventually do a pass to emphasize and smooth out areas of dots to give it a bit more of a handmade look. But not long after this point, my plastic test print arrived!
Despite the fact that I was rather pleased with how it looked, this proved to be far too small to slip over my own or nearly anyone else's hands. I checked and rechecked my proportions from my reference and scale from other bracelet models on Shapeways. My conclusion was that it need to be considerably larger though I could make the hole a bit more squat. I didn't think I had the time nor did I want to drop the cash to do a second test print so I wanted to approach this as carefully as possible so the final result would be as it should.
So I ended up doing some measuring and cutting to make some cardboard guides to see exactly what it'd take to get this bracelet to slip over my, or a slightly smaller, hand. Also there's a sheet of math that was involved in some way or another.
My conclusion was that the bracelet would need to be scaled up so much that it would fit too loosely about the wrist. Steel is a bit heavier so having this fit very loosely would probably be a nuisance. Also, scaling it up raises the price tag pretty quickly so that was less than ideal. I did handle the original in person and I feel that it was fairly close in size to my original and there was no hinge to be seen so I can only gather that it was meant specifically for anyone with exceedingly small hands.
So! I needed a more dramatic solution to this issue. After some debate, I decided that my best option would be breaking from the original design and changing it to a cuff bracelet that could be put directly on the wrist without having to go over the hand at all. After a repeated process of trimming and smoothing the ends of the test print, I ended up with this. And whaddya know, it fits on my wrist!
As a Christmas present, I wanted to make a bracelet to 3d print based on one I saw in a shop. I chose to this through Shapeways.com and their ever-expanding metal printing options. The original was silver, but with price and shine being what they are, I thought steel would do nicely. Also, this entire project doubled as an exercise for me to improve with zBrush.
I wanted to keep the general design as it, but there were some design changes, to come later. So here is the reference I gathered both online and in person:
And a sloppy blockout in Maya. Just trying to place the tentacles and be super sure that there are actually eight of them.
Once it resembled something I was trying to make, I dropped it into zBrush, hit Dynamesh, and smoothed it out.
Initial sculpting, still mostly concerned with proportion and volume.
After getting comfortable with my shapes and remeshing it, I started in with tests for the suckers. Here's where I got after a couple attemps. I did this with an alpha and just drag-placing them around. Camera work proved to be the hardest part with the tentacles.
Got tired of that boring matcap and wanted to see curvature better, as a lot of time was put into simply smoothing it right.
Finished adding the suckers and did a touch of head sculpting. I ended up exaggerating the proportions of the eyes because, well, I liked them. It was also at this point that I hollowed the model and sent it off to be printed in plastic as test for size and fit. More on that in a bit, as I continued the sculpt while waiting.
And here is where things started getting tedious. I wanted to cover a lot of the surface with the organic dotted pattern seen in the original. I tried a few different approaches to this, using an alpha, brush settings, procedural noise, etc, but nothing really seemed to beat the look I was getting by doing it manually. I decided to suck it up and do the whole surface manually because I felt it would yield the best results. I would take a section and mask a bunch of dots at varying scales and then sculpt down and smooth the areas in between. Here I was not concerned with the dots matching the placement of the original.
Hooray! I finally finished covering (mostly) all of it with stupid little dots and the sculpt was coming together. I would eventually do a pass to emphasize and smooth out areas of dots to give it a bit more of a handmade look. But not long after this point, my plastic test print arrived!
Despite the fact that I was rather pleased with how it looked, this proved to be far too small to slip over my own or nearly anyone else's hands. I checked and rechecked my proportions from my reference and scale from other bracelet models on Shapeways. My conclusion was that it need to be considerably larger though I could make the hole a bit more squat. I didn't think I had the time nor did I want to drop the cash to do a second test print so I wanted to approach this as carefully as possible so the final result would be as it should.
So I ended up doing some measuring and cutting to make some cardboard guides to see exactly what it'd take to get this bracelet to slip over my, or a slightly smaller, hand. Also there's a sheet of math that was involved in some way or another.
My conclusion was that the bracelet would need to be scaled up so much that it would fit too loosely about the wrist. Steel is a bit heavier so having this fit very loosely would probably be a nuisance. Also, scaling it up raises the price tag pretty quickly so that was less than ideal. I did handle the original in person and I feel that it was fairly close in size to my original and there was no hinge to be seen so I can only gather that it was meant specifically for anyone with exceedingly small hands.
So! I needed a more dramatic solution to this issue. After some debate, I decided that my best option would be breaking from the original design and changing it to a cuff bracelet that could be put directly on the wrist without having to go over the hand at all. After a repeated process of trimming and smoothing the ends of the test print, I ended up with this. And whaddya know, it fits on my wrist!
Replies
Next, I'd focus on hollowing the model by using a dynamesh shell and then decimating it to meet the Shapeways upload limit. The site requires meshes be under 1 million tris and, with the surface being what it is, I felt like this was the most limiting factor in detail, rather than sculpt detail or print quality. Though after much tinkering in zBrush, I got it decimated to 1 million without a noticeable drop in quality. Next, I uploaded to the site and that took many attempts as the upload would seemingly timeout regardless of my computer or connection (tried a few different places). Nevertheless, it eventually worked, I ordered the final print and waited.
Just about a month later, this is what I got:
And it fits!
It's a bit looser than I'd like it to be, but I don't think it's in any real danger of falling off. It's also on the heavy side, being steel, but not quite as heavy as I was anticipating and after wearing it for a bit, it isn't noticeable.
I know it seems quick with all the images here but this project was weeks worth of works spread over a few months and was more difficult than I had anticipated. Much of the difficulty was in improving my zBrush skills enough to meet the task and I feel like I have accomplished that and I'm fairly proud of the final result. 3D printing is awesome.
I'm currently dealing with Shapeways too, I find this 1 Millions polygons limit ridiculous, I never heard of this before. I feel like they do this to make the customer pay more by separating the models into pieces.
Could you tell me how you did the hollowing ? Does it change the already sculpted parts ?
But, how did you manage to make the initial measurements? Thats one thing thats still throwing me.
Raw steel print:
After manual smoothing and polish:
The finishing options on Shapeways have improved so much that I didn't have to do any manual work on the final bracelet.
supremedalek925: Thanks, it's a weird but awesome feeling
Texelion: Yeah, apparently the limit used to be 500k and they raised it; hopefully they'll do that again. The 1 mil limit really just meant I had to do some crafty decimation, but I could see it being an issue with like a full character sculpt. To hollow it out I used Dynamesh's Shell function seen [ame="
Azure: Thanks, I used a pre-existing bracelet model as a base and then measured my own wrist and then estimated the wrist measurements of my girlfriend, as it's for her. Generally I felt like I was winging it but it took a lot of effort to get accurate. And it fits!