I thought that the two focus knobs would be horizontal and vertical focus, but they weren’t. Not sure how they work but there are only four options to try out (increase or decrease knob 1 or 2) so didn’t take long to do.
The 21” version was supposed to be solid but I have never seen anyone with one or seen it for sale. The price tag was probably too steep compared to the faulty 17” one. Sad to hear that the graphite model was just as bad though.
Sorry, my wording was unclear. The USB RTL-SDR was for receiving, and I did the capture and analysis on my desktop. The Raspberry Pi was for transmitting using the GPIO pins.
I went into the weeds of reverse engineering my Minka Aire ceiling fans with a USB RTL-SDR. In my case the manufacturer's remotes worked fine, but I wanted to integrate it in with Home Assistant. If you're interested, you can see what I did: https://www.riveducha.com/decode-wireless-signal-with-usb-tv...
I used a USB RTL-SDR ($15) and a Raspberry Pi to receive and transmit, but I think the Flipper Zero can do it all-in-one.
Even better, why not find people with a wireless mouse/kbd dongle plugged into their computer, then send radio signals mimicking their mouse/kbd to do the same thing? That way you don't even need to touch their computer or bypass the check on whether the USB device is preauthorized.
The reason you can't is because many newer wireless devices have encryption, but how many people know or care when they buy one?
I spent a not insignificant amount of time looking into this, but for the purpose of making a game AI, not for hacking other people's computers.
The total games market is expanding at a rate where you will have more quality indie games from new game makers than you could possibly play. If Supergiant gets bought and turned into a cash grab, there will be plenty of others out there to take their place.
This reminds of old factories that had belt-driven tools running off of a shared axle hanging overhead. The axle would be powered by a combustion motor. Turning on/off your tool meant engaging or disengaging it’s belt from the axle. I’ve only seen these in museums and don’t remember what they’re called.
It sounds like otrahuevada is describing the same thing, doesnt it? Choose a commercial license if you want support, or the open source license if you don’t. That is how I understood it.
I used Blender to edit a bunch of my YouTube videos and can’t really recommend it. (Using Kdenlive now and it’s OK.)
Blender’s VSE is too unlike normal editing programs and the performance is bad as well. It makes your workflow more clunky for no gain unless you’re also mixing 3D scenes into your video.
There are ambitious proposals to overhaul the VSE so hope might be on the horizon.
I made a similar little GPT-3 toy (for Linux/bash) that also ended up generating a Rickroll URL.[0]
In vague cases, GPT-3 functions approximately as well as a markov chain - it will give you the most “probable” sequence of tokens. There’s some implementation details regarding how it deals with tokenization, but unless you really increase the randomness (“temperature”) you’re likely to get very popular YT video IDs.
In other cases, it will “hallucinate” things like URLs, UUIDs, hashes, and other things that are basically a random string of characters. In my experience it will make UUIDs that have the right number of characters but seem suspiciously non-random and don’t fit any of the defined UUID formats.
I did this for the FPS game Valorant, not for cheating, but to see if I could make a 100% computer-controlled opponent.[1]
It involved some weird Logitech mouse hacking for the control side of things and was overall really rough, especially with how much latency was involved.
With a normal USB capture device, the latency is around 50-100ms, so it’s hard to do lightning-fast reaction. Even more so if your view is moving at the same time. Everything that normal aimbots do by directly accessing the game’s memory suddenly becomes much harder when you don’t have memory access.
For anyone interested though, the computer vision models publicly available are both very good and also easy to fine-tune for a specific use case (e.g. a specific game).
I am a small time Youtuber with a monetized channel. There are two things to consider for anyone thinking about ads on YouTube.
One is that subscribing to YT Premium will get rid of all the ads. (You can’t get rid of sponsor shoutouts with Premium, though.) From the data that YouTube gives me, only a tiny fraction of my revenue comes from Premium users. I assume this means only a tiny fraction of users have Premium.
Second, if you are in YPP (monetized channel), you can choose which videos have ads and when the ads appear. One workaround I’ve seen people use is to get accepted into YPP and then use that to turn off ads. The Zig channel would qualify given its viewership.
Not advocating for one way over another, just wanted to add info.
If you want to control non-smart wireless home appliances, you can actually make the Raspberry Pi GPIO do it without the Bond Home.[1] It’s a huge hack, to put it lightly.
I don’t think the OP is actually doing any sort of control though. I personally would recommend a Zigbee door sensor as part of a Zigbee + Home Assistant (open source) automation system which gives you the flexibility of not having to run wires and not having to write code to configure the automations.
They have a very aggressive retention team to try to discourage you from cancelling. If you subscribe to (just) the crossword and tell them you want to cancel, they will offer you a lower monthly rate. Not sure if this violates California law requiring easy cancellation, but it’s another way to subscribe for less.
Flexibility is what made Jira into the monster it is! Jira can model any workflow so that any company can purchase it. If you are running into problems because transitioning issue types loses fields, why not make all types have the same fields? or have only one type? In a lot of teams, the answer is - “because that’s what the person who configured Jira made it do”.
And fulfilling laundry lists of feature requests will turn your system into a Jira - can do anything if you have someone whose job it is to configure it. Which is fine if you want to be another Jira, but that might be a tougher challenge than being an opinionated alternative.
Hi HN - looks like someone beat me to posting this here! This was my most recent weekend project, and I was curious to see how GPT would do when asked for output that a computer had to run, versus output for humans to read. The area of AI results that are wrong but wrong in plausible ways fascinated me.