The Volt also contains a gasoline-powered generator, which would be really useful if you could connect it to your house.

I haven’t had the courage to run executable code from P2P networks since the early 2000s. Even then it was probably a bad idea.

chrome : chromium :: vscode : vscodium

That’s a good pun. Clearly the authors have mastered the second hardest problem in computer science.

Google is at fault here for creating the software-defined garbage, but they’re not literally selling the products, are they?

AOL came on floppies originally, but the quality was so poor that you could barely rewrite them.

I’m made of meat.

You can even have your printer run shell commands

I briefly considered using Klipper to make a clock that prints 1 layer per minute, but gave up after realizing it’d be unreadable after the first hour:

When my Ender 3 S1 (not plus) had bed leveling issues, the problem was caused by backlash on the Z axis. It’s important that the Z axis be just loose enough that downward motion is driven by gravity. If instead the Z screws have to “pull down” on the gantry, then the height will be too sloppy for ABL to make fine adjustments.

I didn’t find an alternative, when I looked a few months ago.

There is a USB-C IR blaster that exists, but the Tiqiaa/ZaZaRemote app is awful.

Well, if you currently have this problem and want to fix it, I’ve shown you the way. OpenWrt is free software.

Otherwise, there’s no point arguing about it.

Multi-hour downloads have been a thing since capacity was measured in kbps. If a simple TCP transfer causes excessive queueing, then the queueing algorithm is broken.

A router with OpenWrt and luci-app-sqm can fix this problem, at least for an internet connection with a fixed speed limit.

It’s not clear to me what you’re trying to do. Are you looking for Dual4010_Satsana_by_Gorroth.stl but in Fusion360 format?

The original model has some .step files.

One major AAA game update will likely break your connection
One person in the house uploading anything will cripple your ability to make ANY request

You are describing symptoms of bufferbloat, not capacity problems.

When I load your model and enable “Preferences > Features > vertex-object-renderers”, the Preview pan/zoom/rotate goes much faster, and “vertex-object-renderers-prealloc” seems to reduce the Preview computation time.

I was using voip.ms last year when they were DDoS’d for over a week, by a group demanding payment via anonymous crypto. The DDoS ended when they switched to CloudFlare (which was probably pretty difficult because they’re a SIP provider.)

Almost any website with a small number of servers is vulnerable to this attack, which happens to be great business for CloudFlare. I wonder which companies are most effectively competing with CloudFlare?

independent X and Y accels

This requires patching Klipper, right? I don’t think I want to bother with maintaining a fork.

Though I designed a Z axis brace and plan to add some X/Y linear rails.

Android still doesn’t support DHCPv6 and will be left without a valid address.

RFC 7934 explains their reasoning, though it’s not exactly an ironclad argument.

Learning 3D modeling is slow process, but there are loads of free models online.

I learned how to use Fusion360 a couple months ago by asking GPT-4. The key is to start with an object you want that’s simple enough to describe, and ask lots of questions.

How do you know about the secret planet at L3?