Interesting project! I am relatively skilled at FPS with a mouse, but would like to improve with my trackball (so that I don't have to change my setup when playing an FPS). This could cone in handy, though I imagine that there aren't multiple "mouse profiles" built-in? It could be useful for players to understand how much of a difference different mice make.
I don't have the time to provide such a detailed answer, but there are a few options.
At least on Sway (and I imagine other wlroots-based compositors) it's very easy to do: create a virtual output and start a VNC server attached to it. Start your vnc client on windows. A comment I posted 5 years ago: https://news.ycombinator.com/item?id=25892128
The comment linked above is on the topic of Deskreen, which is a tool that does that.
Funilly, I also commented something similar there: https://news.ycombinator.com/item?id=34010970 the article explains how to do it on a lower level (pretending a display is connected to the GPU), works under X11 too.
A few days ago, an update locked me out of my server. I used a cheap USB HDMI capture card to use my laptop as a screen, which worked very well: that could be another option.
Then I guess that there is this option, which is new to me. It feels like this could be a toogle in sunshine, somehow :)
I still use my Samsung n130, though I went with Alpine after Arch dropped support for 32 bit (there were a few pain points with Arch 32 in the early days, I tried guix but it was too slow and guix uses a lot of RAM during updates).
It works decently, is sufficient for ssh-ing into other hosts. Though web browsing is a pain. I used to mostly use Dillo and elinks, MPV+yt-dlp for videos.
Unfortunately I left it sit for a bit too long and the battery is dead now. I'm thinking of fixing it and upgrading the power port to USB-C.
Sometimes I also think about building a compute module-based motherboard for it.
There were interesting bits in the setup (blacklisting defective RAM addresses for instance), maybe I should make a short writeup :)
> I wouldn’t mind a PD aware charge-only device but inquire require some MITM circuitry and I’m too lazy to make one.
Just get an USB data blocker. I bought one off the shelf at an airport for 10€. This is the "condom" you are talking about, just a small dongle. Then travel with a fully featured cable!
I also use it to charge my phone at my workplace on the docking stations (which would be a big no-no otherwise). I think our IT provides them on request too, I've seen them advertise it.
Not OP, but I can imagine piloting it with a temperature sensor: too warm=>increase speed. It could avoid some sweat-covered wake-ups, depending on the circumstances (trade it for a noise-triggered wake-up, perhaps).
Another use-case I have for this is in a house where the mezzanine gets much warmer than below when heating during winter: detect this with a couple of temperature sensors and turn the fan on when needed.
Funnily enough, I did almost exactly that last week, but I kind of gave up for now as my 433 MHz transceiver does not seem to work well with esphome? It does seem to output signals when I use the remote. Oh well, I will revisit when I have more time.
A major difference is that I relied on a flipper zero I had to do the 433 MHz capture instead of an RTL-SDR (I have one too, but have not needed it yet).
For reference, the transceiver is the one that came with this product: https://www.athom.tech/blank-1/8ch-inching-self-lock-relay-f... (which I already use for controlling a heater). Weirdly, it is not listed in the official esphome config (possibly because RF is a recent addition to esphome). My next step is to test it on another esp board & compare timings with what esphome generates.
I should start doing write-ups too. I came across quite a few interesting sources during this mini project (2-3 hours max).
2. Go to previous app view. This is app-dependent though it will probably, successively with each press:
a. Close menus if open (context, sidebar, etc)
b. Go to previous (web)page if web/file browser
c. Go out of submenus (ex: settings/WiFi -> settings) if not in a browser or if the oldest page has been reached. Keeps walking the tree upwards.
3. Reach the main app view (usually the one you land on when opening the app)
4. One more press minimizes the app.
It is fairly consistent, but some apps decide otherwise:
* some will minimize as soon as you press it (I've seen games do it)
* some will open a new menu (again, games: pause menu)
* some will seemingly walk you the history of visited pages instead of the hierarchy -- which may make sense but can be confusing
* some old apps will display a toast "press back twice to exit". This used to be common back when physical buttons were the norm, but I haven't seen this message a lot.
So, mostly consistent with some weird-behaving apps. Same as on desktop I guess?
Ah, that's interesting, the requirements are similar in the CAD industry. Dassault Design Sync is used a fair bit for semiconductor design databases, for instance. An open alternative would be welcome!
Edit: I do feel a bit uneasy about epic games, though.
I like Windward a lot, even though it does not feel like a complete game. It would benefit from a storyline IMO, and the dev has stopped updating it. I am longing for an open source project similar in scope.
Should I consider using Iroh for intranet communucation ? Is it a viable use-cases? The use-cases I have in mind is an app being deployed on a HPC cluster, onto many nodes, cut off from the internet.
I had been looking at zeromq, but I very much agree with using keys rather than IPs where possible (after years of using yggdrasil, wireguard, tailscale, tor), so I am tempted to try Iroh.
OTOH, this seems overkill if I'm using a client-server approach where the server IP is known.
Thank you for the detailed answer. This is an interesting engineering problem.
I wish I had more free time, a gearbox or a 4-way differential can probably be used on the extruder head to control the mixing ratios.
Regarding the necessary research, I can picture it being cracked either by hobbyists, or industrial-scale (patent-heavy) players. I imagine that consumer-facing players such as Prusa do not have the means to heavily invest in R&D.
Hence why I was asking the question, thank you for answering, I did not consider the high viscosity of the filament when I wrote this. As you say, it can be worked around, but that would require extra complexity/cost.
Thank you for the clear & concise answer. A rotating nozzle and/or alternating colors might work then, but that would be a relatively complex engineering project.
Hmm, I am not in the 3D printer space anymore, but I am surprised they went with alternating layer per layer, as that severely limits resolution. It's probably the simplest way to achieve reproducible results, but I can think of a few other ways:
* the simplest is just mixing filaments, like one mixes paint. The article doesn't spell out the reason it doesn't work, I am curious as to why.
* together with alternating layers, colors could be alternated in the same layer. Some purging may be necessary, but I think you could either: accept some mixing (compute its impact to compensate) / take into account the volume in the nozzle (extrusion "latency") / discard the unwanted part in the infill (at the cost of less smooth edges)
Of course, the hard work with any approach, including their current work, is calibration, as the article highlights. I wonder if off-the-shelf monitor calibration sensors could help with measuring the filament you have at hand.
I ended up being able to authenticate using a passkey, IIRC. Can't remember if that was my yubikey or bitwarden. I was surprised that it worked at all.
It didn't use to be complicated, but an update messed stuff up a few months ago (halo infinite).
You can contact me at hn at mayeul dot net
I live in France, working for a semiconductor company.