🇳🇱
I really hope I dont get screwed over on this kickstarter... I should be getting my #Chess board next month if all goes well.. Excited about this one.
https://www.kickstarter.com/projects/chessnut/chessnut-evo-human-ai-powered-chessboard
@freemo
I expected for it to move the pieces automatically 😢
@skobkin all the stuff out there that does that has internal moving parts and is rather big... so I specifically dont want that, at least not until someone makes a solid state one.
@freemo
Yeah. I mean... electromagnets are with us for a... couple+ of centuries 😁
I don't see a big problem implementing that. They obviously have some kind of NFC/RFID identifiers in every piece, just need to also implement moving them. Yes, they'll also need some algorithm temporarily moving pieces that are on the way, but it's really not a new challenge and we have a lot of code like that in games.
It'll become more expensive I guess because of amounts of copper coils needed for that, but... wouldn't it worth it? I mean really seing the pieces moving by themselves...
@skobkin yea i similarly thought using electromagnets should work.. you would need a grid of them smaller than the pieces for the motion to be smooth, but im not sure why people use other approaches.
@freemo
Yeah. We already have Qi chargers which utilize coil grids to charge multiple devices at once almost in any position. It most likely is just about scaling this and using more power 🤔
I personally also wouldn't mind if the pieces move in one quick slide from one position to another. I think it's possible to implement with only 64 coils (or even less). If we need to move a piece to an adjacent position, we can power up target coil with high voltage and other coils nearby with lower so only the piece which is not being held back would move 🤔
Using pulse modulation in combination with real-time position computation can also allow more smooth movements with reduced number of coils...
Looks like we just invented a magic chess board 🤣
@skobkin well there are chessboards that move pieces though.. a few models.. the reason i think it may be harder to do is becasue why would they build all these fancy robots inside the board that use magnets to move it.. surely the first thing they tried was a grid of coils.. surely, at least before designing a hoard of robots inside the damn thing.
@freemo
I mean... Mounting one magnet to a caret moving inside by two servo motors VS dozens of expensive copper coils 🙄
Let's compare the profits...
@skobkin im not so sure of that.. i mean wire isnt that expensive, and those robots need to communicate nad have processing power... id imagine the coils would be cheaper, at the very least affordable.. just taling about some chunks of iron and a bunch of wire.
@freemo
We still need at least the same amount of computations if not more when working with coils. Especially if trying to achieve smooth movements. We also need something to manage high currents on such a large scale. Let's say, we won't be considering fully paralleled operation of all coils at the same time. But we still need to be able to power at least ~8 coils at the same time.
Main challenge from my PoV is potential magnetic interference which can reduce the efficiency of each coil. But it's a task for some physics modeling, not just sudden brainstorm in Mastodon comments.
@skobkin yea this is one of those things we would have to test.. im not even sure how much power youd really need.
@freemo
Yeah. After solving the power problem, we may also need thick wire to avoid overheating or even burnout which can make this not only expensive, but also quite heavy.
But... Moving pieces are still fun!
@skobkin usually electromagnets using thin wire wth many turns
@freemo Yeah, it's more efficient, but minimally sufficient thickness of the wire is still a thing to determine. Also a subject for modeling and calculation...
All this sounds like a very interesting engineering project.
@skobkin very doable without much advanced EE... but still a tehcnical challenge. Sure, would be fun.
@freemo
... to find out that it's already patented in the end.
But still fun!
checkers with some ferrofluid looks easy enough :)
@skobkin being done with a small ball:
https://incoherency.co.uk/blog/stories/electromagnet-grid.html
@freemo
Ahah
> The idea for this project came from the Regium KickStarter scam. Regium is an automatic chess board.
Bingo 🤣
@freemo
How interesting and cute. And also using pulse magnets for movement.
@freemo Yes. I was thinking about jerky movements in the first PoC model. Then it could be fixed by utilizing magnetic field pulsation with real-time position tracking.
We'd obviously need teflon mats for each piece to reduce friction.
Of course it's be harder than that, but... The sole fact that it doesn't exist doesn't mean that it's very hard to do. I mean if these guys didn't slap a tablet to the chessboard while making it RGB-lit at the same time, I wouldn't think that someone needed that at all... But here we are.
Good luck with this campaign by the way. Hope it'll be released and delivered to you.