Of course there has been no progress on reverse engineering the epaper display driver in the past three months. Furthermore it is questionable why anyone would need to RE it, considering that the driver is in-kernel. The assembly file that rockchip shipped with their sdk seems like a GPL violation to me.
Good point. That page says the BSP ships with a GCC-produced driver blob, without source. Would be good to hear from the Pine64 folks on whether they've contacted the vendor and demanded source.
As for blobs without source, the same happens with x86 all the time. If you really want a fully deblobbed computer, you can try one of the POWER boxes from Raptor.
Disagree. I didn't hear about it back in September, and am glad to know they're at least working on this. So, posted it in an attempt to spread the awareness. :)
If you do something really stupid and somehow fry your SD card in your Raspberry Pi, that's fine because you can always pop in another SD card. I wouldn't want to accidentally burn out the internal storage on this tablet and have no way to replace it.
Procedure for "entering maskrom mode", from the link above, is so cool! Reads like some magic ritual:
> Flip the device around so that the display faces down
> Lay the pen on the right side, with its tip pointing towards the speaker grill, and its magnet pointing towards the upper right corner of the label on the back.
The old cylindrical mac pros had a similar trick where the case had a magnet in it and the computer itself had a detector for it. So if you ever wanted to turn it on without the case to see the diagnostic lights, you had to know where to put a magnet to fool the sensor.
This is an exceptionally overpriced way to do it: Dell PCs tend to just have a little push switch that's depressed when the case side panel is on. (For a corporate environment, these are also very helpful for tamper protection, of course.)
The key difference is that apple cares what their computers look like when they're open. Dell doesn't. An ugly switch would ruin the line of the Mac's case and look out of place.
Now, you may disagree with that philosophy, but Apple is wildly successful for it. Personally, I don't like Macs anymore, but I do absolutely respect that attitude.
Apple believes that even when the computer is open, it should look good. Thin Steve loved joking that the back and insides of his computers looked better than the front of the competitors.
Not really as they have always been fairly open about working closely with factories in Shenzhen that product the products if you subscribe to their monthly email digest. Their previous US based offices seemed kinda pointless.
If you're worried about that, you can go with Purism. But, because their employees make San Francisco salaries, and they lack expertise and connections within China's supplier/manufacturer scene, Purism can't even come close to competing with Pine64 on price.
Little confused. So eventually the PineNote will come with a custom Linux distro for end-users, I presume, that works out of the box with the hardware. But what are devs at this point supposed to flash, any flavor of mobile Linux?
At this point, whoever buy the device are the ones who figure out what is supposed to be flashed. That's basically how Pine64 works: they make the hardware, and let the community work on the software, including the OS and drivers. It worked decently for phones, albeit slow, the software seems to be coming along gradually.
The PineNote is very close to the design of the Quartz64, there is a Linux BSP for that, you would flash a Quartz64 distribution and use a serial console.
It's too bad it needs Linux as a base, would be great to build a Smalltalk-based system using something like the circle library for low-level access (https://github.com/rsta2/circle). Hopefully someone can reverse engineer the binary blob for controlling the eInk display.
Well, something like the smalltalk-78 that Alan Kay demos here (https://youtu.be/AnrlSqtpOkw?t=242) would be an ideal starting interface for e-ink based tablet.
I really want an 8.5x11 or bigger, very low power e-ink device that I can use as essentially a photo display. I was inspired by this[0], but as I'm not a hardware guy at all, trying to make one is out of my reach. Hopefully this PineNote will become that device sometime.
I want an 8.5x11 device with large-size microUSB.
Nothing fancy, no markup, no net connection, huge battery.
Simple page turning (could be touch or button).
Simple file handling. Memory of where I stopped reading.
You could sell these things in volume, the first venue
being libraries and bookstores.
Students with school library access could use them to carry
all of their papers. Profs could use them for research.
Boox: Totally would if the price came down a bit. I know that's a big ask with eInk displays being proprietary but a guy can dream.
remarkable: Don't know enough about it but maybe the same caveat applies.
The trouble with sheet music is full size is quite large. 8.5" x 11" is probably "close enough" much of the time but even that is bigger than most eink displays.
I'd be interested in buying and tinkering with one... I'm not afraid getting my hands dirty with embedded Linux stuff, the only thing I am afraid of is that I end up with a board that (despite saying "production run") has some issue on the PCB that makes core functionality (e.g. CPU, RAM, display, USB-C or WiFi) unable to work in hardware.
What are the options in case there indeed is some sort of defect on the PCB?
Why not buy from brands that make consumer products?
To me, Pine products are for hackers & makers. I've three pinecils, both pinepowers, a couple "woodpecker" serial dongles, and some related accessories. Everything I've bought has worked great and been worth every cent. I wish they keep doing exactly the kinds of things they've been doing.
I'd argue that anyone wanting a fully open source anything isn't an average person.
If you want something finished and mostly open, a reMarkable tablet already works. If you want something with android and are okay with GPL being completely ignored, let me point you towards Boox.
I totally get releasing products like this to a 'developer audience' when its not fully working / complete but charging that much for it in that state...
What a joke. You can get a remarkable 2 for less than the cost of this device when "Most of the device’s functionality, including the e-paper display, do not work at this time".
You're just not the target audience, they make it clear that it's a device in its early stage of development.
The goal is clearly to build a community that will help building an open-source ecosystem around this kind of device.
The kernel drivers should already be open source. Just not upstreamed because rockchip doesn't develop upstream first, they only do it later and partially.
Nobody has attempted to reverse engineer the boot blobs. Hard to do.
My question is: why rockchip? Why not focus on amlogic or mediatek? What does rockchip offer the open source community that other vendors don't?
The move from Allwinner to Rockchip was mirrored by many of the other Pi competitors in what seems to be mostly a need for more performant SoCs. Given a lot of this market is driven by fabless Shenzhen based companies it probably comes down to the right mix of drivers/production/price.
Although given Mediatek has been moving up a lot in the world and powering some really nice budget devices (Lenovo Chromebook/Chrome Tablets come to mind) I really hope for some sort of mediatek SBC to materialize as ARM adoption grows.
For now chromebooks remain one of the only ways for me to get linux (via crostini) ARM devices with any sort of performance vs slowly tinkering around on a Pinebook Pro.
It seems insane that they're trying to build a community of support, but then have little to no documentation for so much of the hardware. If your product depends on the community for success maybe you should try to help that community succeed?
I dunno. After reading HN's more recent take on the remarkable 2 and its move towards a subscription model[0] I can see a use case.
Also, comparing stats[1] they are different devices[2]. The PineNote has a more powerful CPU, 128G of storage vs the Remarkable 2's 8G, and a bigger battery.
So you get more hardware and freedom for the $. I think also that price point, coupled with the warnings, forces people to think hard before they make the jump to try it out.
On the flip side, for those of us who enjoy this sort of hardware hackery, this is a reasonable price for this collection of parts all conveniently wired up and ready for the fun part.
I can appreciate that, as I have paid for similar 'hacker focused' projects e.g. playdate.
It just seems very expensive to me considering how lacking it is even in terms of basic drivers but I suppose if people who are into this stuff think it's fairly priced then it's all good! Replies to me certainly seem to confirm that!
Remarkable tablet has a closed soure eink display driver in userspace. Rockchip has a closed source eink display driver in kernel space. At least remarkable isn't violating the GPL.
As a developer I'm a little confused. Are they skirting around having to pay developers to build the software for their device? I'm supposed to pay $400 USD for a development kit that comes with no tooling or support and... what? Do I get to release software for the platform under the Pine brand? Will Pine buy my drivers and software back? Will there be an ecosystem where I get a share of profits for the work?
Your outrage seems misplaced. People are excited to have an open platform to play with, that is not trapped in some form of digital garden.
Many people contribute to mainline and to distros because they believe in those projects, no-one expects that these projects co-finance their hardware.
People will spend a lot more money than 400$ in other things, this is how they choose to spend both their money and time.
Sorry, not outraged just genuinely confused what the draw is here for folks and what Pine's plans are for the "open source community."
Will they eventually release a consumer version that packages up all of the software other folks have written on their own dime? Or do they only release these as developer kits?
> Will they eventually release a consumer version that packages up all of the software other folks have written on their own dime?
That has been the Pine M.O. for their product line. The PineBook was a 'developer' release about 4 years ago. Now it is sold commercially as a full fledge ARM laptop.
The PineTime has now reached a stable hardware but still under SW development and the PineNote is I think the first HW revision that they are shipping.
> I'm supposed to pay $400 USD for a development kit that comes with no tooling or support
No, you're not. In fact, if you're not a developer who specifically signed up in advance to get your hands on the hardware in order to start building the stuff that will make it usable, you can't buy it right now at all.
Ya I’m a little confused, why is this thing $400 for something that doesn’t even work? Is the target audience developers who want to develop drivers and stuff for this? If so shouldn’t it be as cheap as possible? I don’t see the point in trying to make money at this stage when your target is people interested in a non functional product.
Producing electronics at small quantities is expensive. Prototypes are expensive. Economies of scale is everything when it comes to producing cost-effective electronics.
>Is the target audience developers who want to develop drivers and stuff for this?
Correct that's the way Pine64 works. Really it's the way the vast majority of niche ARM powered dev boards work. It is often stated on the product pages for people that don't seem to understand this. Their devices are often sold for very little margins above cost as they are mainly to drive developers to the platform. (as development on these devices benefits the SoC manufacturers)
If you want something more consumer friendly come back in two+ years when some of the kinks have been worked out.
There is no indication they will ever leave this stage. I believe they generally sell at cost, even for working products. The purpose is to advance the technology, not to make a profit. But, the manufacturing and NRE costs need to be paid somehow.
People seem to get really upset when its pointed out that they (Allwinner & Rockchip) are just profiting off of Pine64's community developing linux support for their SoCs. They seem to miss the point that the community wants products like these to tinker with and don't mind doing some legwork to get things functional.
I believe Pine64 is making profits off of people buying developer devices that have little-to-no actual functionality as a means of funding the development of the products in question.
To then follow that up with allowing most of the multi-million dollar development to be done by hobbyists who have purchased your non-functional hardware and you have a recipe for... something.
“ The main purpose of Pine Store creation is servicing PINE64 community. Most of the profit generated from Pine Store uses for production cash flow, new product engineering fee, sponsor community event expense, maintain server cost, and also supporting other open source communities.”
> Developers will need to use a coupon code to complete the purchase.