This kind of limited device is something I've been thinking about with respect to what interactions I want my children to have with computers. I remember when I was 9 years old and we got these computers at the lab at school and we wrote some LOGO and BASIC and it was a mind-blowing experience. We were drawing SQUARES! And we were making TRIANGLES of ASTERISKS! Hahaha, what a glorious thing that felt like.
I got so much joy from computers and I'd like my kids to have that kind of experience too without accidentally detouring into social media (which has my mind in a vice grip).
Still a couple of years away, but I think I'd like to evaluate this kind of device then and see if it's the right model to use.
> This kind of limited device is something I've been thinking about with respect to what interactions I want my children to have with computers. I remember when I was 9 years old and we got these computers at the lab at school and we wrote some LOGO and BASIC and it was a mind-blowing experience. We were drawing SQUARES! And we were making TRIANGLES of ASTERISKS! Hahaha, what a glorious thing that felt like.
Well, Minecraft 'redstone' works a bit like that?
Of course, it's embedded in a much bigger program, but I'm not sure that makes a difference to the kids?
I don't think comparing redstone to something like BASIC is fair. Redstone is easy to get started with but actually making something interesting with it is significantly more complicated. Minecraft Education edition is a better example where you can use Python or something like Scratch to interact with the game.
I recall those days as well, but I think it's much harder to replicate today. In that era, computers were new and rare. Any interaction with a computer was notable, and then the ability yourself to not just interact, but actually CONTROL one was amazing.
Today, kids are surrounded by all kinds of tech. They see people interacting with tech in all kinds of ways from the moment they are cognizant. It's much harder to create that wow moment now.
The Basic variant on a TI-84 was magic. One science teacher at my high school allowed us to use any program as long as we wrote it ourselves. I remember making programs with nested menus for all types of physics algorithms. Through debugging, I ended up memorizing them anyway, but it was more fun (and error-free) to use the program itself.
I'm not a parent, but I think encouraging anything creative is probably just as good.
Half the fun was not knowing how to do something. There was no other way to satisfy curiosity than to tinker endlessly and constantly seek out information. Stumbling upon unusually good programs made it seem like anything was possible regardless of the machine it ran on. Video games and the demoscene were like that for me, and now any modern machine really can run almost anything.
Programming can still be fun like that, but often in the context of existing ideas. My parents had similar feelings about new music and cars. The sense of wonder decreases when the bar is raised. That's not to say there isn't a ton left to explore, but that's the impression when curiosity is too easily satisfied. You have to keep up and find new ways to stay curious. We consume way more than we create these days.
The New Yorker had a memorable single panel comic by David Sipress with an old man saying "everything was better back when everything was worse". I just had to mention it.
I remember a large part of the fun was that we couldn't just look something up on the web - it didn't exist in our home in the eighties.
Instead we'd pore over BASIC on floppies. Changing a thing, for example GRAVITY=0.1, and finding out the banana now flies almost straight up.
Or meticulously typing over the source code printed in hand-me-down magazines. Evenings of me and friend one typing the other reading out loud. Then way more evenings of finding all the typos and bugs. And then one or two evenings running the game we just "wrote" and get bored immediately. And start changing things.
This is how I learned programming and what has paid my bills for over 25 years now. There were few university careers for programming, but mostly, young-arrogant-me was like "well, I have learned myself programming, so I'd better follow a university program that teaches me stuff I don't already know XD".
The tinkering and creative part has been lost by now for me. I lament that. So I've put aside a fund, finishing off some contracts now and from this summer on, will do unpaid work of "creative coding". Making "art" with software - something I now do in spare time, fulltime. Because that tinkering is what drew me in. Not the scrum-rituals, spagetti-code-wrangling or layers of architectural enterprise abstractions. But the fun of nesting nested loops and seeing my name fly over the screen in weird patterns, or the joy of making the matrix printer play a "song" by letting it print weird ascii strings.
When I was kid thinking about this as a career, I knew what I was getting into. I had the internet (when it was a lot smaller). I had seen big tech flop hard. I saw companies like Apple for what they were before and after their iDevices. By the time I was wrapping up my CS degree I had seen social media destroy itself and legacy media, the rise of web 2.0, SaaS startups, mobile apps, etc.
I've been working professionally now for over a decade, but got started long before that as a child. Despite the endless negative things I could say about the modern era, I don't feel like any of it impacts my enjoyment of my work or gets in the way of my creativity.
I think this is because the closest I've ever been to truly being alone with the machine is writing programs for my TI calculators, but even then I still had ticalc.org. Some programs on there were brilliant, but most were awful. It was the perfect balance for people my age at the time. Despite what people believe today, especially with their LLMs, I don't think the landscape has changed much in that regard. There's still a lot of awful code with few brilliant examples. That leaves room for me to work on new interesting stuff or improve what's there without having too much help spoiling it.
I've been using it for a few months. Great project. I especially love adding i2c peripherals from M5. E.g. a bank of 8 rotary encoders.
Also love how absolutely minimal it is in size and if you didnt notice, the screen is a touchscreen. And they have a basic set of ui widgets.
Also interesting, the gfx lets you overlap sprites, bitmap, and text mode. You can tell the designers have lot of XP on 8-bit systems. And the bitmap is a little larger than the screen so you can do some superbitmap stuff. It's bot terribly larger, just a bit.
I havent been using it as much for its synth capabilites, ironically, but for making sequencers for external instruments. I believe it also has audio in...
What I love about this is the reduction in complexity compared to how something like this would typically be built today.
If I were to build a synth a year ago I probably would have used Rust compiled to WASM and running in the browser. This thing has a lot of the same functionality, but you have about -30 million lines of code for the OS, -30MLOC for the browser, and another -30MLOC for Rust/LLVM.
And that doesn't even get in to the cost of materials or power savings.
Obviously it's not apples to apples but it really makes me wonder how much of that stack we need for most programs.
I agree AI is interesting here. It raises the level of abstraction in a similar way to the OS/Browser/language, but it does so by depending on a lot of data, as opposed to depending on a lot of code.
There’s a sort of graph isomorphism problem of mapping APIs onto each other that seems solvable since a lot of them do the same thing but in different ways. Though it’d take something more keen on the minutiae than the LLMs for this I think
Yes, I got a tour of their factory back in the day when I was editor for a number of IT-related magazines. Close to everything was made there in that factory from the metal housing for the machines to the circuit boards - photoresist, exposure, etching, cleaning, printing, conformal coating, through the pick-and-place machine, through the wave solder bath, testing and mounting in the chassis. In the Netherlands, in a relatively modest factory hall. If it could work then - and it did, for a while - it should be possible to do that now without the compulsive urge to outsource everything.
I expected this thread to be about a vintage computer from them when I clicked.
I'm pretty sure I had seen some promotional material of theirs the last time I was in NL, so I didn't know they had gone out of business in 2008/2009 already.
The thing about trademarks is that, if you want to prevent other people from using them, you generally have to still be using it yourself and be able/willing to justify to a court that you're still using it. (At least in most legal systems that I'm familiar with)
Since the original company both changed names and was subsequently liquidated in bankruptcy nearly 20 years ago... that seems unlikely. There's only so many names out there, and occasionally they get fairly recycled.
That it's built off Micropython is a huge deal. I've started using it by default for my hardware projects and it makes everything easier - writing drivers, playing with user interfaces, etc. Loads of regular Python libraries work and I can even grab them over the Internet. Like how I imagine it was running Forth or smth on embedded kits back in the day (ah maybe not the networking bit)
Looks interesting. I'm interested in the T-Deck Tulip CC and would love to use it for coding whilst im traveling. Any experience with using such a device for light programming?
If you're staying in python or another dynamic language it could probably work. Unfortunately I don't think there are a lot of native compilers that run on esp32s, though there are some[0]
I have used the T-Deck with tulipcc for coding and writing, although these days I mostly use it as a calculator. I wrote a GUI text editor for it, which you can find here:
I preordered the hardware version sometime before it was released. At a $50 price point, I thought it was worth a gamble. I wasn't disappointed! It's more of a hacker's device at the moment - for better or worse - but it's fairly powerful under the hood. AMY is a very impressive synth. And the device is clearly a labor of love from the creator.
Why? That's a decent amount of power for running one thing. It's also dual core! You can emulate early consoles on this microcontroller. Someone made a 3D game for the similar (but less performant) RP2040 running at 250 MHz -> https://github.com/bernhardstrobl/Pico3D
It's seems like the Tulip could definitely be used for something like that, though you might have to write quite a bit of your own framework code in python.
Seems a little too technical for me but I am really curious. Seems pretty interesting. My brother is into music and is looking forward to get started with music production as well. And he himself is a developer. He might like this. Let's see.
> You can use Tulip to make music, code, art, games, or just write.
Am I wrong to think statements like these are just aspirational warm-and-fuzzies about the product without any real substance?
You could do all those things on anything, but they are typically incongruent with one another. If you are a beginner or a pro, you’re going to be better off doing it on a “more-standard” device.
* I can read and understand every line of code that is running
* I can understand all of the hardware it's running on
We've gotten so used to computers not working. Weird stuff breaks all the time and even experts can only guess why beyond turning it off and on again, which takes minutes for most devices.
I dream of a world where we trust computers to work and be fast. It's completely possible, but step one is reducing complexity by several orders of magnitude.
Agree with the above. As someone who has never heard of this before, the description of "a portable programmable device for music, graphics, code and writing" reads to me as "a computer". I'm kind of unsure why I would want to use this instead of the computer I'm typing on right now.
This seems to be targeting the market of users with the following intersecting interests:
* DIY hardware enthusiast
* musician
* python developer
* maybe also wants graphics...?
Seems a small segment to me, but I assume I'm missing something here.
An immediate benefit I see is that they're cheap enough to use once - you could make/find/buy a software instrument that you like, then put it in your gear bag and never reflash it. Now it's just like any other synth. Then you can get a second Tulip and do the same thing later if you like. You could do this with laptops of course but it starts to get expensive.
The Pocket Operators have something similar (the KO at least, maybe the others). If you've written samples into them you want to preserve for playing live, you can snap a tab off and then they're read-only - no surprises on gig night.
I'm particularly sensitive to shallow critiques of new ways of computing, particularly those that encourage and enable people to be creative. Whether a project is successful or not, it's nice to see something that isn't a "bootup your general purpose comouter and then immediately open a browser" style of computing.
Attempting to get people to interact with the real world and also be creative should be commended.
It’s the “You can do anything (… at Zombo.com [1])” angle.
I’ve been lulled into novel only-limit-is-your-imagination work environments that try to convince me to think they will be “transcendental” in my abilities.
A little while later, I run into software or hardware limitations only to face a physical malaise because I’d been troubleshooting a hardware or software problem for hours.
Take the “just write” angle. I do technical writing. I don’t trust that the built-in dictionary for that device is ever going to meet my spellcheck needs. I need mah Microsoft Word that I know how to navigate like the back of my hand, and I’m set.
Don’t promise a “Zombocom” device that doesn’t actually deliver.
This kind of limited device is something I've been thinking about with respect to what interactions I want my children to have with computers. I remember when I was 9 years old and we got these computers at the lab at school and we wrote some LOGO and BASIC and it was a mind-blowing experience. We were drawing SQUARES! And we were making TRIANGLES of ASTERISKS! Hahaha, what a glorious thing that felt like.
I got so much joy from computers and I'd like my kids to have that kind of experience too without accidentally detouring into social media (which has my mind in a vice grip).
Still a couple of years away, but I think I'd like to evaluate this kind of device then and see if it's the right model to use.
> This kind of limited device is something I've been thinking about with respect to what interactions I want my children to have with computers. I remember when I was 9 years old and we got these computers at the lab at school and we wrote some LOGO and BASIC and it was a mind-blowing experience. We were drawing SQUARES! And we were making TRIANGLES of ASTERISKS! Hahaha, what a glorious thing that felt like.
Well, Minecraft 'redstone' works a bit like that?
Of course, it's embedded in a much bigger program, but I'm not sure that makes a difference to the kids?
I don't think comparing redstone to something like BASIC is fair. Redstone is easy to get started with but actually making something interesting with it is significantly more complicated. Minecraft Education edition is a better example where you can use Python or something like Scratch to interact with the game.
Perhaps redstone is a bad example, I don't play Mindcraft [sic].
I recall those days as well, but I think it's much harder to replicate today. In that era, computers were new and rare. Any interaction with a computer was notable, and then the ability yourself to not just interact, but actually CONTROL one was amazing.
Today, kids are surrounded by all kinds of tech. They see people interacting with tech in all kinds of ways from the moment they are cognizant. It's much harder to create that wow moment now.
The Basic variant on a TI-84 was magic. One science teacher at my high school allowed us to use any program as long as we wrote it ourselves. I remember making programs with nested menus for all types of physics algorithms. Through debugging, I ended up memorizing them anyway, but it was more fun (and error-free) to use the program itself.
I am probably going to give my toddler my Callisto 2 https://www.thingiverse.com/thing:4846997
when he is old enough to get some joy from it.
Haha, this is incredible. I love it. Many of the parts will need to be substituted since they are no longer stocked but what a concept. Oh my.
I'm not a parent, but I think encouraging anything creative is probably just as good.
Half the fun was not knowing how to do something. There was no other way to satisfy curiosity than to tinker endlessly and constantly seek out information. Stumbling upon unusually good programs made it seem like anything was possible regardless of the machine it ran on. Video games and the demoscene were like that for me, and now any modern machine really can run almost anything.
Programming can still be fun like that, but often in the context of existing ideas. My parents had similar feelings about new music and cars. The sense of wonder decreases when the bar is raised. That's not to say there isn't a ton left to explore, but that's the impression when curiosity is too easily satisfied. You have to keep up and find new ways to stay curious. We consume way more than we create these days.
The New Yorker had a memorable single panel comic by David Sipress with an old man saying "everything was better back when everything was worse". I just had to mention it.
Same here.
I remember a large part of the fun was that we couldn't just look something up on the web - it didn't exist in our home in the eighties. Instead we'd pore over BASIC on floppies. Changing a thing, for example GRAVITY=0.1, and finding out the banana now flies almost straight up.
Or meticulously typing over the source code printed in hand-me-down magazines. Evenings of me and friend one typing the other reading out loud. Then way more evenings of finding all the typos and bugs. And then one or two evenings running the game we just "wrote" and get bored immediately. And start changing things.
This is how I learned programming and what has paid my bills for over 25 years now. There were few university careers for programming, but mostly, young-arrogant-me was like "well, I have learned myself programming, so I'd better follow a university program that teaches me stuff I don't already know XD".
The tinkering and creative part has been lost by now for me. I lament that. So I've put aside a fund, finishing off some contracts now and from this summer on, will do unpaid work of "creative coding". Making "art" with software - something I now do in spare time, fulltime. Because that tinkering is what drew me in. Not the scrum-rituals, spagetti-code-wrangling or layers of architectural enterprise abstractions. But the fun of nesting nested loops and seeing my name fly over the screen in weird patterns, or the joy of making the matrix printer play a "song" by letting it print weird ascii strings.
When I was kid thinking about this as a career, I knew what I was getting into. I had the internet (when it was a lot smaller). I had seen big tech flop hard. I saw companies like Apple for what they were before and after their iDevices. By the time I was wrapping up my CS degree I had seen social media destroy itself and legacy media, the rise of web 2.0, SaaS startups, mobile apps, etc.
I've been working professionally now for over a decade, but got started long before that as a child. Despite the endless negative things I could say about the modern era, I don't feel like any of it impacts my enjoyment of my work or gets in the way of my creativity.
I think this is because the closest I've ever been to truly being alone with the machine is writing programs for my TI calculators, but even then I still had ticalc.org. Some programs on there were brilliant, but most were awful. It was the perfect balance for people my age at the time. Despite what people believe today, especially with their LLMs, I don't think the landscape has changed much in that regard. There's still a lot of awful code with few brilliant examples. That leaves room for me to work on new interesting stuff or improve what's there without having too much help spoiling it.
I've been using it for a few months. Great project. I especially love adding i2c peripherals from M5. E.g. a bank of 8 rotary encoders.
Also love how absolutely minimal it is in size and if you didnt notice, the screen is a touchscreen. And they have a basic set of ui widgets.
Also interesting, the gfx lets you overlap sprites, bitmap, and text mode. You can tell the designers have lot of XP on 8-bit systems. And the bitmap is a little larger than the screen so you can do some superbitmap stuff. It's bot terribly larger, just a bit.
I havent been using it as much for its synth capabilites, ironically, but for making sequencers for external instruments. I believe it also has audio in...
Also the discord is helpful. 10/10
What I love about this is the reduction in complexity compared to how something like this would typically be built today.
If I were to build a synth a year ago I probably would have used Rust compiled to WASM and running in the browser. This thing has a lot of the same functionality, but you have about -30 million lines of code for the OS, -30MLOC for the browser, and another -30MLOC for Rust/LLVM.
And that doesn't even get in to the cost of materials or power savings.
Obviously it's not apples to apples but it really makes me wonder how much of that stack we need for most programs.
It's a trade off as always. I agree though.
I wonder the same thing a lot. I also wonder how AI will fit into this problem.
I agree AI is interesting here. It raises the level of abstraction in a similar way to the OS/Browser/language, but it does so by depending on a lot of data, as opposed to depending on a lot of code.
The cost of abstraction is always dependencies.
There’s a sort of graph isomorphism problem of mapping APIs onto each other that seems solvable since a lot of them do the same thing but in different ways. Though it’d take something more keen on the minutiae than the LLMs for this I think
Funny! In the '80s Tulip Computers NV[1] was a Dutch computer manufacturer that manufactured PC clones.
[1] https://en.wikipedia.org/wiki/Tulip_Computers
It was a great innovative company in the Netherlands. They designed and manufactured everything themselves. Hardware boards and software. See https://www.homecomputermuseum.nl/collectie/tulip/?srsltid=A...
Yes, I got a tour of their factory back in the day when I was editor for a number of IT-related magazines. Close to everything was made there in that factory from the metal housing for the machines to the circuit boards - photoresist, exposure, etching, cleaning, printing, conformal coating, through the pick-and-place machine, through the wave solder bath, testing and mounting in the chassis. In the Netherlands, in a relatively modest factory hall. If it could work then - and it did, for a while - it should be possible to do that now without the compulsive urge to outsource everything.
Wow, that’s very cool. Was there an ecosystem in NL for this sort of company at the time, and is this where ASML came from / has its roots?
No, ASML was spun out of Philips (as was e.g. NXP)
I expected this thread to be about a vintage computer from them when I clicked.
I'm pretty sure I had seen some promotional material of theirs the last time I was in NL, so I didn't know they had gone out of business in 2008/2009 already.
Maybe the trademark is still owned by someone (?)
The thing about trademarks is that, if you want to prevent other people from using them, you generally have to still be using it yourself and be able/willing to justify to a court that you're still using it. (At least in most legal systems that I'm familiar with)
Since the original company both changed names and was subsequently liquidated in bankruptcy nearly 20 years ago... that seems unlikely. There's only so many names out there, and occasionally they get fairly recycled.
Was thinking that too.. :)
Cute of you to think that the american developers behind this would care about that.
https://github.com/shorepine/tulipcc/graphs/contributors
I think I'm in love.
That it's built off Micropython is a huge deal. I've started using it by default for my hardware projects and it makes everything easier - writing drivers, playing with user interfaces, etc. Loads of regular Python libraries work and I can even grab them over the Internet. Like how I imagine it was running Forth or smth on embedded kits back in the day (ah maybe not the networking bit)
Looks interesting. I'm interested in the T-Deck Tulip CC and would love to use it for coding whilst im traveling. Any experience with using such a device for light programming?
If you're staying in python or another dynamic language it could probably work. Unfortunately I don't think there are a lot of native compilers that run on esp32s, though there are some[0]
[0]: https://github.com/valdanylchuk/xcc700
I have used the T-Deck with tulipcc for coding and writing, although these days I mostly use it as a calculator. I wrote a GUI text editor for it, which you can find here:
https://github.com/coolcoder613eb/notepad
I haven't done too much coding with it, but I have on occasion fixed bugs in the text editor on the T-Deck itself.
I preordered the hardware version sometime before it was released. At a $50 price point, I thought it was worth a gamble. I wasn't disappointed! It's more of a hacker's device at the moment - for better or worse - but it's fairly powerful under the hood. AMY is a very impressive synth. And the device is clearly a labor of love from the creator.
Given the name I thought it was someone reviving a PC brand.
https://en.wikipedia.org/wiki/Tulip_Computers
Also interested if this supports strudel REPL or tidalcycles. This would be a really awesome device to use for livecoding sets if it does.
That’s a tall order for a 240 MHz CPU.
Why? That's a decent amount of power for running one thing. It's also dual core! You can emulate early consoles on this microcontroller. Someone made a 3D game for the similar (but less performant) RP2040 running at 250 MHz -> https://github.com/bernhardstrobl/Pico3D
Specifically because both Strudel and Tidal Cycles seem to struggle on moderately complex compositions on my ThinkPad X270.
Oh interesting. Thanks!
Old man asks: Does this support what the kids call "livecoding"?
Funny how this has co-opted the Runtime Revolution folks re-naming their Hypercard clone as:
https://livecode.com/
I didn't know about livecoding; thanks for that!
It's seems like the Tulip could definitely be used for something like that, though you might have to write quite a bit of your own framework code in python.
I have one. Bought it in July of 2024.
It is not an instrument for me and I don’t use it. I prefer to interact with music in other idioms.
But I appreciate the passion of its designer.
Seems a little too technical for me but I am really curious. Seems pretty interesting. My brother is into music and is looking forward to get started with music production as well. And he himself is a developer. He might like this. Let's see.
> You can use Tulip to make music, code, art, games, or just write.
Am I wrong to think statements like these are just aspirational warm-and-fuzzies about the product without any real substance?
You could do all those things on anything, but they are typically incongruent with one another. If you are a beginner or a pro, you’re going to be better off doing it on a “more-standard” device.
Here's a few I care about:
* Boots almost instantly into a usable system
* I can read and understand every line of code that is running
* I can understand all of the hardware it's running on
We've gotten so used to computers not working. Weird stuff breaks all the time and even experts can only guess why beyond turning it off and on again, which takes minutes for most devices.
I dream of a world where we trust computers to work and be fast. It's completely possible, but step one is reducing complexity by several orders of magnitude.
Agree with the above. As someone who has never heard of this before, the description of "a portable programmable device for music, graphics, code and writing" reads to me as "a computer". I'm kind of unsure why I would want to use this instead of the computer I'm typing on right now.
This seems to be targeting the market of users with the following intersecting interests: * DIY hardware enthusiast * musician * python developer * maybe also wants graphics...? Seems a small segment to me, but I assume I'm missing something here.
An immediate benefit I see is that they're cheap enough to use once - you could make/find/buy a software instrument that you like, then put it in your gear bag and never reflash it. Now it's just like any other synth. Then you can get a second Tulip and do the same thing later if you like. You could do this with laptops of course but it starts to get expensive.
The Pocket Operators have something similar (the KO at least, maybe the others). If you've written samples into them you want to preserve for playing live, you can snap a tab off and then they're read-only - no surprises on gig night.
Do you have a more substantive critique?
I'm particularly sensitive to shallow critiques of new ways of computing, particularly those that encourage and enable people to be creative. Whether a project is successful or not, it's nice to see something that isn't a "bootup your general purpose comouter and then immediately open a browser" style of computing.
Attempting to get people to interact with the real world and also be creative should be commended.
It’s the “You can do anything (… at Zombo.com [1])” angle.
I’ve been lulled into novel only-limit-is-your-imagination work environments that try to convince me to think they will be “transcendental” in my abilities.
A little while later, I run into software or hardware limitations only to face a physical malaise because I’d been troubleshooting a hardware or software problem for hours.
Take the “just write” angle. I do technical writing. I don’t trust that the built-in dictionary for that device is ever going to meet my spellcheck needs. I need mah Microsoft Word that I know how to navigate like the back of my hand, and I’m set.
Don’t promise a “Zombocom” device that doesn’t actually deliver.
[1] https://zombo.com/
I get the impression that the Atari AMY chip was an inspiration? Wonderful to see how the Alles speakers are implemented!
Super cool!
Go brian!