Card Shuffler interest (3 Viewers)

[demonnic]

So ever since @jamesjkim posted this thread regarding a 3d printed card shuffler I've become a bit obsessed. I have started working on making one of these for myself without any of the bits for supporting the cheating functionality. Removing all of that not only makes it cheaper and more trustworthy, but easier to build and design.

Which also got me to thinking about making it something my roommate and I could sell for some extra cash on the side. That being said, I'm curious what it is people want out of a shuffler and what they might be willing to pay for one. My main ideas right now are:

Operation should be simple.
Preferably a button to lower the deck and begin shuffling, and the same button at the end of the shuffling cycle presents the deck. Does it need to be any more configurable than this really? I know the shuffletech lets you choose whether to shuffle 3 or 7 times, but I feel like this method of shuffling is inherently more random to start and could probably get away with having it just set one way and left alone. Does anyone actually set their shuffletech to 3? Real question.
I also have no intention of having this deal anything. I want it to take a deck of cards, shuffle it, and return the deck for use. That's it.

Flush mountable should definitely be an option.
Should be able to put it in your table or a sidecart relatively simply. I don't think this will be at all difficult to make happen, given it's already a top loading design.

Maintainable
Should be able to get replacement parts, especially for things which are likely to wear down eventually over time. Things like the o-rings on the rollers and the belts and even the gears should be relatively easy to obtain so that you can keep it working.

Trustworthy
I'm not trying to sell this to casinos or anything, but I'm hoping some people might have ideas as to how to increase trust in the device. After all even after we've refined the design and tightened up build quality, I'm still just some schmuck on a poker chip forum. I was thinking I might send a prototype unit to a few users here once I get to that point, if they're willing. @minedelta seems obvious with their deckmate experience, but I'd love to hear suggestions for other people whose opinions hold weight on the subject here, as well as other things which could be done to build up that trust. I wonder if there's a company which can do an evaluation for fairness that doesn't cost an arm and a leg?

Affordable
My goal is to get this at or below the cost of a brand new shuffletech, with similar or better build quality/durability. How attainable that goal is remains to be seen, but I think I can manage it. The price jump from "smashes two halves of a deck together" to "takes a whole deck and splits it in two, then smashes them together; repeat" always struck me as a bit nuts. Assuming it doesn't sound like a freight train coming through and produces a well shuffled deck of cards every time, what would you be willing to pay for such a thing? Or would you rather just stick with the existing options, better the devil you know and all that?

I'm waiting on parts to arrive to build the first one and printing other parts in the mean time. I'm definitely making at least one of these for my own use, as I want a shuffler and I don't want to shell out for the ST1000 or deckmate. But I think there's a hole in the market I could fill here and feel good about doing it.

Thanks for reading my novella.

 

[jamesjkim]

My only minor concern would be time to shuffle the deck. It sounds like cost would be relatively affordable which is a plus. I’d be happy to be a tester too.

Following with lots of interest.

 

[demonnic]

My only minor concern would be time to shuffle the deck. It sounds like cost would be relatively affordable which is a plus. I’d be happy to be a tester too.

Following with lots of interest.

The one in the video took about a minute and a half iirc. Which isn't too bad, but could also probably be optimized for. What would folks consider the maximum allowable time to shuffle?

 

[jamesjkim]

The one in the video took about a minute and a half iirc. Which isn't too bad, but could also probably be optimized for. What would folks consider the maximum allowable time to shuffle?

Probably shoot for <1 min

 

[ekricket]

Just hook one of these up to a track that moves up and down and “deals” into your sorting rack. It’s already got the single card dispenser roller system and the deck holder.

Or you could move your sorting rack up and down in front of this.

https://www.justpokersupplies.com/p...dlZfgRjwgSM8a5l10djn3tYoiyT3FgTRoCoTgQAvD_BwE

 

[demonnic]

I think trying to interface with that might be more difficult to do than getting the single card dispensing working as intended, but it's certainly a thought. I think I would prefer not to rely on a fully assembled third party product like that though, if I can't find them or they change their shape we have to change everything to fit a new.

For a little bit of context/background I'm a software developer with a penchant for tinkering and the like, and my roommate is a mechatronics engineer. So I'm confident we have the technical skills to see this through.

 

[buzzmonkey]

A simple, if not cumbersome test, would be to put a sorted deck in, have it do the shuffle, then record the order. Once you've recorded a sufficiently large sample size, you can analyze, possible with a LLM AI to look for any repeating patterns.

 

[demonnic]

A simple, if not cumbersome test, would be to put a sorted deck in, have it do the shuffle, then record the order. Once you've recorded a sufficiently large sample size, you can analyze, possible with a LLM AI to look for any repeating patterns.

That's a good idea and worth doing I think.

 

[buzzmonkey]

For the manual portion, it would be most efficient to have a number of identical decks, a person to record the order, a person or 2 to re-sort them and a person to feed them into the machine.

 

[demonnic]

Yeah I've got a stack of used 2018 WSOP decks that would work well for that. Intend for it to work with both bridge and poker size cards

 

[Undream]

Watching.

Plenty of issues with the shuffletech. Obviously, they are a little delicate, I've had to had mine serviced once already and I need to send it back in again soon.... Durability of parts is a concern too. My grandkids broke the clear door off mine since the hinge was thin plastic.

My biggest complaint with the shuffletech when it is working well (and I do absolutely love it) is the noise. It's just kinda of annoying. I feel like it could have been designed with a different material than plastic on the parts where the cards hit /are stopped during the shuffle. Something more solid or thicker at least, that wouldnt echo inside the thing. It's not a huge deal at a table of 8+ guys yapping away during a game with a couple TVs on. But, it does kind of get old later in the evening if we are short handed.

 

[demonnic]

Watching.

Plenty of issues with the shuffletech. Obviously, they are a little delicate, I've had to had mine serviced once already and I need to send it back in again soon.... Durability of parts is a concern too. My grandkids broke the clear door off mine since the hinge was thin plastic.

My biggest complaint with the shuffletech when it is working well (and I do absolutely love it) is the noise. It's just kinda of annoying. I feel like it could have been designed with a different material than plastic on the parts where the cards hit /are stopped during the shuffle. Something more solid or thicker at least, that wouldnt echo inside the thing. It's not a huge deal at a table of 8+ guys yapping away during a game with a couple TVs on. But, it does kind of get old later in the evening if we are short handed.

The noise is a complaint I see fairly frequently, though I see almost as many people say it isn't that bad. But to me that means half of the folks who might be interested in what I'm making think the the noise is an issue, and that's a lot. So finding a way to keep it quiet is definitely on the list of goals.

Since I also intend to use this myself, it'll need to be durable. I don't want to have folks needing to ship them back to me for repair or anything. I mean, I stand behind my work. But I'd like to stand behind it from way over here rather than up close with my hands in its guts. With all the testing I'm certain we'll get thousands and thousands of cycles run and where things want to break down or go awry.

 

[Rakrul]

Just hook one of these up to a track that moves up and down and “deals” into your sorting rack. It’s already got the single card dispenser roller system and the deck holder.

Or you could move your sorting rack up and down in front of this.

https://www.justpokersupplies.com/p...dlZfgRjwgSM8a5l10djn3tYoiyT3FgTRoCoTgQAvD_BwE

I like how they fucked up the FX here. It looks like the loose card is hurling itself towards the dealer.

 

[Rakrul]

So ever since @jamesjkim posted this thread regarding a 3d printed card shuffler I've become a bit obsessed. I have started working on making one of these for myself without any of the bits for supporting the cheating functionality. Removing all of that not only makes it cheaper and more trustworthy, but easier to build and design.

Which also got me to thinking about making it something my roommate and I could sell for some extra cash on the side. That being said, I'm curious what it is people want out of a shuffler and what they might be willing to pay for one. My main ideas right now are:

Operation should be simple.
Preferably a button to lower the deck and begin shuffling, and the same button at the end of the shuffling cycle presents the deck. Does it need to be any more configurable than this really? I know the shuffletech lets you choose whether to shuffle 3 or 7 times, but I feel like this method of shuffling is inherently more random to start and could probably get away with having it just set one way and left alone. Does anyone actually set their shuffletech to 3? Real question.
I also have no intention of having this deal anything. I want it to take a deck of cards, shuffle it, and return the deck for use. That's it.

Flush mountable should definitely be an option.
Should be able to put it in your table or a sidecart relatively simply. I don't think this will be at all difficult to make happen, given it's already a top loading design.

Maintainable
Should be able to get replacement parts, especially for things which are likely to wear down eventually over time. Things like the o-rings on the rollers and the belts and even the gears should be relatively easy to obtain so that you can keep it working.

Trustworthy
I'm not trying to sell this to casinos or anything, but I'm hoping some people might have ideas as to how to increase trust in the device. After all even after we've refined the design and tightened up build quality, I'm still just some schmuck on a poker chip forum. I was thinking I might send a prototype unit to a few users here once I get to that point, if they're willing. @minedelta seems obvious with their deckmate experience, but I'd love to hear suggestions for other people whose opinions hold weight on the subject here, as well as other things which could be done to build up that trust. I wonder if there's a company which can do an evaluation for fairness that doesn't cost an arm and a leg?

Affordable
My goal is to get this at or below the cost of a brand new shuffletech, with similar or better build quality/durability. How attainable that goal is remains to be seen, but I think I can manage it. The price jump from "smashes two halves of a deck together" to "takes a whole deck and splits it in two, then smashes them together; repeat" always struck me as a bit nuts. Assuming it doesn't sound like a freight train coming through and produces a well shuffled deck of cards every time, what would you be willing to pay for such a thing? Or would you rather just stick with the existing options, better the devil you know and all that?

I'm waiting on parts to arrive to build the first one and printing other parts in the mean time. I'm definitely making at least one of these for my own use, as I want a shuffler and I don't want to shell out for the ST1000 or deckmate. But I think there's a hole in the market I could fill here and feel good about doing it.

Thanks for reading my novella.

It has to be able to mount in table for it to be interesting for me. Bonus points if you can reuse the shuffletech flush mount plate. As programmer you probably have some tools to ensure and verify "true random" which I see as the biggest hurdle.

Other things;
Reliable
Shuffle time
Noise
Bridge and pokersize compatible
Unjamming function
Easy to self-repair

 

[demonnic]

It has to be able to mount in table for it to be interesting for me. Bonus points if you can reuse the shuffletech flush mount plate. As programmer you probably have some tools to ensure and verify "true random" which I see as the biggest hurdle.

Other things;
Reliable
Shuffle time
Noise
Bridge and pokersize compatible
Unjamming function
Easy to self-repair

Thanks for the input!

Being able to reuse the shuffletech plate is a good idea. Will definitely keep that as a potential goal. Perhaps an adapter if direct use doesn't make sense. Definitely want to be able to flush mount if desired either way.

Shuffle time and noise definitely come up in every card shuffler thread I've seen on here, which to me says they're top of mind for a lot of people. I'll definitely be keeping those in mind while refining the design and optimizing it.

I use both poker and bridge sized setups myself, so it will have to work with both. I'm hoping to make it so it doesn't really need a special 'unjamming' function but easy to self-repair is a must, as I'll be self-repairing my own and I don't want to make work for myself.

 

[CoolDude53]

@demonnic I have been doing the same. I got custom PCBs made and assembled, if you want one lmk.

I have all the mechanical parts, just waiting on the 3d printed parts to ship. Will keep you updated on my prototyping - excited someone else is working on it too!

 

[demonnic]

@demonnic I have been doing the same. I got custom PCBs made and assembled, if you want one lmk.

I have all the mechanical parts, just waiting on the 3d printed parts to ship. Will keep you updated on my prototyping - excited someone else is working on it too!

View attachment 1435408

I don't intend to use their design directly, necessarily. There's a lot of extra complexity there I just don't need for a simple card shuffler. That being said if you had a MOQ and have an extra one of the PCBs printed I'll happily take one off your hands, might be able to use it for initial prototyping while I trim the fat.

 

[demonnic]

Alright, did some scouring of videos and such on the internet and it seems the st1000 takes about 90 seconds to do the 7 times shuffle. And it looks like it jammed a couple times at least in each video. It was able to unjam itself and continue operation but it definitely slowed the shuffler down.

I think my first goal will be sub 90 second shuffle time, and then work for sub 60 second shuffle time after that.

I also think with this general design, jamming shouldn't be a huge issue. And I'm not nearly as concerned with dispensing only one card at a time every single time as the original project since I'm not trying to set card order or anything. The occasional double up would just be an additional random element. No huhu as long as it's not happening frequently.

I'm thinking that tuning for Copag out of the box makes some sense, since they're probably the most ubiquitous plastic card out there. I have no real intention of supporting paper cards, the st1000 manual says to steer clear of paper cards and honestly if you're paying hundreds for a shuffler you can invest in some plastic setups. Paper cards are more hassle than they're worth.

 

[spartan037]

Have you see the leShuffler: https://www.pokerchipforum.com/threads/leshuffler-a-competitor-for-shuffle-tech.123566/

 

[demonnic]

Yeah, I'm familiar with it. Was one of the more frequent commenters. I think it's a pretty nice product.

But I also think there were design decisions made for it which will make it a less attractive option for many people as demonstrated in the threads here.

Having a continuous shuffler that will dispense cards for you and take them back in without the need to actually physically mix the cards between each hand is in many ways a smart decision, but personally I just want a machine I can put a deck of cards into, and then get a mixed deck of cards out of.

Must sit on top of the table or side cart: I'd really prefer a flush mounted option.

The unit does not look particularly end user serviceable. One of my goals is to be able to provide instructions and parts so users can fix it themselves without having to send anything in for servicing if they have even the smallest modicum of mechanical gumption.

And also it's a kickstarter, which I know turns a lot of people off. My plan is to make one for myself, get it where I can start fulfilling orders, and start shipping with the first order rather than do the whole crowd funding thing.

 

[Dbxela]

I'd been thinking about this too. One slightly different idea I have is that I don't have the same space limits that casinos or Shuffletech are aiming to satisfy.

I can imagine something with an artistic furniture factor and be based on a floor standing mobile cabinet, so if it worked like the youtube DIY example it could have a 52 slot tray system sitting tall like a CD shelf tower. This would get it down to 1 shuffle step for any seed, it could probably move slower and more elegantly, and still be quite quick (1 pass to fill the trays in order, then the shuffled collection process, or fill the trays shuffled and find a cool way to knock them all back out for collection). I'd try to show its shuffle process working (without revealing card faces obviously), kind of like a building with external lifts.

 

[SetToBean]

Alright, did some scouring of videos and such on the internet and it seems the st1000 takes about 90 seconds to do the 7 times shuffle. And it looks like it jammed a couple times at least in each video. It was able to unjam itself and continue operation but it definitely slowed the shuffler down.

I think my first goal will be sub 90 second shuffle time, and then work for sub 60 second shuffle time after that.

I also think with this general design, jamming shouldn't be a huge issue. And I'm not nearly as concerned with dispensing only one card at a time every single time as the original project since I'm not trying to set card order or anything. The occasional double up would just be an additional random element. No huhu as long as it's not happening frequently.

I'm thinking that tuning for Copag out of the box makes some sense, since they're probably the most ubiquitous plastic card out there. I have no real intention of supporting paper cards, the st1000 manual says to steer clear of paper cards and honestly if you're paying hundreds for a shuffler you can invest in some plastic setups. Paper cards are more hassle than they're worth.

90 seconds and regular soft-jamming seems to be the characteristic of an old and/or poorly cleaned shuffler possibly using sub-optimal cards. My ST1000 unit, delivered in October 2024 and used for ~5 poker nights so far with COPAGs and similarly stiff cards like Desjgns, 7-shuffles in just under 60s and only experiences soft-jams a few times a night.

 

[Undream]

I've used my st1000 for 4 years. The machine gets about 22 card games a year, and each game lasting about 6-7 hours, going to say at least 550 hours of 20+ hands an hour. We really keep the game moving.

I've had it ruin about 12 cards in that time (hard jam to the point where it permanently creases a card). Tried other brands than copag and get similar performance. As for soft jams, I probably get 6-12 a night. More humidity of summer usually is higher end.

Cleaning the rollers doesn't really seem to completely fix the jamming. I actually think that part of the reason they start jamming the older they are is the plastic "catch trays" that have the edges of the cards hit them tens of thousands of times, slowly wear and get shiny and slippery, causing the cards to slip around just enough to occasionally pop up and out the edge of the trays.

I actually put some painters tape on the sides of the trays to help create a bit of stopping resistance to prevent this. It helps a little.

I think with the design of the shuffler where the cards are inserted into the "pockets" it would significantly reduce the possibilities of jams. There's just too much open space in a shuffletech for things to go wrong.

 

[demonnic]

I'd been thinking about this too. One slightly different idea I have is that I don't have the same space limits that casinos or Shuffletech are aiming to satisfy.

I can imagine something with an artistic furniture factor and be based on a floor standing mobile cabinet, so if it worked like the youtube DIY example it could have a 52 slot tray system sitting tall like a CD shelf tower. This would get it down to 1 shuffle step for any seed, it could probably move slower and more elegantly, and still be quite quick (1 pass to fill the trays in order, then the shuffled collection process, or fill the trays shuffled and find a cool way to knock them all back out for collection). I'd try to show its shuffle process working (without revealing card faces obviously), kind of like a building with external lifts.

That's certainly a thought, and if you decide to build that I'll definitely follow the thread with interest. I do think there's probably good reasons most "casino grade" (for lack of a better term) shufflers utilize a drum rather than the vertical stack like this one, especially once you start having a cubby per card or multiple decks involved. This is one of the design decisions which makes sense in leShuffler when you look at how it's operating. A 52 slot stack bin like that would be pretty tall, and reducing the height of the slots to make the overall height smaller would introduce more complexity in ensuring you only get one card per slot every time.

I'm not sure I'd want to build such a thing as a product, maybe just a one off for myself or something.

90 seconds and regular soft-jamming seems to be the characteristic of an old and/or poorly cleaned shuffler possibly using sub-optimal cards. My ST1000 unit, delivered in October 2024 and used for ~5 poker nights so far with COPAGs and similarly stiff cards like Desjgns, 7-shuffles in just under 60s and only experiences soft-jams a few times a night.

Fair enough, and worth keeping in mind. Their manual states 70-80 seconds so < 90s still feels like a good first goal during prototyping though. It's entirely possible the videos I found were of mostly older machines, it's not as easy as I might have hoped to find videos of people just running their shufflers =)

I've used my st1000 for 4 years. The machine gets about 22 card games a year, and each game lasting about 6-7 hours, going to say at least 550 hours of 20+ hands an hour. We really keep the game moving.

I've had it ruin about 12 cards in that time (hard jam to the point where it permanently creases a card). Tried other brands than copag and get similar performance. As for soft jams, I probably get 6-12 a night. More humidity of summer usually is higher end.

Cleaning the rollers doesn't really seem to completely fix the jamming. I actually think that part of the reason they start jamming the older they are is the plastic "catch trays" that have the edges of the cards hit them tens of thousands of times, slowly wear and get shiny and slippery, causing the cards to slip around just enough to occasionally pop up and out the edge of the trays.

I actually put some painters tape on the sides of the trays to help create a bit of stopping resistance to prevent this. It helps a little.

I think with the design of the shuffler where the cards are inserted into the "pockets" it would significantly reduce the possibilities of jams. There's just too much open space in a shuffletech for things to go wrong.

Thanks for the info. Physically interleaving cards reliably from two stacks is not an easy task for a simple machine to carry out, which is why I didn't even consider it for anything I might build myself. For what it is, the ST1000 does a pretty good job of working around the difficulties of the task.

I believe that trying to adhere to this "like a human does it" interleaving design is a mistake though if your goal is just a randomly mixed deck of cards and not specifically to replicate actual riffle shuffling.

I am hoping that once I get close to go time I'll be able to find someone relatively local with an ST1000 they are willing to let me borrow for noise, speed, and size comparisons. Not wanting to spend $700 on an ST1000 is part of what sent me down this path so I'd like to avoid having to do it anyway for research purposes heh.

 

[demonnic]

3d printer go brrr. Honestly love the first prototypes and using up bits of 'leftover' filament for it.

 

[TheChipVault]

watching with interest

 

[Relik]

Following with interest

 

[suited67]

Following

All my Best Jeff

 

[Lts923]

 

[demonnic]

Got all the remaining bits and bobs ordered that I couldn't print or already had lying around. With any luck I'll be able to start testing by the new year =)