In Traveling Witch, the player brews potions to bring happiness to people around the world. The game provides a portable cabinet full of wicked ingredients and a half-sized cauldron that hasn’t been cleaned for generations. Players have to figure out what potion to brew based on their conversations with visitors.
The physical controller playset was designed to simulate the act of brewing potions. Based on conversations with visitors, players refer to their notes (the clue book), select the correct ingredients, and stir the cauldron in a particular way to complete the potion. These physical interfaces provide an intuitive, tactile experience that elevates the gameplay’s sense of playfulness and immersion.
I made a cauldron controller in 2020 (here). Originally, the controller was meant to do two things — read multiple ingredients (RFID tags) and use stirring motion (KY-040) as an input. The multi-RFID reader board wasn’t that reliable at the time. Depends on how the RFID tags was stacked, not all tags were read properly. However, stirring both clock and counter-clock wise was working and felt very satisfying. I have wanted to build an example of a simple ID system, I decided to come back to the cauldron and try a different approach to id the ingredients.
The old cauldron was made of KY-040 and Adafruit Feather. I decided to build the ID input module and cauldron with M5Stack this time because a stack of Limit Switch units (SKU: U145) is perfect for creating a simple ID system. I made the ID input module in the form of an alchemy cabinet. The player can insert the ingredient data card from the top of the cabinet and press down to send in ID. Switch-based ID system is a simple but effective method used in many interactive toys. Every switch has an on and off state. I used 5 switches to create 2 to the power of five -1 IDs for this play experience. -1 is because 00000 won’t work in my setup. I am only using 5 ingredient data cards at the moment.
Working with the M5Stack unit – Limit Switches: I wanted to create an ID system using 5 limit switch units. However, I found out only 4 of the 6 ports on pHub unit can be used as digital inputs. The rest two are analog only. This is not a road blocker, it only makes reading switches a little more complicated. https://shop.m5stack.com/products/limit-switch-unit
Working with KY-040: I started out with the basic example code that can be easily found on the internet. There were little unwanted jumps on directions here and there, but it was acceptable. However, when I started to add more code to the sketch, the amount of jumps increased drastically. At first, I had suspected it might have something to do with debouncing. The amount of jumps remains the same after trying both software and hardware debouncing. I then found out the code I added to the sketch, especially in the loop(), slows down each loop cycle by about 6 ms compared to close to 0 ms with just the KY-040 example code. I eventually found a workaround using interrupts which in a nutshell is like threading for Arduino. It works perfectly without any jumps on directions when stirring the cauldron. https://github.com/ownprox/Rotary_Encoder_KY-040_Fixed/
3D models – Medicine Cabinet
3D Model – Cauldron
What goes into a witch’s cauldron?
Macbeth, Act IV, Scene I (William Shakespeare 1564 – 1616)
The three witches, casting a spell [First Witch] Round about the cauldron go; In the poison’d entrails throw. Toad, that under cold stone Days and nights hast thirty one Swelter’d venom sleeping got, Boil thou first i’ the charmed pot. [All] Double, double toil and trouble; Fire burn and cauldron bubble. [Second Witch] Fillet of a fenny snake, In the cauldron boil and bake; Eye of newt, and toe of frog, Wool of bat, and tongue of dog, Adder’s fork, and blind-worm’s sting, Lizard’s leg, and howlet’s wing, For a charm of powerful trouble, Like a hell-broth boil and bubble. [All] Double, double toil and trouble; Fire burn and cauldron bubble. [Third Witch] Scale of dragon, tooth of wolf, Witches’ mummy, maw and gulf Of the ravin’d salt-sea shark, Root of hemlock digg’d i’ the dark, Liver of blaspheming Jew, Gall of goat, and slips of yew Sliver’d in the moon’s eclipse, Nose of Turk, and Tartar’s lips, Finger of birth-strangled babe Ditch-deliver’d by a drab, Make the gruel thick and slab: Add thereto a tiger’s chaudron, For the ingredients of our cauldron. [All] Double, double toil and trouble; Fire burn and cauldron bubble. [Second Witch] Cool it with a baboon’s blood. Then the charm is firm and good.
These gruesome-sounding names are most likely used for dramatic effect rather than literal ingredients. For example, “Tiger’s chaudron” refers to tiger guts or entrails. However, when the ingredient is listed in modern analyses, it is interpreted to mean lady’s mantle, a common herb.
eye of newt: a mustard seed.
fillet of a fenny snake: leech or root of a plant
toe of frog: bulbous buttercup
wool of bat: moss and holly
Owlets wing: garlic
I thought I started with these 5: Hemlock root, toad, fillet of a fenny snake, eye of newt, and wool of bat.
Interestingly, upon examining how D&D books might simplify potion recipes, I was struck by the remarkable similarity between potion-making and the practice of traditional Chinese herbal medicine.
Chinese herbal medicine utilizes ingredients derived from plants, minerals, and animals to restore balance and harmony to the body’s vital energy, or Qi. These components are typically combined into formulas tailored to an individual’s specific condition. Much like the exotic ingredients one might find in a fantasy cauldron, the herbs themselves often bear wonderfully imaginative names.
Earth Dragon: Earthworm
Ground Turtle: A bone-connecting-insect that lives in soil
Dragon Coat: Snake skin
Chicken Inner Gold: The inner lining of a chicken’s gizzard
Blood Residue: A carbonized substance obtained from charred human hair.
Yellow Dragon Extract: Essentially liquid from aged human feces.
In order to prepare a Chinese herbal decoction, first soak the herbs in water for about 30 minutes. Then, boil and simmer the mixture for 45-60 minutes before straining the liquid. The process is repeated with the same herbs, using less water to extract a second, more concentrated decoction. The two liquids are combined to create the final drink. The brewing process and the end product is very similar to potion-making in my mind. The only notable difference from Chinese herbal decoction-making is the specific advice against using a cast iron pot which is what most cauldrons made of.
I had experimented with MR on Pico 4 but the resolution, image stitching, and controller tracking were not ideal. Since my code-free tool kit doesn’t support Meta 3 MR yet, I am going back to native OVR development for this experiment.
Even me can tell it was out of tune. The first prototype was based on a scale found on the internet. I decided to made one myself. I found out there are tuning apps available on my android phone. After testing a few free ones, I eventually found Tuner – Pitched!, it was free and didn’t have the complicated free trail model. I tasted it with a piano first, the app worked very well. My scale turned out to be very different than the one I used in the first prototype, and it sounds so much better! Maybe the scale I found on the internet was made for a different Otamatone model?
Okay, now I got the basic down, kinda, I am going to challenge myself with Le Festin from Pixar’s “Ratatouille”. The Cola Original Taste cover was super inspiring and made an impression of the song on me. Here is 10 hours of it.
“Since TMRCies were known for their creativity and ingenuity Digital was kind enough to donate to TMRC the first PDP 11 to appear at MIT; this made the Club the first MIT student group to ever have a computer! Soon TMRCies combined the operation of WECo’s donations and Digital’s PDP 11 to create cab assignment and switching though the computer. The computer also allowed the implementation of ‘phone operation’ – that is, switches could be thrown via the telephone system within the club room! (This phone system, to be named MaRoto, was also a TMRC customized item.)”
“Here’s what I found: the building was hastily-constructed of plywood. It leaked. It had bad acoustics and was poorly lit, inadequately ventilated, very confusing to navigate (even for people who had been working there for years) and was scorching in the summer and freezing in the winter.”
Occupants of Building 20 in 1963 (from Staff Telephone Directory, compiled by MIT Institute Archives) Army, Military Science Campus Patrol (20C-128) Data Processing (20C-220) Ice Research Lab (20E-206) Industrial Hygiene Lab (20B-245) Lab for Nuclear Science, linear accelerator (20D-014) MIT Press (20B-120) Model Railroad Club TMRC (20E-214) Occupational Medical Services (20B-238) Physics Labs Research Corporation (20B-111) Research Lab of Electronics
PDP-1 attached to a model railroad after its retirement to the University of Massachusetts at Amherst in 1968.
“The PDP-1 that originally went to Bolt, Beranek & Newman to be used in a more formal environment than at MIT was eventually retired to the University of Massachusetts at Amherst in 1968 during my own tenure there and was connected to a model railroad. It still had SpaceWar! as the primary program. (J.A.N. Lee, ibidem, p. 272)”
Posted by the user NoLand Notably, the MIT wasn’t the only university which featured a model railroad, which also leads to a few confusions. E.g, in this context, we have to point out a couple of photos that are frequently distributed with the tag line “The PDP-1 at the Tech Model Railroad Club.”
Here’s one of them and even the CHM has it under the title “History – PDP-1 at the Tech Model Railroad Club” in its catalog.
There are some details that don’t add up. Notably, the PDP-1 was never at the TMRC and there wasn’t even a possible time window for this to happen. As it turns out, this is at the University of Massachusetts Amherst in 1968, and the person seen operating the PDP-1, which was indeed used to control the layout seen in the foreground, was J.A.N. Lee, Head of the then new CS program. The machine seen in the picture, is the PDP-1B production prototype that was originally installed at BBN in 1960 and was once used for Ed Fredkin’s pioneering timesharing system.
(Marginally interesting is that, while the machine had originally a detached operator’s console, this was now mounted to what looks like the integrated paper tape reader found on the production models, which it seems to have acquired over the years and which promotes a somewhat deplorable impression regarding the state of the computer, as if dismantled half-ways. BTW, the PDP-1 is the white (or rather cream colored) cabinets seen in the background of the photo. What’s balancing on the combined paper tape reader and console assembly may be a paper tape punch, as indicated by the rails around what looks like a slot seen at its top. The tape punch is the vertical slot seen at the right of the top front panel of the production models. — And, yes, as the machine retired to UMas, it was found to still have a copy of Space War in its non-volatile core memory.)
And, to close the circle, the image first publicly appeared (with a correct tag line) in a review of Steven Levy’s “Hackers”, the very book that popolarized the notion of the TMRC. This review was written by the same J.A.N. Lee, we see in the photo, and is found on page 271 of the Annals of the History of Computing, vol. 7, no. 3, July, 1985 (J.A.N. Lee, “Reviews: Levy, Steven. Hackers”, pp. 270-272).
When I was reading Dragon Quest: The Adventure of Dai (DRAGON QUEST –ダイの大冒険–) as a kid, I had always fascinated by all the magic artifacts in the story. The magic bullet gun (魔弾銃) was my all-time favorite.
The gun belongs to Maam (マァム), one of the main protagonists in the story. It takes form of a gold-colored large-caliber single-shot pistol. There are a total of 8-10 bullets depends which version it is. For each bullet, a sacred stone that accumulates magical power is embedded at the tip, and by touching it and chanting a spell, it becomes charged with magic. The charged bullet can be loaded into the pistol from the the back by opening the tail cover, one bullet at a time.
After the defeat of Flazzard (フレイザード) on the island of Bulge, large firepower was needed to save Princess Leona (レオナ姫) who was frozen in ice by a forbidden spell. Because of this, Dai (ダイ) charged the bullet through his dragon emblem (竜の紋章), and although he was able to shoot it and melt the ice, due to the recoil from using a bullet with more power than the allowable amount. The gun barrel breaks in half. Badak (バダック), the self-proclaimed “King of Inventions,” tried to repair it, but it was broken beyond repair. Badak said, “Avan (アバン), who created this gun, was a genius.”. The Magic Bullet Gun was no more which is a pity.
There are 2 major gun visuals. The biggest difference is the design of the trigger area and handle. Little torn here, because I like both designs. I am going to focus on Gun Visual No. 2 for this round because I had boxed out a prototype a while ago.
V3V3
This is designed specifically to with M5Stack system. I did hack it a little so I can use the B port for both a limit switch unit (SKU: U145) and a Neopixel ring. I used a RFID 2 unit to detect the bullet. In my first iteration, I had put the unit flat on the button of the bullet chamber. It doesn’t detect the RFID tag at all. Now I have it stand vertically and it works beautifully.
I don’t have solution for sound effects to be built into the gun, but I am going to try the thing I learned from the Chiba Hogwarts — send sound to a Bluetooth speaker. Take advantage of M5Stack’s built-in Bluetooth and connect it to a Unity app that plays the sound effect.
Assembly
[Sound Testing Video]
千葉ホグワーツのハリーティッカー
The next thing I want to improve is the battery life. M5stack Fire comes with a 500 mAh battery bottom and it doesn’t last long. I am planning to create a type C port at the handle and attach one of those mini 5000 mAh power bank. That will provide about 10 times more battery life that what I have now.
Not all pins on the Leonardo and Micro support change interrupts, so only the following can be used for RX: 8, 9, 10, 11, 14 (MISO), 15 (SCK), 16 (MOSI).
“… addForce apply the force at the center of mass.” “simply cheat and do both a AddForce and Addtorque, then you have simple control over your inputs without having the headache of doing proper physics computation for achieving your controls.” “the other option is to use “atPosition” property, this will in effect apply a self force but not from the center of mass and thus imply a torque” http://docs.unity3d.com/ScriptReference/Rigidbody.AddForceAtPosition.html
