Hacker News new | past | comments | ask | show | jobs | submit login
Popcorn Computer’s Pocket P.C. Open Sourced (popcorncomputer.com)
126 points by nullagent on Aug 8, 2020 | hide | past | favorite | 62 comments



> Our goal is to ensure that we get all aspects of our products mainlined so that custom kernels will not be required.

I love these kinds of projects, but after being burned by the PocketCHIP, who also said this but never made it there, this is isn't somewhere I want to go again.


The big problem with the C.H.I.P. is that it used raw NAND flash as the main storage (and MLC NAND on the original boards at that), which just wasn't something that could be supported upstream. I think more or less everything else required for upstream support was there a long time ago. This should hopefully avoid that problem; it's using EMMC which doesn't have that problem and it looks like it's based on the Allwinner A64 which already has fairly good mainline Linux support.


Yeah, ditto. Why can't mainline Linux support come first? Then they could sell a working product.

Crazy, I know. Investing in difficult R&D problems to produce a quality product is not a good business model. It's so much easier to churn out a working devkit and ask for pull requests.

Sorry for the snark and pessimism, but that's what happens when you keep getting burned. This statement does not inspire confidence:

>Finally, to ensure that Pocket P.C. has great software support out of the box, we have requested embedded development help from the community.

Why is software always an afterthought with these projects? Especially a project like this, which has clearly seen an enormous amount of effort put into the hardware design.


> Why is software always an afterthought with these projects? Especially a project like this, which has clearly seen an enormous amount of effort put into the hardware design.

This can go either way, in my experience. I’ve worked with software developers who do a terrible job with hardware, and hardware developers who do a terrible job with software. Best I can tell, it’s a problem that’s existed since we started making computers.

In some ways, I think it’s arrogance. The rockstar software developer believes that software is infinitely more complex than the hardware it runs on. The ninja EE believes that software is trivial. And they both assume that they can effectively do each others’ jobs.


The only example I have is the LibreComputer S905X based board - it took years of work with a small team to get the kernel changes upstreamed. Hell, look at the Raspberry Pi - I think mainline support is pretty recent.


This project shows a lot of promise.

> TRYING TO BUILD THE ULTIMATE RASPBERRY PI COMPUTER (ZERO TERMINAL V3)

https://n-o-d-e.net/zeroterminal3.html

https://www.youtube.com/watch?v=wiJqUWfR90I


I suspect if someone were ready to pounce on the Oxide source code when it comes out, that one might make a go of a properly open portable handheld device by adding some drivers and snipping a few bits off. Radios might be the hardest things to sort out, since people seem to like to keep those as proprietary as hell, and WiFi is merely okay now.

Sounds like starlink antennae won’t be small enough for a long time, so that leaves who besides Qualcomm, Broadcom, and Apple?


Mediatek make modems


Oxide?


Brian Cantrell is off trying to build server hardware based on Open firmware. Which likely means they’ll have to write a lot of it themselves.


Looking at the circuit board is reminding me if a thought experiment I entertain from time to time: we keep trying to make the single purpose keyboard smaller so we have room for everything else in a compact device.

What if we didn’t? Instead of removing the volume, can we put something in it? Take this one for instance. How many layers does a keyboard trace need? 2? 3? How many things can you stuff on the backside of the keyboard? Between? Under?


What are some interesting use-cases for this?

I can imagine sys admins would like this for remote admin.

I can't think off the top of my head why general hackers would keep this in their pockets most of the time.


It's got integrated LoRa, so I assume they're targeting a lot of hackers or people working on IOT stuff "in the field", as it were.

Well and I'm not one of those, but I'm still seriously considering getting it - or rather, the cheaper version without LoRa. I can't really tell you anything from the top of my head, but back when I had my N900, it found so many unexpected uses constantly. Having a full-blown Linux machine in your pocket is a seriously different game than any of the current gen of smartphones.


I just want someone to take this form factor[1] (and crucially build quality)

https://en.wikipedia.org/wiki/Psion_Series_5#/media/File:200...

and shove modern hardware/screen into it running stock gnu/linux.


As others have mentioned, Planet Computers "Gemini" is what you're looking for.

Truth be told, its only really usable when running android. BUT:

* its easily rooted (the bootloader is open, and Planet provide a rooted android image)

* with Termux (or a full chroot of your choice), you do get most of the way towards "the linux experience"

Early batches may have been hit or miss. Mine was from the 2nd indiegogo batch, and all I can say is, hardware-wise, its really good.

I've used it HEAVILY over the last couple of years. Dropped it a few times. They keyboard is good. It does have genuine tactile feedback. Given the size limitations, its brilliant.

And yes, I did have more than 1 Psion back in the day (from the calculator-style model, all the way to the famed Psion 5).

And as far as keyboard feel is concerned, on my desktop, I use ergonomic mechanical keyboards (ergodox & atreus62) with various switch types - which means that yes - I am a bit picky when it comes to keyboards.

And I have to say that, given the size limitations, the Gemini keyboard is a marvel. Its genuinely usable. I also have a GPD Pocket. Thats significantly larger, with a chiclet style keyboard.

The Gemini keyboard blows it out of the water.

I have no qualms over opening up termux, and neovim or joe, switching between multiple tabs, buffers and embedded terminals, and banging out some code.

(although I do wish they had used an ortholinear layout for the keyboard, and I wish I could properly remap the layout a la qmk)

As a phone, it has its limitations. But, if you pair it with a smartwatch, many of those limitations can be overcome. Its been my only phone for over a year.

On a final note, nothing beats being able to sit in front of a large screen android-enabled tv, pull out my phone/pda/psion-remake, connect via miracast, and just bang away ....

Its a hoot :)


this may right up your ally! https://www.www3.planetcom.co.uk/gemini-pda

As far as I can tell, it can dualboot linux and android (or even sailfish OS!)


Yes it dual boots, I have one. The issue is that only last week they got the modem to work under Linux and there are still many open issues. I wish they opened their development including Android as there are issues that the community could fix much faster.

As for use cause. I don't know. Nice to have a keyboard but I'm carrying two phones around now which is kind of stupid. The front display it total garbage, slow ui and useless.

Their next model seems more practical as the screen is on the outside allowing use without having to fold open the keyboard which is kind of annoying on the gemini.


What are some useful/fun things to do with an N900 (or Linux with LoRa in your pocket) that a non-jailbroken smartphone like iPhone/Android can't do well?

(genuinely asking for ideas)


I guess you can try librism 5 linux phone, or pinephone.


All I can say about this is 'Raspberry Pi 4 8GB @75$US' and available right now.


what can i do woth it, its twice as expensive as a flipper


The more expensive version supports LoRa:

“Pocket P.C. w/ LoRa packs a LoRa radio capable of low power, long range communication ideal for IoT developers. Build custom LoRa peer-to-peer applications or build apps for the rapidly growing worldwide LoRaWAN community. With access to LoRaWAN at your finger tips you can prototype and build cutting edge applications faster than ever. Ready to operate on frequencies EU868, US915, and AS923.

Pocket P.C. w/ LoRa is the tool for developing LoRa solutions on the go.


It runs Linux, has a keyboard and has a 1080p screen. I doubt the flipper even has a MMU.


Some Flippers have an i.MX6 running Kali Linux, but they do seem like completely different products.

Many more radios, but many fewer buttons and pixels.

It's nice that we have so many options for tiny computers these days, but sooner or later, someone is going to have to find something useful to do with them.

The UI and developer experience has never gotten quite good enough to drive a market the size of smartphones or laptops, so...what are we doing here? Is it time to acknowledge that the masses don't particularly want general-purpose computing machines? That there isn't much of a market for anything that isn't plug-and-play? That's a depressing thought.


> someone is going to have to find something useful to do with them

Looking at that pocket pc website... If that usb c port could drive a display it might be interesting to dock that thing and get a full desktop experience. Then take your work with you when you are ready to go mobile. On buses and trains or other places where it might be clumsy to use a laptop.

I guess this was a thing many people have fantasized about for a long time but nobody has really made practical. I guess laptops, phones and tablets being separate entities is good enough.


Is that going to be a great experience with an Allwinner R8 and 512 mb of RAM? The PocketCHIP had similar specs but a much more low res display, and it chugged under load.


What are you planning to run on it?


What is the flipper? URL?


Some hyphy SBC that's all flash and style. It's got a basic SDR and a few common SPI/i²c devices. It's 99% a marketing play.

All off the shelf and any engineering team could clone it in a few days while substantially undercutting their price. I'd place bets someone already has. The cheaper clones will probably beat their flashy device to market and work better.

But they'll do fine regardless because branding matters way more than it should.

They're selling a fashion accessory targeting the fanboy tech crowd. They haven't thought of putting a clip on the bottom so you could wear it in public yet, but it's probably coming soon. They're running a bunch of ads for it on social media.

It's really irritating how transparent and obvious of a smoke and mirrors act it is AND how successful that strategy is.

Good on them for making the play but bad on us for hitching on to any caravan that sells silicon like it's the National Enquirer.

Anyway, here you go https://flipperzero.one/one

And the $2.3 million they've raised in 9 days https://www.kickstarter.com/projects/flipper-devices/flipper...


TL:DR; Flipper Zero has at least some purchasers who are only going to be users of the U2F or NFC features, not hacking on or with it much. The fact that it has useful applications out of the box separates it from some of the SBCs etc that need many more parts and code to have a useful application IRL.

> All off the shelf and any engineering team could clone it in a few days while substantially undercutting their price

Perhaps. And perhaps Dropbox is just a glorified rsync FTP target. (Sarcasm here, boiling down to saying that talk is cheap, shipping is hard)

> They're selling a fashion accessory targeting the fanboy tech crowd

I guess that makes me a 'fanboy tech' then.

Unlike the breadboard + parts + wires + needed case of some kind + SBC/microcontroller +need to learn yet another proglang or dialect to make whatever you built DO something, this device comes with a working battery, controls, screen, Bluetooth and U2F, NFC cloning out of the box. No instructables or breadboard required!

While I'm interested in circuit design, there are no simple books that will take you from plugging in a battery to a single led to something that can run c code. Maybe NAND to Tetris? Have had a hard time finding a physical book for that.

Learning to do embedded stuff as a result would require more time than I'm willing to invest right now vs what I could get done by reusing old phones or retro computers to do stuff, or by purchasing a Turn-Key device like this.

And for once the marketing may mean that these people have answers to questions, and do not disappear for months on end while the engineering types silently work and don't dare talk about their problems. Sort of like the opaque development of signal versus the open development of this week in matrix.

at the very least, the dolphin character and marketing or branding on this device shows that there is at least some ux care taken, and there will be at least one or two glossy coats of paint applied over top the bare metal. I compare this to something like a Chinese pen plotter that while technically being only a cheaper version of the do-it-yourself pen plotter or axidraw, has much worse software that barely works and is not intuitive.

Regarding the flipper itself, I purchased it to function as a simple NFC clone for my work badge if I ever go back to work, and as a u2f key manger since I don't already have one of those, not as a hacking or embedded development aid.

Those capabilities and gpio pens are optional to me as a user, which may confirm your suspicions, I don't know.

Like most finished products, this device can provide a high-value to user supplied code ratio. There's definitely a place for devices that are 'merely' hardware and a software platform that enable a one-to-one value to user supplied code device... even I run things like pihole on a dedicated raspberry pi zero since the cost benefit is there. Why, at some point, I'm going to learn native Linux development so that I can write applications for the pine phone whenever it comes out with a keyboard. not everything has to be a completely finished product with applications after all.

Having the ability to hack on something is great, but this fanboy prefers to purchase products that can actually do something for him, yet are extensible, as opposed to buying some pcbs and hoping that one day some way somehow you can make them do something. I have enough almost started weekend projects already :)

/End rant , thanks for your patience with my clumsy delivery skills


While I'm interested in circuit design, there are no simple books that will take you from plugging in a battery to a single led to something that can run c code. Maybe NAND to Tetris? Have had a hard time finding a physical book for that.

I don't want to address any other point, but I can share an experience here.

I went from very basic electronics knowledge (what most discrete components do in the abstract for, though not how to use them properly) to building a couple of SBCs by googling "home brew 6502" and reading whatever came up. I was a very rewarding journey, though there was certainly no book, I had to follow up on a lot of things myself.

Obviously a home brew 6502 won't do much modern, but you'll be left with way more knowledge than you need to build something with bare microcontroller / cpu and the kind of SPI/i2c bus peripherals you're talking about here.

Certainly it's a bunch of effort, but everything is there is you want to research it, and the 6502 forums are stupidly helpful.

Just in case you decide you do want to persue more DIY digital logic. It's well within reach!

PS. Nand to tetris is an awesome project, but it's waaay overkill for what you're talking about. There is a huge amount of cpu and compiler design that you don't need. And actually, not enough electrical engineering...


It's far simpler. Flipper is closer to a custom built PC, where you buy the parts online and then assemble them.

They got someone to probably use solidworks for the case design and they have some brilliant artists but beyond that you can buy all of these components online and basically snap them together, read some short documentation and be on your way.

Their initial goal was $60k, which would have covered 500 units which they probably presumed they'd be manually involved with each of them so it was basically set up for exactly what I described. They just had some runaway success, which is the problem because they still only have a cheap project that takes no special talent or time

I'm mostly jealous how much of success is just pure luck like this


Just out of curiosity, what kind of resources would be needed to be able to develop something like this with a risc-v processor? I know the architecture is in its infancy, but it would be awesome to have stuff like the popcorn pc or even a raspberry pi on a risc-v architecture for tinkering with.


You can’t render anything complex on a FullHD screen with a low-power CPU. Need a GPU. Risc-V doesn’t have it yet, Pocket P.C. does. ARM Mali ain’t nVidia or Apple, but still, it’s not too bad for a mobile device, and orders of magnitude faster than rendering stuff on CPU.


Holy cow look at that silkscreen. Five-digit component numbers? Populating this board is like Where’s Waldo...


And the PCB has 10 layers! They must have had fun designing it.


Kinda does beg the question whether ten layers were really necessary, but I'm not a board designer.


I see a pretty fine-pitch BGA on there, so I'm guessing it is (but also not a designer)


BGA isn’t usually the reason for 10 layers, it’s high speed signals like SDRAM that need isolation and timing.


Not sure why you're getting downvoted, you're dead on.

I just finished routing an LPDDR4 bus (high speed + low voltage => worst case signal integrity) between two BGAs with signals routed on only 3 layers. The catch is that you need each high speed bus to be comfortably sandwiched between solid copper planes. Layers dedicated to reference planes start to add up quite quickly when you've got RAM, eMMC, ethernet, wifi, camera input, display output, SD card, etc all crossing over and through each other.


The stackup pdf only has 3 digit component numbers.

What you're probably seeing is that someone paneled the PCB 16 times.


I wish there was a video on the website. What are the use cases?


Confusing name but not related to Microsoft’s obsolete Pocket PC platform


That was immediately where I went. I loved my little iPAQ back in high school.


I could only afford a Handspring Vizor. iPAQs were so high-end.


I’ve only used WM5/6 but compared to PalmOS it genuinely sucked so


JFTR, There is SunVox[0] — free modular music studio for Palm OS, Windows CE (Pocket PC & Windows Mobile) and much more.

[0] https://warmplace.ru/soft/sunvox/


I totally forgot about the handspring, I had one in high school. Great device!


I still have two that still run. I wonder if the Computer History Museum in Roswell, GA, would take them as donations.


I could only afford an iPAQ but I really wanted a Jornada.


I wanted a Sharp Zaurus. They had a pretty good Japanese hacking community.


I had an iPaq and a Zaurus: the quality difference was miles removed: the Zaurus was so much better at everything (for me) but the iPaq and gprs so that was a reason to hold onto it. Zaurus is still working and running Linux today, iPaqs, multiple, are all dead even though used less. Build quality was robust even though it had a swivel screen (which was also great).


I still have a zaurus 860C in the box never used, and I can’t find anything to do with it.

The build quality was amazing, and so was the Silicon graphics display.


Through work I think I had them all. Various palms, psions, iPAQ, Dells. Good times.


I preferred my Casio Cassiopeia.


Yes! I had an E105... What an awesome little computer.


'numpad0: “Confusing name but not related to Microsoft’s obsolete Pocket PC platform”

I suspect such retrospective revisionist reports are designed to precisely do that, render confusion. For instance, according to wikipedia, Windows NT was never designed with the Internet in mind. Whatever that's supposed to mean.


I saw the title and remember how when the battery died it wiped the memory. Brutal


According to Wikipedia, Palm copied the “Pocket PC” from the PDA, who knew. Or was it Microsoft first named the “Pocket PC” the “Palm PC”. So as to render confusion in the public mind and steal market share. That's probably when Palm became “a newer entrant into the Pocket PC market”

https://en.wikipedia.org/wiki/Pocket_PC


Sharp may have predated that around 1980. They were called pocket computers but the model name was PC-nnn surely implying it was a PC.

https://sharppocketcomputers.com/


I would buy something with slightly different specs and design.

- square screen, possibily of higher resolution - high quality tactile switch keyboard, optimized for thumb typing - useable in bare PCB form (maybe with a two PCB sandwich) - similar design aesthetics as the pocket operators from Teenage Engineering, but built as a powerful full featured pocket computer for hackers


That was the Blackberry Passport hardware-wise.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: