I knew it, I’m surrounded by assholes.

KEEP FIRING, ASSHOLES!

The major thing this would necessitate is more communication between the slicer and the printer. I run my printer from an SD card, because I’ve got an ancient printer.

If you mean, why not install a load cell in the spool holder instead of an RFID reader, well…

I just happen to have four empty 1kg spools lying around, because I’m a total packrat. Let’s weigh them:

1. 343g
2. 319g
3. 300g
4. 254g

So that’s a range of 89 grams, out of just four spools. And these are all visually similar 1kg black plastic spools. I’ve seen skeletonized spools that tried to reduce plastic, I’ve seen cardboard spools, and I’ve seen spools of different sizes. How is a printer supposed to tell a mostly empty 2.5kg spool from a full 1kg one?

Then…What happens if you load one new spool, use some of it, unload it, use a different spool for awhile, then switch back to the first? Will you have to manually key in a tare weight for that first spool?

If you install a load cell in addition to an RFID reader, well then the spool’s RFID chip could store the weight of the spool, the initial weight of the filament loaded, and the weight of the remaining filament, and the printer could weigh the spool to verify that, which could catch and correct errors caused by oozing, miscalculation, using the spool on another printer, having to cancel a job mid-run because of a problem, etc. I’d kind of like this for reloadable spools. Somebody is coming up with split spools that you can buy just the filament for, and then you could reload the spool with another load of the same filament, and a printer with a load cell could automatically weigh and recalibrate a reloaded spool including an updated tare weight.

All told though, given how much it matters, I’d be fine with the dead reckoning approach done by the slicer. I mean, my personal 3D printer just turned 11, it has no auto bed leveling system, no filament runout sensor, no auto loading system, hell I haven’t updated the firmware since Barack was president, and I’m in the habit of running one spool all the way empty, and just shoving in the start of the next spool as the printer runs. I’ve done that for two-color signage and such, something with colored raised lettering on a white background or something. You can get away with shit on a primitive old clanker like mine that the newfangled units won’t put up with.

For new users, if it doesn’t exist in the repos, you’ve gone too far.

I don’t think this holds up under scrutiny. Theoretically sure, installing using your distro’s package manager is the beginner skill, compiling from source is the advanced skill.

The reality is, people transplanting from Windows often own hardware they want to continue to use, that require software that isn’t in a distro’s package manager. For me, this included a DisplayLink docking station, an Epson printer and a SpaceMouse. For some, it will include gaming keyboards or mice, stream decks, who knows what else. A lot of times, there are folks making open source software for these things, but they don’t package them. So you end up on Github as a beginner looking for the thing to make your thing work.

As you migrate into the ecosystem, you start buying hardware that is well supported by the Linux ecosystem, that problem starts to fade away.

by rpm vs deb, I wasn’t meaning downloading individual files…though I’ve done that. DisplayLink offered their driver as a .deb. At first, that Epson printer only issued a .rpm, and I had to use Alien to install a .rpm on a Linux Mint computer. With time, they offered a .deb, and eventually the printer was just natively supported by CUPS. I meant, I find that the Debian/Ubuntu repos (the dpkg/APT system that uses .deb files) have more stuff in them than Fedora’s repos (the DNF package manager that uses .rpm files) do.

Does Mint still not use Wayland?

When I built my current PC, Wayland support in Mint Cinnamon was “We’ve just now added it, it doesn’t work worth a damn but you can try it.” They’re coming along, but they’re behind.

Is an older codebase generally good for new users? The first distro I installed on an x86 PC was Mint Cinnamon 17. Quiana. On a then brand new Dell Inspiron laptop. For about 6 months, the kernel that shipped with the OS didn’t support the laptop’s built-in trackpad. I had to manually update the kernel through Mint Update for the trackpad to work. There’s problems at the bleeding edge, but there’s problems at the trailing edge as well.

I see it as useful for some rando with a printer as well. If you’ve got a printer to make little dinosaurs for your kid or the occasional curtain rod bracket, that little warning “this gcode file was sliced for ABS, but you’ve got PETG loaded. You sure?” or “There’s only 177 grams left on this spool, this print will take 190.” can be helpful.

If the tag is read-only, it can allow:

• marginally better loading, as the printer can heat the nozzle correctly for that filament without input from the user.
• Comparing a G-code file to the loaded filament, either to throw a filament mismatch error, or to adjust temperature settings on the fly.
• Allow slicer software with a network or serial attachment to the printer sense what filament is loaded

If the tag is writable, it can allow for keeping track of how much filament remains on the spool, by writing how much was consumed during each print. This means, when you get to the end of the spool, the printer can warn you if there isn’t enough filament remaining without having to manually track the mass of the spool.

At least some of the problems I reported about Bazzite are inherited from Fedora. Bazzite didn’t create Anaconda.

Fedora has the problem of being generally fine, but most of the world for the last decade has been targeting Ubuntu as THE Linux distro, so there’s a lot if Git repos out there that don’t include instructions for Fedora. Way fewer things are packaged in rpm rather than deb. I’ve never seen Linux Mint kernel panic unless I was fucking around with the video drivers, I’ve seen Fedora kernel panic.

The main reason I’m using Fedora right now rather than Mint is Mint tends to have an older codebase, and we’re at a point in PC technology where things like wayland offer support for video and graphics stuff that don’t work well under X11. like my 1440p ultrawide 144Hz monitor sitting next to a 1080p 60hz side monitor. Fedora KDE has it ready to go, Mint Cinnamon does not.

I had one fail fairly early, giving me a cryptic message because apparently it couldn’t cope with how I’d set up the partitioning.

I’ve had a Linux Mint install fail because it couldn’t cope with a BIOS setting, the error message gave a plain English explanation “it’s probably the XMBT (or whatever acronym) setting in the BIOS, see this page on the Ubuntu wiki for details:” and it gave a hyperlink, because the installer runs in a live environment, it had a copy of Firefox ready to go, AND it gave a QR code so you could easily open that link on a mobile device. THAT’S how it’s done.

Having played with it for a little while now that I’ve got it installed…I think it’s alright for a mostly or entirely gaming machine. I wouldn’t want to use it, or any immutable distro, as my main computer.

I’ve attempted to stay out of the trendy distro of the month club, remember Garuda? Remember Peppermint? Remember Endeavour?

Bazzite offers KDE or GNOME, and in the menu mentions KDE is what is used in SteamOS.

I installed Bazzite on my HTPC recently. It was the worst install process I’ve seen in over ten years of using Linux. I shall enumerate the problems I had:

1. The image is weirdly large, it’s like 9GB in size. It takes awhile to download and a weirdly long time to write to a USB stick.
2. Once written, you boot the image, and GRUB has the options to Install Bazzite or Test Media And Install Bazzite. By default, Test Media is selected. It always fails this test.
3. If you use the typical non-live environment image, the scaling is tiny on a 4k monitor, and there’s no way to adjust this.
4. If you use the live environment image (in beta at time of writing), it might just lock up. I had that happen twice just while clicking through the Anaconda installer.
5. The Anaconda installer, which I think they inherited from Fedora, was I think designed by one of the contrarian idiots who work for Gnome. There’s a DONE button up in the far upper left hand corner of the screen that sometimes acts as a back button, sometimes acts as a forward button. You have to move the mouse from the top corner of the screen to the center of the screen a lot, for no reason. The top-left corner of the screen is a dumb place to put a DONE button because most languages read top to bottom, left to right, the DONE button is where a START button should go.
6. There isn’t a simple way to tell it “put / on this drive, put /home on that drive.” There’s an automatic installer which will do god knows what…fail, most likely. There’s a “custom” partition dialog which I couldn’t make heads or tails of, and then there’s a “custom advanced” one that lets you set the size and position of each partition to the byte. Doing it this way apparently REQUIRES you to not only set up a /boot/efi partition, but also a /boot partition separate from /root.
7. If you’re in the habit of putting /, you know, operating system and software, on one drive, and /home on another drive, you have to learn from osmosis that part of Bazzite’s immutableness means that there is no /home, there’s a /var/home symlinked to /home.

And if it doesn’t randomly lock up, you’ve got Bazzite installed!

Bazzite markets itself as a newbie friendly Linux. They’ve got that configurator on their website that gives you a little Cosmo quiz about what system you have, what desktop you want etc. which is good! That is good user friendly design. But the actual software you get rattles like a Chrysler. How many noobs are going to bounce right off that?

Precisely.

30 knots indicated would take a plane out of the sky too, that’s well below most airplane’s stall speeds. We can probably fix that before doing the Monsanto Slam though.

Things like Minetest exist.

Fun fact: In several episodes of the original Star Trek series, Spock can be seen holding and using an E6B flight computer, which is a slide rule invented for the US Navy in World War 2.

Putting my flight instructor hat on here, this is mostly correct.

“Indicated airspeed” (IAS) is what number your aircraft’s airspeed indicator is pointing to. As Rivalarrival described, the airspeed indicator is a barometric instrument that compares ram air pressure with static pressure to measure the dynamic pressure, which is a function of airspeed. Indicated airspeed is an indication of how the airplane will “feel,” how much force will act upon the aircraft in maneuvers, which is why force limit speeds such as maximum flap and landing gear extended speeds, stall speeds, max normal operating speed and never exceed speed are marked on the airspeed indicator.

“Calibrated airspeed” (CAS) is indicated airspeed corrected for instrument error. The airspeed indicator and the plumbing it is hooked to aren’t perfect, so they’ll be off by a few knots especially near the lower edge of its range. You find a chart in the POH that says “IAS 45, 50, 55, 60 etc” on one line and “CAS 43, 49, 54, 60 etc” on another. Pilots use this for, if we’re being honest with ourselves, nothing.

“True airspeed” (TAS) is indicated airspeed corrected for air density. The airspeed indicator is flawed in concept: It’s a pressure gauge calibrated in units of speed. To actually determine the relative velocity of the aircraft through the air, we have to do a bit of math comparing the outside air temperature with our pressure altitude, this will give us our density altitude. You then do a bit more math to correct calibrated airspeed for density altitude and get true airspeed. E6B flight computers have little windows for this.

Here is my old cardboard E6B from when I was a student. I’ve set an air temperature of -40C over 30,000 feet in the right-hand window, the center window is showing…pretty much exactly 30k feet of density altitude, and we can read true airspeed over calibrated airspeed on the A and B scales. So for 100 knots, we can look at the 10 on the B scale, and read about 164, maybe 165 knots on the A scale. At 30,000 feet and +50C, which literally never happens, your density altitude is ~38,000 feet and 100 KCAS will get you 194 KTAS. Not quite 400 😜

“Ground speed” is true airspeed corrected for wind. To calculate your ground speed, you need your true airspeed as we just calculated, and winds aloft forecasts from one of the government agencies the Republicans are desperate to destroy, and then we do some trigonometry. You can whip out your Ti-83 Plus Silver Edition from high school and SohCahToa this bitch, or you can flip the E6B over to find a handy dandy vector plotter, which does ground speed and wind correction angle calculations by accurately drawing and measuring the triangle. My high school physics teacher called using this thing “cheating,” I call it “a required aeronautical skill.”

I always appreciate the girls who go to conventions cosplaying as PS1 era Lara. It’s not a difficult costume to source, you need a pale blue tank top, brown shorts, brown boots, fingerless gloves, a little backpack, the pistol belt with the huge square buckle, and half a shoebox.

Mechanical/civil engineering software, music production, and digital art. Those are the big ones.

I’ve been looking into robotic lawnmowers, and they’re basically the same. The more primitive ones have a hall effect sensor under their snout feeling for a wire you bury around the edge of your yard, and do the “go until you hit something, turn a random amount, repeat until low battery, follow perimeter to dock” or they require phoning home in some way, shape or form.

Meanwhile, some guy’s got an open source system that runs on a Raspberry Pi on the mower itself.

I guess I’m willing to believe that some of the LIDAR or camera-only guided mowers need some serious processing power to create the maps they use for guidance around the yard, and that’s more practical to do on the company’s servers than on the device itself…except not really; we’ve got decently powerful ARM SoCs that don’t cost much, don’t take a lot of power to run, and can do that job. The reality is, you can’t get a pedometer app for a smart phone that doesn’t broadcast sensor telemetry to two continents these days.

It’s pronounced “gulp-noo”

WINE is true, it Is Not Emulation.

The one I hate is LAME, which definitly IS an MP3 Encoder.