They are between builds.

B41 stable has been out for ages with multiplayer support. It’s the current version and what you get if you don’t opt in to experiential builds.

On unstable builds, the devs remove multiplayer initially until they think it’s good enough. B42 unstable is in active development and just recently added multiplayer support.

B42 is still a bug-ridden crapshoot though, stick with 41 if you want to play online.

Aye it’s on the list to try & potentially swap out when time allows. Probably over the holidays - no work until the new year after the 23rd, so no excuse really :)

It’s fine, did the job for me at the time. Just wanted the ad and nasty blocking. Keeping it and the filters up to date is easy.

Now have a pfSense box with pfBlocker-NG, which does essentially the same thing. Also runs Snort as an additional layer, and makes penning in IoT stuff possible.

Well it’s exactly what I asked for, can’t argue 😂

We settled on Puck for the name. It rhymes with what’s said when weird noises come from its direction…

Why not. I would love to see your detached robotic ass.

I printed special little stilts to raise one of our tables the 10mm extra it needed to be able to go under without getting wedged. It still tries fornicating with the cat tree - a different problem to solve.

Never lost it’s ass though, one to look out for there.

It would definitely be a size thing for adding Ethernet (PoE or otherwise) to small boards like these. The ones I am using are already bigger than they ought to be - the bottom half is just a glorified serial interface and power input for USB. The esp plugs into this through pin/header. If I were less lazy, they could be about half the thickness in a final product. No PoE I suppose also keeps them cheap, which is always good for me. The casings were my first ‘proper’ design and entry into resin printing.

The Tapo kit I have found to be a good balance of price, features and quality. I have a Tapo C310 mounted outdoors at another building, which has done great in all weathers. Initial setup does require the app/service last time I checked, but it can be made to serve RTSP locally after that. Very good for the ~£30 price point.

Eggs, baskets, and lessons being taught about how it’s not a good idea to heavily skew their ratio.

Ok so the combination is:

• This camera board
• This external antenna
• This project
• A shell I designed myself in SketchUp (skp download). Note that’s not the final version, as I lost some design files.

And the finished item:

All assembled, they will give a decent enough feed to frigate for the basics. Just don’t expect miracles in the resolution or framerate departments. 3fps does fine for my use case of tracking critters.

Gladly. I’ll collate a few bits later - time for work.

New to me & bookmarked. I am sure I have some crap lying around that this would work with.

Thank you!

For hardware, anything that can provide a local rtsp stream is a good place to start. I run cheap and cheerful mix of tapo, unbranded and homebrew esp32 cams. Offload the motion/object detection and alerts to something that can pull in the feeds, and isolate the cams to local network only.

WiFi usually ok, but at least hardwire the power to save future grief.

Using frigate to manage mine, which is running under Homeassistant - another project worth looking up.

A few images, featuring Freddie the visitor:

There’s a world of difference between day 1 and the current build as well. Definitely take another look if you haven’t recently. Devs finally waking up haha.

Favourite part so far has been the lighting system & ragdolls. If you whip a car fast enough in reverse and hit a sweet spot on the swing around, you can send Z flying like baseballs.

B42 multiplayer can’t come quick enough…

Have you tried unstable in SP yet?

To add, see here for a continually updated litany of fail:

https://static.xtremeownage.com/blog/2022/reasons-to-avoid-cloud-based-automation-products/

Hardcore, I love it.

Insane. At the core it’s no more than an extension lead with a shaped plug. Some basic brains for safety.

If I had to go that route, I’d wall it off from the main network/internet at large.

Sometimes they’ll even remove these kinds of feature. See: Yeelink Yeelight removing LAN control from their bedside lamps, as a particularly egregious example.

The goal for me would be to not have to break out the programmer in the first place. The same way I have never felt the need to operate on my toaster, fridge, TV etc.

On a ‘good’ device, having that relatively easy to access is still a bonus though :)

If a device relies on any kind of external service to initially set up or function thereafter, do not buy. Regardless of brand.

Or accept that it has a finite lifespan that you cannot control. It’s not a matter of if the rug will get pulled, but when.

There is a grey area for things that can be reflashed or rebrained, but I prefer to not rely on this. Local access methods like ZigBee, Z-Wave and 433Mhz are immune to this kind of enshittification by design. Even WiFi devices can fit in here, with appropriate restrictions in place.

An acceptable middle-ground would be for EOL devices to be offered (with a big disclaimer) a final update that removes the reliance on the service but retains the core function. That’s a pipe dream though.