LED driver nerd here! What you're referring to as "frames per second" here is really the PWM cycle time, which is a separate measurement to frame rate (the rate at which the colours can be updated). With a 400kHz timing resolution, that 12-bit depth is running at an effective PWM frequency of just 98Hz. This puts you well into the increased flicker risk zone as per IEEE1789, meaning you're going to get significant stroboscopic effects that may be quite unpleasant to viewers, and would potentially pose a small risk to people with photosensitive epilepsy (PSE). This is slightly alleviated by the population count approach (multiple shorter pulses per cycle instead of a single on/off duty per cycle) but the worst-case flicker remains identical to plain PWM. At 8-bit depth and 1.5kHz you're back in the low flicker risk zone, so that's much better, although with just 8 bits of depth and no dithering your perceptual dimming ratio only ends up being 20:1 (based on sRGB EOTF) due to the nonlinearity between radiant intensity and perceptual intensity. If you decouple your update rate from the PWM frequency and implement temporal dithering, you can get 10-bit depth (8-bit linear, 2-bit dither) for a 79:1 dimming ratio, with the lower bound of the dither ripple frequency being 390Hz (and therefore the maximum FPS also being 390), which is ok. I built a whole calculator for this stuff; if you search "LED PWM calculator" it should come up as the top result. Getting high dimming ratios out of linear PWM without stroboscopic effects is hard work. To reach 256 perceptually linear steps without dithering you need about 12 bits of PWM. But to fully avoid stroboscopic effects, you also need 3kHz or more of PWM frequency. That leaves you needing a timing resolution equivalent to 12.3MHz. That means your driver rise/fall times need to be on the order of 20 nanoseconds, which means about 1A/us of dI/dt per 20mA LED, so you start to have to consider parasitic inductance in your power delivery network and both conducted and radiative EMI in the HF bands. A 2L board won't cut it, you'll need 4L with an SGGS stackup and good layout. It's quite the challenge!
As a quick addendum, if you want to get into the really high end of stuff, look up shunt dimming with adjustable constant-current switching regulators. Dimming is achieved by a combination of adjusting the regulator's output current and shunting the LED current by placing a MOSFET in parallel with the LED chain. When the MOSFET is switched on, the current flows through it instead of the LEDs, turning them off. If you drive the MOSFET with PWM, you get a dimming effect. The MOSFET and its drive circuit are selected to produce a similar impedance to the LED chain, so that the inductor current in the switching converter remains roughly constant regardless of whether the LEDs are being shunted or not, thus allowing for very low ripple current and tight forward current regulation from the buck (while also minimising EMI). A fast trench FET (or, more recently, GaN) is typically used, sometimes with an additional protection diode if the LED chain is long enough to have appreciable inductance. Since the two dimming controls are multiplicative, you can achieve extremely good dimming ratios. It takes a lot of engineering work to get right though.
1) It's reassuring that even a professional can make silly mistakes 2) It's *incredibly valuable* (especially to someone like myself that's never made a PCB) what some of the "gotchas" could be in making one and that you've documented it 👍
Optical links might be more stable for data connections. That's what we do for the majority of data in our slip rings. Two chunky brushes for ground and +300V, and all data goes through a fibre connection which is free to rotate in the centre.
@@spherarec I imagine a brush, but big and chunky! A brush is something they often use in motors, hence "brushless" motors. A brush is a way of passing electricity, which could be power, or a signal, to something that's rotating. You have a metal ring attached to the rotating part, and a "brush" attached to the static part. The brush can be a copper contact, that rubs against the rotating ring, so is in constant contact, to pass the electricity. Old electric drills, mains not battery, are a good example. They use spring-loaded blocks of carbon for the brushes. These push against the rotating copper rings of the motor to pass power. The carbon, graphite, wears away over time, so you can undo a screw and replace them with new ones. Graphite is nice because it's smooth, soft, and essentially self-lubricating. It starts off with a square edge but soon wears away into a concave shape, conforming properly to the brushes. Sometimes you can see blue sparks through the vent holes in a drill, that's the brushes making not-quite-perfect contact. Might happen more as the brushes wear away and copper starts pushing against copper (which damages it, so stop!). Old electric drills are great. They're almost indestructible, and the replacable carbon brushes mean they can last forever. Don't make 'em like that any more!
Seeing your mistakes right off the bat makes me feel a lot better about the silly mistakes I make while working on projects. I tend to beat myself up over them even if I know that it isn't fair to myself to do so
I love your solution! I wrote Windows automation many years ago, and wrote a ChatSleep() function. It would take a minimum time to wait -- so it doesn't hammer on the CPU, like your solution -- and an interval, and function to call to check a state, which it would then start checking. Each time through it would double the interval. Something of the reverse of yours, but the use case was different: it was to catch the next state change, so the automation could proceed, so doing it as fast as possible was a goal. Neat to see a reflection of it! Thank you!
These remind me of an old electronic game i had as a kid, Scrabble Flash. They were little tiles with mono LCDs and a button that would display a letter and then communicate either wirelessly or via IR to tell if they were in the right order to be a word. They would also function as a multi display menu for the 3 different games. Oh the memories.
I did something similar to that a few years ago, not with pogo pins and magnets, just pin headers and jumpers, and programmed a game of life for it. The total playing field would configure itself depending on the placement of adjacent matrices. Not at runtime, but I think that would be a trivial addition to the software.
11:50 I wonder if you could display an image at a higher resolution for a camera this way, since you can basically control each line at image capture. Or even include secret messages only for video cameras, not visible to the naked eye.
This is super inspiring. I've been imagining how a magnetic LED tile would work and this is almost exactly how I would have done it. I'm excited to see how much of a difference the magnetic connectors will make compared to the pogo pins.
I have been looking for a high-side driver for years... YEARS!!! THANKYOU for putting me out of my misery!!! ❤❤❤😭😭😭 Also thankyou for going through the PCB assembly process, I'm only just starting to learn about it and have found it very intimidating, so again, thankyou! ❤
This reminds me of the MBLed project which was done by three french students about a decade ago, they used used similar matricies and used optical transceivers for data. They have a yt channel by the name of "MBLedRose".
Nice magnetic LED matrix, really feel excited when seeing the LED matrix work and shining after a series of fix work, thanks for your great sharring, super cool👍👍👍
It's cool to see someone using the pogo pin connectors and separate magnets. My brain has wanted to use these for something at some point so it's really cool to see them in action with such success! Also, I 100% understand the feeling of being judged by parts I bought many years ago... the struggle is all too real.
I have always enjoyed your content, this blows me away at how smart you are! I have a bunch of those matrix units and have not done anything with them for five years now. pls correct the small bugs and make them available, I would buy 10 assembled boards.
You got me into led matrices, recently made a 16x18 led matrix with the ws2812b leds in a nice frame. Playing with that using esp32 atm. I also have one of the led matrices you've shown in this video, now I will definitely also play around with that one.
Pogo pins with stronger springs or a very low tech method would be to run the contacts on some pure silver kind of like the pencil trick but with a more conductive material. Obviously graphene would be better. But a quick and dirty silver rub might fix the connections . Electroplating is super easy. You would be surprised at what you can get metal to stick to with a tiny amount of voltage. No fancy requirements. White vinegar and whatever you want to plate with add a few volts at low amperage then add table salt until your item to be plated starts attracting your positively charged coating material. Electroplating might be a fun little side rabbit hole to go down when you need a break from coding. You do end up with a bunch of nickel plated pennies and copper coated quarters in similar fashion to all the plastic fidget type dealios you printed out when you first get a 3D printer. You get to use the coins still at least and they make a good conversation piece. Super easy to do, barely an inconvenience.
this is super super cool. I have a lot of those 8x8 black framed matrices lying around as well and have always had something like this in mind. well done
HDR systems can use a limited bit depth to be signalled but then you use extra bits to indicate the dynamic range mapping of the scene. This allows you to position the limited bit depth within a window.
That’s really cool. One thing you could do to ensure better connections between blocks is to use 2 pins for + and 2 for - instead of just 1 each. Then use a data over power scheme with capacitors to separate the a/c from d/c. So the left 2 pins could be + and transmit and the right 2 pins could be - and receive. Good luck!
How cool, in my Disco i so often think of something like that for behind the DJ, so it could be easily transported and adapted to different Rooms, especially with sound controlled Light would be nice!
I love this! I really like how the display adapts automatically as you attach multiple units, I'm trying to think of a way that it could be made into a fun childs toy or game, maybe you could create Tetris on it but have the playing board adapt to the layout of the modules? This is the only idea I can think of right now
I like the idea of no controls, except rearranging the modules. Maybe not Tetris but some puzzle game, or even something where a snake goes off the edge of one screen, and you have others that can divert him and change his direction, maybe you have to get him home. At the simplest, you could even juggle balls. Adding motion sensors would be nice too so you can tilt the displays and have them recognise that. With enough screens you could do all sorts of stuff! Possibly each screen can change between several options, to save the need to have so many, but then that needs a control method. You could use tilt, you turn the screen around to change between configurations. That might be part of a static puzzle, or a moving action game. Yeah these'd be a hell of a toy! And I think they've got a lot cheaper since advertising companies started making billboards out hundreds of them. PATENT PENDING! PATENT PENDING! PATENT PENDING!
Oh of course, Pipemania would be the natural game for these! But you'd need, what, 30 or so? Would also be nice, while I'm spending imaginary money, to use a controller with Bluetooth or just local RF. So at the start of the game, one block could assert itself as the controller, the boss. And it could tell each other screen what pipe shape to turn into. It'd be better than a normal video game since it involves real objects, even if they're just screens for virtual objects. Instead of a joystick, you're physically connecting parts together. Get kids away from the screen while still letting them play video games. Give them some tactility, practice their fine motor control for the other 8 fingers that aren't stuck to a tiny thumbstick. Now I realise, they'd each need a battery on board. Oh dear. BOM went right up! Wonder how cheap it could be made? If you put 30 in each box, you'd be up to the 10,000 quantity just to make 300 units. You might just make a prototype and take it round toy companies, you'd need their support anyway unless you wanna sell it on Etsy to a tiny fraction of customers. Oh, and if you could integrate it into one of those programming platforms for kids, like the MIT one, you'd be able to offer it as something to learn to program. Of course 98% of kids wouldn't do it, just like we got our 8-bit computers by lying about doing homework on them! Then again 100% of kids would only be 50% of the customers, as adult nerds try figure out how to hack them.
you'll get much better color resolution is you use PDM (Pulse-density modulation) instead of PWM. the effective bit resolution approaches infinity amortized over time, even for low actual refresh rates. the only limit is the size of the error calculation, which can trivially be 32-bits per channel.
you shouldn't try to push pins against pins, if you can avoid that. you'll want the data pins pushing onto pads. so make the bottom and left side be pins, and the top and right side be pads. that'd make that work a lot more reliably I bet.
I love this , can be used to display pixel art , would be cool if you give every led their own room with like milk glass above it. You will get a very square pixel. Would love to give you some pixel animations I made , they're mostly 16x16 or 33x32
Ha! I have a box full of LED matrix displays that I bought 20 years ago, and they are also still taunting me. They're rectangular, and only red, but enough to make a small poster. I need to get back to it finally! Part of my problem back then with all my ideas was that microproc tech just wasn't ready yet. The clock speeds were just too slow to do anything I wanted.
Looks really good, what about transmitting power through the magnets itself? You could have more pins reserved for future updates or more bandwidth and the magnets would handle the power throughput just fine making misalignment less of an issue.
I think you just need longer pogo pins and a better mechanical mate to get your connection to be more stable. some alignment pins in the 3d print would help a lot
I was going to do a similar thing myself then i read the comments and i think that's alright. it was either play with some LEDS or some FOC/ESC, and I think it 's time to play with some motors lol
Another thing you can do to free up processor time in this kind of a setup, if the MCU supports it, is using timer-triggered [circular] DMA to automatically feed the GPIO from memory without any processor involvement. I've done it on STM32F0x (to drive 8 7-segment displays from one GPIO bank), but I'm not familiar with CH32V, so not sure if it'd work there. Of course, it'll still use a lot of memory bandwidth if you're running at ludicrous speed for bit depth reasons, so the bit planes + variable-duration timer might still work better in this case. Of course, if you can figure out a way to drive the DMA with an automated variable-duration timer, you could have the benefits of both. (Not sure off the top of my head if you could update the timer configuration with another DMA - or use another DMA-driven GPIO to output the wanted timing pattern on a pin, which is just looped back to another pin used as a trigger for the second DMA)
i wonder if you could do gamma correction on the device side by adjusting the timing of when you check each bitplane and write the LED state. for example, instead of checking at inverse power-of-two intervals, check at inverse power-of-three intervals.
dunno if anyone brought this up... but brightness (as perceived by the eye) is logarithmic. maybe if you vary the PWM width to make it non-linear it would help with the minimum brightness?
LOL. Hi Luni. Its quiet funny that I also currently have a project with BAM instead of PWM to drive some LEDs on a branded merchandise project of our company. Might be able to send you one since I owe you something from the Keysight oscilloscope I got from you :D
Sure, but throwing more power at something eliminates learning opportunities. The clever PWM scheme being used here is a result of having limits, and learning to work within them. Having cheap and easy resources results in lazy programming, because you never end up limiting yourself to what you NEED, instead of just using what you HAVE.
That's a slick tileable RGB matrix! Great video, well presented and a very cool project! And that dimming technique was brilliant! Thanks for sharing! (Now I feel kinda stupid for not thinking about that! Might have to update my RGB PWM function now). From reading a few comments below, apparently its called Binary Code Modulation (BCM), or upon further reading is the same as BAM (Bit Angle Modulation). But it might have a flicker problem in transitioning around the most significant bit value (and also other transitions, but less noticeable, also depending on FPS I would guess). But it is solved by mirroring it - MIBAM - Mirror Image Bit Angle Modulation. Unless you used MIBAM, did you experience any flicker problems with fading or transitions with your implementation of this? Or maybe your update frequency is so high it doesn't matter anyway. A few years ago I made a timer based RGB PWM function for an Arduino thing I made, that I later adopted to a really small 3x3x3 RGB cube. I remember I was happy that I got 52.5 Hz with 128 levels on the 27 RGB LED's (3x3x3) on the Arduino. I couldn't use the hardware PWM outputs, it's flipping bits for a 3 by 9 2D matrix, times 3 for RGB, then packed together to be a 3D cube of 3x3x3. Basically just as a test, directly driven from the Arduino, no buffers or anything. But now I'm wondering if I could maybe improve that by a lot, I have to think a little bit about it first. I also never implemented any gamma correction, which I also wanted to do at some point. Maybe by using BCM, or rather MIBAM the PWM resolution could be increased for a smooth, gamma corrected output. (Actuallly I was pretty happy with it, but now after seeing this video I'm not so happy with it 😂 Jk, but there's always room for improvement)
Having the LED bulbs closer together to make a higher resolution display would be preferable, I saw the video you made some years ago to make a video display, I was hoping you would have improved upon that project.
Is there a video about that project with the magnetic pogo pin connector coming soon? I am interested which parts that is exactly. I found some on Aliexpress but they are quite expensive and they don't show its a SMD or through hole component.
8:58 And this is almost exactly how DMDs (DLP chips) work. ;) Although, DMDs have a feature where you can "Reset" the positions of all mirrors at once, so you can shorten the display duration for each bitplane. At the same time, you can be shifting in the pixel data for the next bitplane. tesla500 made an interesting vid on DLP tech, as did Ben at Applied Science.
Wait, can you try put a fresnell lens in front of such an array of rgbled so as to make a smaller back projection onto white canvas like 2-3in away? Can you make like an rgbled plasma tv?
"woah, that's a high refresh rate, i wonder how much it would cost to make a HD display with those" 12 * ( 1080 / 8 ) * ( 1920 / 8) = ~$388,000 "nevermind"
LED driver nerd here! What you're referring to as "frames per second" here is really the PWM cycle time, which is a separate measurement to frame rate (the rate at which the colours can be updated). With a 400kHz timing resolution, that 12-bit depth is running at an effective PWM frequency of just 98Hz. This puts you well into the increased flicker risk zone as per IEEE1789, meaning you're going to get significant stroboscopic effects that may be quite unpleasant to viewers, and would potentially pose a small risk to people with photosensitive epilepsy (PSE). This is slightly alleviated by the population count approach (multiple shorter pulses per cycle instead of a single on/off duty per cycle) but the worst-case flicker remains identical to plain PWM. At 8-bit depth and 1.5kHz you're back in the low flicker risk zone, so that's much better, although with just 8 bits of depth and no dithering your perceptual dimming ratio only ends up being 20:1 (based on sRGB EOTF) due to the nonlinearity between radiant intensity and perceptual intensity. If you decouple your update rate from the PWM frequency and implement temporal dithering, you can get 10-bit depth (8-bit linear, 2-bit dither) for a 79:1 dimming ratio, with the lower bound of the dither ripple frequency being 390Hz (and therefore the maximum FPS also being 390), which is ok. I built a whole calculator for this stuff; if you search "LED PWM calculator" it should come up as the top result.
Getting high dimming ratios out of linear PWM without stroboscopic effects is hard work. To reach 256 perceptually linear steps without dithering you need about 12 bits of PWM. But to fully avoid stroboscopic effects, you also need 3kHz or more of PWM frequency. That leaves you needing a timing resolution equivalent to 12.3MHz. That means your driver rise/fall times need to be on the order of 20 nanoseconds, which means about 1A/us of dI/dt per 20mA LED, so you start to have to consider parasitic inductance in your power delivery network and both conducted and radiative EMI in the HF bands. A 2L board won't cut it, you'll need 4L with an SGGS stackup and good layout. It's quite the challenge!
As a quick addendum, if you want to get into the really high end of stuff, look up shunt dimming with adjustable constant-current switching regulators. Dimming is achieved by a combination of adjusting the regulator's output current and shunting the LED current by placing a MOSFET in parallel with the LED chain. When the MOSFET is switched on, the current flows through it instead of the LEDs, turning them off. If you drive the MOSFET with PWM, you get a dimming effect. The MOSFET and its drive circuit are selected to produce a similar impedance to the LED chain, so that the inductor current in the switching converter remains roughly constant regardless of whether the LEDs are being shunted or not, thus allowing for very low ripple current and tight forward current regulation from the buck (while also minimising EMI). A fast trench FET (or, more recently, GaN) is typically used, sometimes with an additional protection diode if the LED chain is long enough to have appreciable inductance. Since the two dimming controls are multiplicative, you can achieve extremely good dimming ratios. It takes a lot of engineering work to get right though.
Super interesting comment! Thank you!
My dude was not joking about being a nerd and i envy
Isn't it also the effective frame rate though?
He is addressing all the pixels at once, so 1 PWM cycle = 1 frame.
awesome comment
1) It's reassuring that even a professional can make silly mistakes
2) It's *incredibly valuable* (especially to someone like myself that's never made a PCB) what some of the "gotchas" could be in making one and that you've documented it 👍
Optical links might be more stable for data connections. That's what we do for the majority of data in our slip rings. Two chunky brushes for ground and +300V, and all data goes through a fibre connection which is free to rotate in the centre.
I'd be curious to find out more what 'chunky brushes' are :)
@@spherarec I imagine a brush, but big and chunky! A brush is something they often use in motors, hence "brushless" motors. A brush is a way of passing electricity, which could be power, or a signal, to something that's rotating. You have a metal ring attached to the rotating part, and a "brush" attached to the static part. The brush can be a copper contact, that rubs against the rotating ring, so is in constant contact, to pass the electricity.
Old electric drills, mains not battery, are a good example. They use spring-loaded blocks of carbon for the brushes. These push against the rotating copper rings of the motor to pass power. The carbon, graphite, wears away over time, so you can undo a screw and replace them with new ones. Graphite is nice because it's smooth, soft, and essentially self-lubricating. It starts off with a square edge but soon wears away into a concave shape, conforming properly to the brushes.
Sometimes you can see blue sparks through the vent holes in a drill, that's the brushes making not-quite-perfect contact. Might happen more as the brushes wear away and copper starts pushing against copper (which damages it, so stop!). Old electric drills are great. They're almost indestructible, and the replacable carbon brushes mean they can last forever. Don't make 'em like that any more!
@@spherarecbrushes are often made using some type of carbon material. It gets very dusty though. This is how generator rotors are excited
What about clock springs?
Seeing your mistakes right off the bat makes me feel a lot better about the silly mistakes I make while working on projects. I tend to beat myself up over them even if I know that it isn't fair to myself to do so
I love your solution! I wrote Windows automation many years ago, and wrote a ChatSleep() function. It would take a minimum time to wait -- so it doesn't hammer on the CPU, like your solution -- and an interval, and function to call to check a state, which it would then start checking. Each time through it would double the interval. Something of the reverse of yours, but the use case was different: it was to catch the next state change, so the automation could proceed, so doing it as fast as possible was a goal. Neat to see a reflection of it! Thank you!
These remind me of an old electronic game i had as a kid, Scrabble Flash. They were little tiles with mono LCDs and a button that would display a letter and then communicate either wirelessly or via IR to tell if they were in the right order to be a word. They would also function as a multi display menu for the 3 different games.
Oh the memories.
I did something similar to that a few years ago, not with pogo pins and magnets, just pin headers and jumpers, and programmed a game of life for it. The total playing field would configure itself depending on the placement of adjacent matrices. Not at runtime, but I think that would be a trivial addition to the software.
0:10 The internet has rotted my brain. Even at 2k frame rate, I instantly recognized our good 'ol pal Rick Astley.
It took me so long I thought it was just a moving blob waaaaaa
11:50 I wonder if you could display an image at a higher resolution for a camera this way, since you can basically control each line at image capture. Or even include secret messages only for video cameras, not visible to the naked eye.
This is super inspiring. I've been imagining how a magnetic LED tile would work and this is almost exactly how I would have done it. I'm excited to see how much of a difference the magnetic connectors will make compared to the pogo pins.
Brute force versus elegance, that's how I see the difference between methods.
I love your videos.
I have been looking for a high-side driver for years... YEARS!!! THANKYOU for putting me out of my misery!!! ❤❤❤😭😭😭
Also thankyou for going through the PCB assembly process, I'm only just starting to learn about it and have found it very intimidating, so again, thankyou! ❤
This reminds me of the MBLed project which was done by three french students about a decade ago, they used used similar matricies and used optical transceivers for data. They have a yt channel by the name of "MBLedRose".
3:40 Editing genius, I LOL'd at the sound and the forehead clip. The workaround was also genius!
Nice magnetic LED matrix, really feel excited when seeing the LED matrix work and shining after a series of fix work, thanks for your great sharring, super cool👍👍👍
It's cool to see someone using the pogo pin connectors and separate magnets. My brain has wanted to use these for something at some point so it's really cool to see them in action with such success! Also, I 100% understand the feeling of being judged by parts I bought many years ago... the struggle is all too real.
Nice project! BTW the trick you used at 9:05 is also used in DMD chips.
I have always enjoyed your content, this blows me away at how smart you are! I have a bunch of those matrix units and have not done anything with them for five years now. pls correct the small bugs and make them available, I would buy 10 assembled boards.
You got me into led matrices, recently made a 16x18 led matrix with the ws2812b leds in a nice frame. Playing with that using esp32 atm. I also have one of the led matrices you've shown in this video, now I will definitely also play around with that one.
Never in my life have I ever been so happy to be Rickrolled.
What a great little project! The PWM solution is genius.
You are the first person that managed to rickroll me in a very long time, congrats!
A more reliable way might be to have the tiles slide together with dovetail connections, with springy contacts that are pressed together firmly.
What a cool idea. The execution is clean too. I love this.
Pogo pins with stronger springs or a very low tech method would be to run the contacts on some pure silver kind of like the pencil trick but with a more conductive material. Obviously graphene would be better. But a quick and dirty silver rub might fix the connections .
Electroplating is super easy. You would be surprised at what you can get metal to stick to with a tiny amount of voltage. No fancy requirements. White vinegar and whatever you want to plate with add a few volts at low amperage then add table salt until your item to be plated starts attracting your positively charged coating material. Electroplating might be a fun little side rabbit hole to go down when you need a break from coding. You do end up with a bunch of nickel plated pennies and copper coated quarters in similar fashion to all the plastic fidget type dealios you printed out when you first get a 3D printer. You get to use the coins still at least and they make a good conversation piece. Super easy to do, barely an inconvenience.
this is super super cool. I have a lot of those 8x8 black framed matrices lying around as well and have always had something like this in mind.
well done
HDR systems can use a limited bit depth to be signalled but then you use extra bits to indicate the dynamic range mapping of the scene. This allows you to position the limited bit depth within a window.
That would definitely marketable as an expandable kit.
This is awesome!
I was thinking 12bpp is massive overkill but of course I'd not taken perceptual brightness into account.
I believe the dimming technique is also known as Binary Code Modulation. I recall seeing it used in LED cube projects.
That’s really cool. One thing you could do to ensure better connections between blocks is to use 2 pins for + and 2 for - instead of just 1 each. Then use a data over power scheme with capacitors to separate the a/c from d/c. So the left 2 pins could be + and transmit and the right 2 pins could be - and receive. Good luck!
Hey, I just want to say that I appreciate you! You are awesome! Loved this project.
How cool, in my Disco i so often think of something like that for behind the DJ, so it could be easily transported and adapted to different Rooms, especially with sound controlled Light would be nice!
I love this! I really like how the display adapts automatically as you attach multiple units, I'm trying to think of a way that it could be made into a fun childs toy or game, maybe you could create Tetris on it but have the playing board adapt to the layout of the modules? This is the only idea I can think of right now
Also trying to think how you could integrate some controls into it, maybe create a module with some buttons or a joystick?
I like the idea of no controls, except rearranging the modules. Maybe not Tetris but some puzzle game, or even something where a snake goes off the edge of one screen, and you have others that can divert him and change his direction, maybe you have to get him home.
At the simplest, you could even juggle balls. Adding motion sensors would be nice too so you can tilt the displays and have them recognise that. With enough screens you could do all sorts of stuff! Possibly each screen can change between several options, to save the need to have so many, but then that needs a control method. You could use tilt, you turn the screen around to change between configurations. That might be part of a static puzzle, or a moving action game.
Yeah these'd be a hell of a toy! And I think they've got a lot cheaper since advertising companies started making billboards out hundreds of them.
PATENT PENDING! PATENT PENDING! PATENT PENDING!
Oh of course, Pipemania would be the natural game for these! But you'd need, what, 30 or so?
Would also be nice, while I'm spending imaginary money, to use a controller with Bluetooth or just local RF. So at the start of the game, one block could assert itself as the controller, the boss. And it could tell each other screen what pipe shape to turn into.
It'd be better than a normal video game since it involves real objects, even if they're just screens for virtual objects. Instead of a joystick, you're physically connecting parts together. Get kids away from the screen while still letting them play video games. Give them some tactility, practice their fine motor control for the other 8 fingers that aren't stuck to a tiny thumbstick.
Now I realise, they'd each need a battery on board. Oh dear. BOM went right up! Wonder how cheap it could be made? If you put 30 in each box, you'd be up to the 10,000 quantity just to make 300 units. You might just make a prototype and take it round toy companies, you'd need their support anyway unless you wanna sell it on Etsy to a tiny fraction of customers.
Oh, and if you could integrate it into one of those programming platforms for kids, like the MIT one, you'd be able to offer it as something to learn to program. Of course 98% of kids wouldn't do it, just like we got our 8-bit computers by lying about doing homework on them! Then again 100% of kids would only be 50% of the customers, as adult nerds try figure out how to hack them.
you'll get much better color resolution is you use PDM (Pulse-density modulation) instead of PWM. the effective bit resolution approaches infinity amortized over time, even for low actual refresh rates. the only limit is the size of the error calculation, which can trivially be 32-bits per channel.
What a dope video! Thanks for sharing!
you shouldn't try to push pins against pins, if you can avoid that. you'll want the data pins pushing onto pads. so make the bottom and left side be pins, and the top and right side be pads. that'd make that work a lot more reliably I bet.
That is what he is doing.
He even draws the diagram out at 13:11 showing pins>pads
@@Prophes0r d'oh i missed that, somehow.
@@naikrovek no worries. Good impulse.
I've seen it done wrong and been very confused before too.
I love this , can be used to display pixel art , would be cool if you give every led their own room with like milk glass above it. You will get a very square pixel. Would love to give you some pixel animations I made , they're mostly 16x16 or 33x32
2k refresh rate is bonkers, cool project
Awesome as usual , hope to see you on the live stream on wednesday !
Ha! I have a box full of LED matrix displays that I bought 20 years ago, and they are also still taunting me. They're rectangular, and only red, but enough to make a small poster. I need to get back to it finally! Part of my problem back then with all my ideas was that microproc tech just wasn't ready yet. The clock speeds were just too slow to do anything I wanted.
Now naturally, with that refresh rate and 8x8 grid, you gotta connect 30x30 of these to make an NES resolution display
Been thinking about making something like this for ages! Now I don't have to :)
Reminds of the lightpad blocks that Roli used to sell as configurable midi controllers.
"You are typically covered by the engineering team, they will contact you if something is wrong"
Proceeds to receive 5 board with a flipped driver.
The engineering team only tells you if the PCB isn’t manufacturable, not if your schematics or layout is wrong.
Looks really good, what about transmitting power through the magnets itself? You could have more pins reserved for future updates or more bandwidth and the magnets would handle the power throughput just fine making misalignment less of an issue.
You always have clever ideas. I like this!
Interesting solutions for the flipped ic pinout
If you leave the magnets to flip freely, you do not need to think about polarity, they will automatically adjust
What? Then they would allow misaligned connections.
That removes one of the primary reasons for using a magnetic connector.
Can U play doom on it?
Dead but mode! I hadn't heard that term in a very long time!
SysPCB is also a good choice for those who has higher requirement in assembly
Wow, great project
This on the back of a windshield with the no symbol over high beams would be revolutionary. People need to turn off their high beams.
It be interesting to project into etched plastic. Maybe one on each column to add depth?
I think you just need longer pogo pins and a better mechanical mate to get your connection to be more stable. some alignment pins in the 3d print would help a lot
Nice magnetic LED matrix
I love this, reminds me of the GLiP project :3
I was going to do a similar thing myself then i read the comments and i think that's alright. it was either play with some LEDS or some FOC/ESC, and I think it 's time to play with some motors lol
I am here to patiently await for the Kickstarter that will make you a millionaire, as long as you can make them cheaply enough.
Another thing you can do to free up processor time in this kind of a setup, if the MCU supports it, is using timer-triggered [circular] DMA to automatically feed the GPIO from memory without any processor involvement. I've done it on STM32F0x (to drive 8 7-segment displays from one GPIO bank), but I'm not familiar with CH32V, so not sure if it'd work there. Of course, it'll still use a lot of memory bandwidth if you're running at ludicrous speed for bit depth reasons, so the bit planes + variable-duration timer might still work better in this case. Of course, if you can figure out a way to drive the DMA with an automated variable-duration timer, you could have the benefits of both. (Not sure off the top of my head if you could update the timer configuration with another DMA - or use another DMA-driven GPIO to output the wanted timing pattern on a pin, which is just looped back to another pin used as a trigger for the second DMA)
i wonder if you could do gamma correction on the device side by adjusting the timing of when you check each bitplane and write the LED state. for example, instead of checking at inverse power-of-two intervals, check at inverse power-of-three intervals.
dunno if anyone brought this up... but brightness (as perceived by the eye) is logarithmic. maybe if you vary the PWM width to make it non-linear it would help with the minimum brightness?
LOL. Hi Luni. Its quiet funny that I also currently have a project with BAM instead of PWM to drive some LEDs on a branded merchandise project of our company. Might be able to send you one since I owe you something from the Keysight oscilloscope I got from you :D
@atomic14 would be proud of your pcb fix.
Dual core would've been ideal.
One core to take care of the LEDs, the other to run the main program.
Sure, but throwing more power at something eliminates learning opportunities.
The clever PWM scheme being used here is a result of having limits, and learning to work within them.
Having cheap and easy resources results in lazy programming, because you never end up limiting yourself to what you NEED, instead of just using what you HAVE.
@@Prophes0rSherlock! is that you? Or is it Captain obvious again?
I can't tell you guys apart anymore!!
@@piconano Hey, some times we still need to state the obvious.
Hearing it out loud, or reading it in text, reinforces it.
0:19 the rickroll just got me fr💀
That's a slick tileable RGB matrix! Great video, well presented and a very cool project!
And that dimming technique was brilliant! Thanks for sharing! (Now I feel kinda stupid for not thinking about that! Might have to update my RGB PWM function now). From reading a few comments below, apparently its called Binary Code Modulation (BCM), or upon further reading is the same as BAM (Bit Angle Modulation). But it might have a flicker problem in transitioning around the most significant bit value (and also other transitions, but less noticeable, also depending on FPS I would guess). But it is solved by mirroring it - MIBAM - Mirror Image Bit Angle Modulation. Unless you used MIBAM, did you experience any flicker problems with fading or transitions with your implementation of this? Or maybe your update frequency is so high it doesn't matter anyway.
A few years ago I made a timer based RGB PWM function for an Arduino thing I made, that I later adopted to a really small 3x3x3 RGB cube. I remember I was happy that I got 52.5 Hz with 128 levels on the 27 RGB LED's (3x3x3) on the Arduino. I couldn't use the hardware PWM outputs, it's flipping bits for a 3 by 9 2D matrix, times 3 for RGB, then packed together to be a 3D cube of 3x3x3. Basically just as a test, directly driven from the Arduino, no buffers or anything. But now I'm wondering if I could maybe improve that by a lot, I have to think a little bit about it first. I also never implemented any gamma correction, which I also wanted to do at some point. Maybe by using BCM, or rather MIBAM the PWM resolution could be increased for a smooth, gamma corrected output. (Actuallly I was pretty happy with it, but now after seeing this video I'm not so happy with it 😂 Jk, but there's always room for improvement)
[14:55] Killed me
Not enough of this in the world
Maybe some guiding pins and holes may help or more magnets (in verical plane) ?
woah its been a while since ive seen eclipse!
Using stm32 hardware pwm mode on timers, you could get up to 1098Hz with 16bit resolution, directly on the pins, without any interrupts
Reminds me of the LEDSEE and colourduino stuff
duuuude these are sick.
Whoa long time no see
an rgb matrix module for arduino boards would be nice
Having the LED bulbs closer together to make a higher resolution display would be preferable, I saw the video you made some years ago to make a video display, I was hoping you would have improved upon that project.
Is there a video about that project with the magnetic pogo pin connector coming soon?
I am interested which parts that is exactly. I found some on Aliexpress but they are quite expensive and they don't show its a SMD or through hole component.
8:58 And this is almost exactly how DMDs (DLP chips) work. ;)
Although, DMDs have a feature where you can "Reset" the positions of all mirrors at once, so you can shorten the display duration for each bitplane.
At the same time, you can be shifting in the pixel data for the next bitplane.
tesla500 made an interesting vid on DLP tech, as did Ben at Applied Science.
I love the RGB matrix project, btw. Really nicely done.
great !!!! You can try tled !!!
That looks very similar to the Roli Lightpad midi device, wonder if it's possible to recreate it
11:57 Wondering if there is a way to exploit that effect to transmit data in a way that would be "covert" to a human?
Wow - Das sieht aber cool aus. lg aus CH
Your first issue is coding in java. Would explain why it barely functions, despite some more obvious physical connection issues.
I wonder if you can do the gamma correction by changing the timings of the interrupts
1:05 I would just bend the contacts
Wait, can you try put a fresnell lens in front of such an array of rgbled so as to make a smaller back projection onto white canvas like 2-3in away? Can you make like an rgbled plasma tv?
@bitluni, how you managed to make that logo on that bottom silkscreen? 01:56 and 03:23 Like how you made half-tone graphics in GIMP.
Awesome!!!
Rick Asley '" never gone gif you up "
What I wonder, would it be possible to use the Magnets as a connection? Are they conductive?
"woah, that's a high refresh rate, i wonder how much it would cost to make a HD display with those"
12 * ( 1080 / 8 ) * ( 1920 / 8) = ~$388,000
"nevermind"
is this a sufficient frame rate for volumetric persistence of vision display? I think yass
I think you need to calibrate your dso probe.
Scale it up to an entire wall :)
What is that hot plate you used at the beginning of your video to fix the flipped IC?
Please have a look at "Light Stax Junior Classic Illuminated Blocks" - this would be a mind blowing toy uisng animated LEDs - what a toy !
just reminds me i need to work on my daft punk helmet's XD
Link to pogopins component please? Have been looking for a while for something like that
Which CAD tool are you using?