First iteration 3D Model Printing
So I wasn’t planning on collaborating with game design students with their board game project as I was busy with my own projects and didn’t really see the appeal in board games as the teams were looking for 2D artists, and I’m not much of a 2D artist. I was planning on just waiting for their next project and seeing if it had some 3D implementation for me to work on, that being said I was also pretty busy with some team projects already so it wasn’t a priority at all until I had the spare time.
Well anyway I was introduced to a couple of the game design teams whilst waiting for our KPI’s and I asked around for any 3D work which could be done. I ended up joining in a team with a friend and got some 3D modelling to do for 3D printing, something which I haven’t done before, but figured the gig sounded fun enough so what’s the harm. So basically The teams project is a gladiator style board game with 2 opposing factions (the gladiators) and a king which can influence the outcome of the game. So for the 3D models the team requires 2 gladiator models with an extremely exaggerated style, but they also need to resemble gladiators during a gladiatorial match.
So I quickly threw together some reference as Clash of Clans has always interested me with their cute exaggerated characters (their animated shorts are really fun take a look if you’re interested).
So yeah, I created a quick Art bible as the game design team allowed us to have a look through their GDD, but their art style and their visuals were primarily of weapons, not really any direction that we could use. So they ended up taking this Art Bible and putting it into their GDD (Below).
So after finally receiving some feedback on the ArtBible we finally had some vague direction, something we’d been seeking for a couple of days prior, but I don’t think the game design students use Slack as much as us animation students do. Something which hadn’t been done yet was to check out the 3D printing pipeline. So I went to go find out some details about the pipeline, and turns out We have until Friday, It is currently Wednesday. We literally have 2 days, almost 48 hours on the dot to create the 3D model base, the defining amour and extras, rig and pose it, place on the appropriate base and then finally garentee that it can be 3D printed. That’s a lot to do in such a short time frame.
So below outlines my production steps creating the base character mesh, detailing that mesh, rigging it, posing it, adjusting it, and then finalising it for the team.
Reference image which I used in the 3dsMax scene to use for proportions as we didn’t have time to create a turnaround character sheet, I haven’t got the designer name right now, but i’ll search for it as I found this picture on Pinterest months ago but I really enjoy the exaggerated character and cute art style used, I’d hoped that this would fit the theme they wanted and the team was pleased with the reference.
So basically I took an older character base mesh and completely re-topologised it according to some solid reference which I found on Pinterest. Each of these screenshots are those which I used to update the team on my progress throughout production.
So this is the finalised base mesh, as I have another animation student with me I elected to create the base model as my modelling is a bit stronger so it required less time. So i sent this file off to him so he could start modelling his armour pieces so suit the style and also to blend with the mesh.
I used the base mesh to create armour and weapon sets that could tied with the character and the theme. I started with the helm as a had plenty of reference and I had made it very apparent this was the themed helm I wanted to go with and also the large shield as these came up very often in the retellings of gladiatorial battles. I also added a lot of exaggeration to really push the style.
So by this point I was basically done with the model, I needed to rig it and pose it before the end of the night and with such a strict deadline I couldn’t keep adding armour and extra piece as it would just make more of a mess later on with the rigging and posing.
I had rigged the character by manipulating a pre-existing 3dsMax human cat rig and then Voxel skinning it, I almost did no skinning at the topology and rig were done well enough that the Voxel solve did a really great job and saved me a huge amount of time. So I gave the team a choice of the top 3 poses to choose from as I didn’t believe we had time to mess around trying to get them a perfect pose, that being said Tyler and I decided to create one last pose to suit his character. We did this because we decided that I could pose the character and then convert to an editable Poly, from there i was to send him the file and he could merge and replace areas with his own tailor made armour pieces, which in the end worked really well and saved a huge amount of time.
Once we finally had our characters in the correct pose and Tyler had modified his character model to accompany mine we had to let the game designers decide upon whether to add a turbo smoothing modifier or to leave the low poly as a design choice. In my honest opinion I really liked the low poly character, but regardless the designers voted upon the turbo smoothed models. So I took Tyler and my models and added a turbo smooth modifier and then re-topologised them both so they could look as good as possible as a turbo smooth mod doubles the poly count but modifies geometry by smoothing it out, so some pieces were not looking so great.
So finally we came to our concluded model, We had 48 hours and I used 22 of those to model, rig, pose, re-topologise, and finalise 2 models alongside Tyler. Unfortunately last minute we found some sizing issues which we had completely blanked on, we fixed them and exported the files, but unfortunately the issues remained with the export, luckily issue was picked up and fixed immediately by Ralf, our 3D printing wizard.
I think there are things I would have liked to have fixed up, but all-in-all we were incredibly rushed and communication because scares at early points in the project. So i think in the end it came out all well and good. That being said, I’ll wait for confirmation of 3D printing before giving a solid conclusion.
First iteration post mortem:
I’m still waiting on the pictures of the models so I can show evidence of issues and whatnot and also so I can avoid these issues in the next iteration. We gave out 3D models to the game design team immediately and their showcase was the next day so i didn’t have time to take some photos.
First Iteration 3D printed model issues to be resolved:
In application sizing issues:
For whatever reason the models where in Inches not CM and then after fixing up the unit measurements and re-sizing the models within the program. Unfortunately the sizing issues was rolled into the exported file, luckily it is a really common issue when max exports into the 3d printing file type, so Ralf was able to fix it up easily, I also emailed him stating the issue and gave him measurements just in case the issue did in fact roll into the export. Not sure what caused it but It’s something to keep an eye on for next time, luckily I made him aware of the sizing issue with an email so he didn’t print out huge models that were unable to be used.
Turbosmooth Modifier issues for 3D printing:
We didn’t pay enough attention to the width of our weapons, so the swords for both models were damaged, my model started to crumble around the blade tip and edge, whilst Tyler’s models sword broke off entirely. The shield for the Red model caused an issue as well, the shield width wasn’t great enough so the 3D print out had gaping holes in it, it later fell off the base as it wasn’t secure enough.
Models too fragile:
The models started to fall apart during the play testing. Tyler and I were the 4th or 5th group to playtest and by the time I sat down to play I had noticed the head was caving in, the leg had snapped at the knee, and the shield was peeling off the base. The Red model had both the sword and shield break, but the pose seemed solid. So perhaps we can try a slightly finer print next iteration as well as securing the weapons and giving a solid width to both weaponry and also the knees.
The arm guard scales for both models had to be removed for 3D printing as they were too small and didn’t intersect properly due to the turbosmooth modifier. We also lost some defining armor detail for Red model as the spikes were too fine to print properly. Also the base was a little too plain, maybe needs some curvature or some extra loving.
But considering the limited time we had, and that we have never modeled for 3D printing before It was a good start point, we made plenty of mistakes which means I/we have a heap of things that we can learn from and therefore improve upon.
Second iteration model printing.
So for second iteration models I was heavily considering creating the model in sections and printing it out on attached scaffolding, but after multiple opinions saying it probably isn’t needed I decided against it and just pushed to clean up the models the way they are.
So above you can see that I have cleaned up the body further, and strengthened the joints and pushed the muscles in both the arms and legs. I also believe I have fixed the issue with the head con-caving into itself, I hadn’t closed off the back of the helm properly so there was a large gap between the head proxy and the helm itself as the inner helm wasn’t shelled, only one sided faces. I also cleaned up the base as it wasn’t matching the turbosmoothed model, it looked too clunky and thin, so I’ve gone ahead and polished it up a bit. The last couple of touch ups I believe I should do is to fix both weapons as the shield isn’t quite looking consistent with the smoothness of the body, it looks too low poly. Also I need to strengthen the ankles and thicken the feet, they are looking way too skinny. Finally I need to re-create the sword and have it match the style whilst also retaining a solid width to survive 3D printing.
Sadly it seems that the 3D printer is our of commission so I won’t be able to print out the newest iteration, it’s unfortunate, but that’s just how it goes, at the end of the day i’ve gained a heap of new information regarding printing for 3D and also been practicing my edge loops for higher poly modelling to get a stylized effect which I had been meaning to test out, so 2 birds, 1 stone.
After seeing the Dark Souls board game 3D printed models I’m as motivated as I’ll get to play around with more 3D models for next trimester, so I’ll definitely continue playing with 3D models for printing, maybe some slightly larger model next time around. Perhaps I’ll take this character from an earlier iteration and re-rig him for animation, I feel like he’d be a great test model so I can strengthen my rigging and animation in the break.