When Raspis first came out they were genuinely a massive innovation in having small computers people could learn on or build hobby projects from very cheaply.

Over the years this has changed as they have gotten expensive to the point of being out of this price range these days. There are cheaper alternatives.

I still think they are pretty cool pieces of technology for what they are but probably more niche than they used to be now.

for the current price, yes

when you gould get a decent tiny computer that is capable of real tasks for $60 it was great. $200 not so much. especially when you can get a used enterprise thin client for the same price

Agree with the revolutionary access to simple computers for hobbyists.

I haven't used them much myself but now that I am thinking more about automation at home I'm realizing there are much cheaper and safer (lower current) options for things I want to achieve as compared to electronic or electromechanical control systems I am familiar with from industrial settings.

I think Raspberry Pi played a big part in creating the environment where there was a market for all this other innovation.

I managed to get one second-hand from a friend who had upgraded. It was my first entry into self-hosting and running 24/7 automations, like public media archival scripts. Due to the popularity and forum support, it was pretty low-friction which allowed me to dive in easily without a mentor or excessive barriers. I had no hardware experience beyond assembling a PC from parts, so the accessibility was what gave me confidence to start.

A mini-PC would have been far more than needed, to the point it's wasteful for this kind of work, and the better embedded options (I'm sure there are plenty, and cheaper) were just too intimidating or manual to have drawn me in.

Esp8266/Esp32-gang-with-asneeded-add-ons riseup!

Ancient cat at least walked into your home and killed pests

Yeah, I gave up on Raspberry Pi when they prioritized volume commercial customers over long tail users like me that needed just one.

The real question is "what's next?" The niche is "a single board computer for hobby or technical experimentation". So:

• low cost
• low power
• as much compute as you can get for that low power
• built-in standard peripherals, ditto above — with respect to radio you can get for that low power
• extensible (I/O pins are nice but what's really needed is a standard bus people can build on)
• wide support infrastructure (IDE, libraries, circuits, hats, etc.)
• hate to say it, but "AI ready" in the hushed tones of people running 28.9M parameter Large Language Models (LLM) on an $8 microcontroller (ESP32-S3) https://github.com/slvDev/esp32-ai/tree/main to push the inference to the edge

I'm thinking it will be RISC-V and not based on proprietary System on a Chip (SoC) technology.

I'm keen to learn what readers think it will be.

Change my mind.

It's cool because it draws next to no power at all.... Unless your specific use case needs to survive on a bare trickle of electricity there is no reason anyone should be buying a pi over an off-lease 1L mini-pc

it's small and low power so I stuck it above my fridge and it runs my Home Assistant with a bunch of doodads plugged into it (albeit it's trash that they went with the SD card deal because every so often I just have to swap the card and restore from backup)