They're more fun. The programming is easier (although you can get an Arduino like experience on a ESP32). They have 5V options, which make some projects easier without having to add additional components. The ESP32 API (and the Pico for that matter) are better suited professional programmers.
An Arduino is better if you're doing something and want a quick, easy, simple to program controller. It started as a way for artists to add MCUs to the projects without having to become embedded programmers.
That's like the cost of two burritos. Unless you're bricking these things on the daily why would $1 vs $40 be the deciding factor for a project that is tens of hours at a minimum?
How so? I have a product that you can buy that runs on an ESP32S3[1]. They work very well and you can even do OTA updates. Even my competitor uses an ESP32 :)
Yes, regardless if you are using a pre-certified module/parts or not you need to CE certify your product as a whole. However if you use pre-certified modules the testing is cheaper/less complicated.
I do my certification testing in China by a reputable lab which is much cheaper than doing it here in Switzerland (at least 15k USD). At a minimum expect to spend 1000-2000 USD if all goes well.
There is a workaround for CE but it's a bit of a dirty trick. If you are not expecting to sell very many and your target audience are tinkerers then you can sell your device as a kit. There must be assembly that the end user has to do but they are then the ones "putting the device into the market" and they take on that responsivity of CE. That basically means they can't sell it unless they get a CE. Such an example is https://www.clockworkpi.com/ which sell their products a kits.
peripherals are a dumpster fire. I usually have to resort to bit-bang if i want to use the peripherals in a slightly different way than intended. Way easier than figuring out all the "drivers" thing to see if it's actually possible. Also, the couldn't make a decent ADC if their life depended on that. I use them for projects that require radio and GPIO, or slow PWM, or low accuracy ADC, or screens (but only with the preapproved screen controllers)
Be a little careful on those. It depends on what you're doing. Some of them are not suited to be used with the high data rates for I2C, or I2C only at 100khz. I found out the hard way with some of the SparkFun level shifters, years back.
You need to do a little research. It will usually tell in the spec sheet. Which is why the Arduino is useful. You don't have to buy a level shifter. You don't have to read a level shifter spec sheet.
The fact we don't have viable western competition for Espressif is likely to become far more of a headache than all the angst about AI GPU production.
Where can you get a half decent microcontroller with wifi integrated on it? Espressif. All the others are flat out bad in some very important dimension, which isn't to say the Espressif products are perfect, but they fit in the important ways.
Yea... ST, Nordic etc have been sleeping on the Wi-Fi, letting Espressif corner that market. They both now have standalone Wi-Fi ICs, but no MCUs still; and it took them a while to release the ICs.
It’s a shame. Nordic’s chips blow the ESPs out of the water in terms of power consumption. You can get an nRF bluetooth dongle to run for months/years off a coin cell, almost without trying. Getting an ESP32 to behave is much harder
IIRC their standalone wifi chip is pretty good even… just stick them together already c’mon.
I think Nordic etc. are resistant to telling people "if you want to use Wifi you must use FreeRTOS" or equivalent, so they push the two IC solution instead just so their Wifi stack is partitioned physically from the rest of your system.
It just pushes more integration headaches downstream to the customer, in addition to being inherently costlier. Espressif had the core right idea there, even if it's not the right decision for all designs.
I think they do this though... but it's Zephyr instead of FreeRTOS.
What I want is a Wi-Fi radio that just works like a normal part. No RTOS requirement. No framework or software libs required. Read the datasheet and go.
For some context: This is how LoRa radios work, and this is how Esp-Hosted (Official firmware from Espressif that turns the ESP into a radio IC "coprocessor") works.
MCHP has been slowly coming up with decent radio devices, finally.
If you don't use the radio going bare metal is basically effortless, if you need to use the radio the dev Tools are actually improving, though they are still nowhere as good as IDF in hiding the ugliness.
Of course they are more expensive (not much more, really, compared to simillar specced ESPs) but they are western and the peripheral actually work as intended. In my projects with ESP32 i had to basically bitbang every peripheral that i needed to use beyond their simplest mode.
Built-in no, but the Pi Pico W is decent and inexpensive if the form factor isn't an issue. The RP2040/RP2350 are nice chips to work with and documentation is good. I can live with an external module, and it's certified too.
Have you tried it? It's simply not in the same league of battle tested as the ESP one is, and I will happily agree almost everything else about the RP based ecosystem is superior.
Yes, I've use them for ESPHome and other small jobs like lighting controllers, but not for production. They're cheaper than most Arduino or hobbyist breakout boards like Feather. I can't comment on battle-tested, but I've also bought some pretty shoddy ESP breakouts in the past and I've had trouble with unstable WiFi performance when I've meshed them. The PIOs are cool, and better documented than Beaglebone/TI (maybe that's improved). Toolchain is also decent.
I would probably go Atmega otherwise. It's rare I need something in the gap between 8-bit and a dedicated Raspberry Pi. And I'll take some rough edges to support a local company (though for transparency I do hold some stock in RPI).
