Let's pick up from where we left of in the PREVIOUS POST...
The third prototype
I wanted to get rid of number on edges so I turned the thought process around:
How many faces are there? 120
How many faces are required for d20? 40, because 20 x 2 digits each (we give 1..9 the same space)
How many faces are required for d10? 40, because 10 x 2 digits x 2 occurrence
How many faces are remaining? 40.
That's not enough for d12 + d8 + d6 + d4, however, the d12 is sitting on the corner. Why not round those corners, snoop a tiny part of each elongated triangular surface and place them in there? We'll worry about the size of this circle later...
Can we than fit the d8 and d4 when spread out over 40 faces? Well, yes! because they need only 16 faces considering that the d4 will have double occurrence just like the d6 and d10.
It's probably even possible to put the d8 two times on the die and the d4 four times and still have faces empty!
And so the puzzle began. During these steps I focused on getting the d20, d10, d12, and d6 in their 'final' position and then trying to fit in the d8 and d4 in every way possible.
The goal was to:
Add as little extra lines/symbols as possible;
Keep duplicates of d8 and d4 to a minimum;
One 'digit' per face;
Must be done without the need of special manufacturing (for instance two different plastic base colors.)
This was a difficult puzzle because the d8 spanned over many faces and due to the evenly divisible problem it couldn't fit in the d20 or d12 spaces without the need of extra lines or symbols.
I thought of cutting up some d20 faces, but that caused too much confusion even for me.
Finally I realized the follow - and this was the missing piece that made the final product possible:
The d20's triangles consist of 6 faces;
The d8 would requires 15 faces;
If we put a d8 number in a d20 triangle then we have 6 out of 15 faces;
If we then add HALF of EACH surrounding d20 triangle to this (that's 3 x 3)...
... then we have 6 + (3 x 3) = 15 faces!
The only thing remaining was to split all 'adjacent' triangles in two in the most clear way possible and the third prototype was rushed to the resin printer:
Intro to resin casting
I had no idea how to actually make dice besides 3d printing them, so I did what any person would do: look it up on Youtube, spend a few hundred euros in equipment and materials and start following the tutorials step by step!
I used Rybonators dice making guide first and foremost:
Bought materials (silicone, resin, cups, sticks, etc.) and did some testing. After a few test pieces I quickly realized that I really needed a pressure pot because I was simply unable to cast such a 'thick' object without catching bubbles. Invested in a pressure pot and a compressor and tried it again. When I was happy with the test results I started focussing on the dice.
Resin copies made from the third prototype
Now onto the fun stuff: making some dice prototypes!
Actually this started with the LEAST fun thing anyone can image: sanding and polishing the 'master die'. 3d printed parts always have some imperfections like print lines and adhesion points from supports. For this I used sanding/polishing papers that went from coarse 220 grit up to very fine 5.000 grit. Because the faces of the die are small and I needed it to be as flat and accurate as possible I used popsicle sticks and adhered the polishing paper to it, then spend some time making it as perfect as possible.
How much time?
120 faces, each faces required 8 passes of gradually finer polishing paper, and each pass took me about a minute so that's 120 x 8 = 960 minutes = 16 hours!
Wew, that was a lot of time (and a very soar finger) spread out over multiple days, but the end result looked good enough to be made into a resin copy:
Look at that beautiful die with black lines, red d12/d6, blue d20/d10, and green d8/d4!
I was really happy with these first few results. Tried coloring the resin with some glitter and mica powders which created some nice variations.
It has a hefty weight and makes a *bonk* when rolled just like a billiard ball.
With more and more tests I realized that this was actually a functioning product, so I went online, posted it and you guys responded positively - so I continued!
The d8/d4 combo still resulted in a lot of confusion. On this model you can see the tiny diamonds and triangles located next to the d8 and d4 respectively to indicate the standard dice shape.
This was found confusing during testing, and the explanations of the arrows was unclear to many.
Also, even though the weight felt nice in the hand it was too large which made it difficult to read numbers on 'the other side' of the die and I was afraid to use it late in the evening because it could probably wake up the neighbors.
Finally, making a single copy took me a very long time and the mold started wearing out after about 10 copies - so this was not a viable option in the long term.
Still a lot of bubbles and tiny imperfections in the mold. It works, but the next mold has to be better!
The fourth prototype
Once again I went on to designing. The goal was to fix as many of the 'bad' points identified on the previous set of resin copies.
For this I did three major things:
Reduce the size to 40mm in diameter;
Increase the font-to-face ratio to increase readability;
Simplify and differentiate the way d8 and d4 are displayed.
That last one might need some further explanation. Before the d8 and d4 values were displayed with numbers that had a small symbol next to it to indicate d8/d4. I wanted to get rid of this symbol to keep the dice as clean as possible (no unnecessary clutter!! there's enough going on already!)
On this new design the d8 is simplified to having an 8 bit style font, and the d4 number is replaced by small triangles. This design choice made differentiation between d8 and d4 so much easier and reduced the extra symbols required to identify the numbers. The small triangles for the d4 are chosen because a standard tetrahedron/d4 is literally a triangle. Also this is similar to pips for d6 and still easily and quickly countable. Arguably you wouldn't even need to count it because you can literally see immediately how many of the small triangles there are. So that's a win!
Next I asked a friend to try and 3d print some of these on their resin printer. This was actually much more difficult than anticipated (because of my very strict requirements..) but in no time some of these gorgeous green balls were produced:
Some were made hollow with a hole in it for the resin to pour out and others were made as a two-part print.
I finally chose the two-part print, spend ages sanding and polishing, then glued the halves together and spend even more time sanding and polishing. By now I had perfected my polishing technique in the following way:
Put polishing paper on glass plate and tape it into position;
Put the master die on it with the required face down;
Apply pressure with a finger on the top surface;
Use other fingers to rotate the die like its dancing "The Twist".
The results? Much smoother, more consistent, and nicer than the previous master die at the cost of only 30-something hours of polishing time.
Resin copies made from the fourth prototype
With the master die perfected I attempted to create a beter silicone mold for it as well. This went somewhat right. My conclusion: I still have a lot to learn about silicone mold making and resin casting, but this mold is able to produce very usable prototypes that in my opinion look very closely to my intended final product!
At this stage I started making the website, rendered some more images, shot some photos, made the social media posts, and started asking more and more people what they think of it. I actually intended to get some good feedback and then make 100-something dice and put them on Etsy for a few dice enthusiasts...
But WOW!! The response I received from you all was so positive and great, and soon the requests passed a few hundred people!! I had never dreamed of that!
It was clear what had to happen:
Get a manufacturer that can mass produce these dice;
Launch a Kickstarter to satisfy your cravings for these unique dice.
Thanks for the support so far! Let me know if you have any feedback regarding these blog posts or any questions regarding the content!
Next up: more info about the search for a manufacturer...