There seems to be a model of the ASML TWINSCAN NXE3400C which shows the inner workings, but doesn't show on the ASML site. Would be neat if someone could track down the build instructions.
Or, at the rate of $0.30/kWh in California for residential use, $2.70/year. Lowering power usage on embedded devices is important when you're producing thousands, but for a hobbyist it doesn't really matter.
I've been using the Onyx Boox Max 2 for a couple years and it's a pretty good device. When I purchased it they stated they would update the Android version, but they never did so it's stuck on Android 6. For reading PDFs it's good and there aren't really any competitors with this large of a screen. I'll often read books in dual page format, or read manuals in single page.
The device crashes frequently if you try to use any third party apps, and the pen has major lag if you're not using their notes app. Don't turn on the wifi or else the battery drains extremely fast. Make sure you don't leave it charged overnight or else the battery might balloon and explode. When this happened Onyx replaced my battery for free.
Despite these shortcomings, it's still better than the Sony digital paper (which I've also used and sold). The pen is pressure sensitive, which isn't in the Sony. The writing experience is worse than on the Remarkable, but Onyx has a bigger screen.
I'd recommend Remarkable if the screen size is good for you. The software is a lot more stable and the device is a lot more polished in general.
I've also used the Raspberry Pi + Coral TPU. For a while it was almost impossible to get the USB as it was always out of stock, but those issues seem to be gone. I would not, however, recommend the devboard as the documentation is quite sparse (runs on an NXP processor). RPi+Coral is a good combo, and the Pi5 supports the PCIe accelerators.
The BeagleBoard BeagleY-AI that was just released costs $70 at 4TOPS, equal to the Coral which is $60 for the USB alone disregarding the cost of a Pi [2]. It just came out, but BeagleBoards have quite a good community reputation and because the hardware is more open, easier to develop into a commercial product. The older BeagleBone AI-64 has a processor that supports 8TOPs, and TI has processors that go up to 32TOPs if you have the skills to create your own board (I don't).
I would not recommend developing with the NVIDIA Jetson unless you have particularly deep pockets and like to use an out-of-date toolchain. The latter isn't too unfamiliar to the embedded world though.
Thank you so much for sharing your detailed insights and experiences! I'm particularly intrigued by the Raspberry Pi + Coral TPU combo, and it's reassuring to hear that the availability issues seem to have been resolved. The mention of the newly released BeagleBoard BeagleY-AI also caught my attention, especially given its cost-effectiveness and open hardware approach. I've appreciated the support from the BeagleBoard community in past projects, so the BeagleY-AI seems like an attractive option to explore.
Regarding the NVIDIA Jetson, your comments affirm some of my reservations about the costs and the currency of the toolchain. It seems like it might not be the best fit for my needs at this time.
Do you have any experience or tips on getting started with either the Raspberry Pi + Coral setup or the BeagleY-AI, particularly? Any advice on kickstarting a project and navigating potential pitfalls would be invaluable. Also, if you have any recommendations for resources or communities focused on these platforms, I would be extremely grateful.
Hard links only work in the same filesystem. If your system only has one volume with one filesystem (e.g. a PC) that's fine, but not a very portable option for servers.
Latency is an issue even inside a transistor itself, something known as propagation delay. When you have a bunch of logic gates chained together the propagation delay increases and the chance for a race condition occurring also increases. The propagation delay of a logic gate is one of the limiting factors of the speed of a circuit.
Transistor gates act as capacitors in a way. This capacitance increases the delay between switching on/off. [1]
To lower this delay, we can shorten the transistor gate. We've hit a wall with how small we can shorten this length, and it is exceedingly difficult to make smaller gates without running into quantum effects.
Higher electron mobility means that the delay inside the gate is reduced, allowing faster circuits.
[1]http://ece-research.unm.edu/jimp/vlsi/slides/chap4_1.html
Switching speed of MOS systems strongly dependent [on]:
Parasitic capacitances associated with the MOS transistor.
Interconnect capacitance of "wires".
Resistance of transistors and wires.
If you haven't done a calculation incorporating an estimate of the heat content of the Earth's interior, then you cannot know whether humanity will be dead by then.
Assuming just the inner core, using the lower bound for its temperature, specific heat capacity of iron, 20 TW being extracted, ignoring the extra power inputs in the form of radioisotope decay, crystallisation, tidal heating, I get 800 million years, which is close to when photosynthesis stops because of geological processes locking all the carbon away (trees stop working at about 600 million years, not sure about plants in general).
Converting magnetic force into electrical current (electromagnetic induction) is the principal behind of all electrical power generation. SMES (superconducting magnetic electrical storage) is essentially an inductor made using superconductors. Electrical current gets converted into magnetic force and then back into electrical current when the circuit is discharged.
Have you tried using Fluent Terminal? It is available on github and the Windows Store. To use mosh you need to connect via the quick connect menu in the top left corner of the program.
I haven't done a test yet, but the Fluent Terminal seems fast enough and doesn't eat up that many resources surprisingly. As a UWP program it feels very snappy (as opposed to Hyper being a full on electron app using 200MB for simple text output!) I don't care what technologies they use to build a program, as long as it works. I haven't experienced any hangups using SSH, so it's good enough for me.
I tried a ton of them for a while and the microsoft terminal or wsltty both use the least amount of resources, open the quickest and have the lowest input latency.
There seems to be a model of the ASML TWINSCAN NXE3400C which shows the inner workings, but doesn't show on the ASML site. Would be neat if someone could track down the build instructions.
https://asmlstore.com/collections/ready-to-ship/products/twi...
There is another set available on their site.