Exploring different microcontrollers less than $1
jaycarlson.net
jaycarlson.net
I tried "best microcontrollers reviewed", "best low-cost microcontrollers", "survey of microcontrollers", and a bunch of others. Nothing that Google finds in the first 10-20 hits is as good this resource, other than Wikipedia, though very good, isn't a technical review.
I suppose the lessons are: there's room for lots of improvement in search engines, or SEO really works and you better promote yourself no matter how good you are, or word of mouth (Hacker News in this case) is still the way to find the best stuff.
The way I look at it: SEO really works, and shit sources tend to spend a lot on it, while quality sources tend to not spend anything on it at all - therefore a lot of quality stuff is hard to find, unless you can hear the right words from the right mouths.
(IMO SEO, beyond making your website not look like shit, is user-hostile behaviour. It games search engines into serving users links to those who spend effort on gaming search engines instead of links to content the user was looking for.)
But that's dangerous, as search engines evolve, change behavior and occasionally penalize this kind of behavior. That's why I always tried to explain to customers they should just focus on good content for humans and not care about explicit SEO after they claimed someone had told them that Google had recently "changed their algorithm" and now did XY.
Google is really good for general information, but really bad for domain specific information. Piglet (in contrast) is really good at domain specific, but can be poor in the general case (although I haven't seen that yet). We track discussions around the articles, as opposed to linked pages. This typically provides more information. You can read about it on our about page.
For instance, this article appears right at the top of the search results: https://imgur.com/a/AkUTb
I recommend reaching out to me if you're interested - we are currently doing a private beta. Spcifically, the application is for investing (we are building an AI financial advisor), but I personally use it as a search engine as well.
Google video search is also weaker than a pure youtube search for me too, so much so that I only use youtube search nowadays which is perhaps their intent...
It will pickup your search is technical and embed the best stack overflow answer it could find. Sometimes with a code snippet.
I remember how revolutionary late 90s Google was in how it would show you exactly what you were looking for, or pages with relevant content, whereas the other popular search engines was just be a jumble of stuff.
It seems like in order to monetize search, Google has had to regress what once worked well.
Should probably exclude some operations like "inurl" or simply searching the entire page url.
You can also see the ugly results of their new ranking system any time you search for a simple answer that deserves maybe two sentences, and are greeted with ten search results all of which are a dozen rambling, content-free paragraphs, with a title closely matching your search and with your answer buried somewhere inside (maybe). That reads as "high value" and "high relevance" so google eats it up, even though it's 100% grade-A garbage content.
Your proxy-search would be a solution but then I wonder what "indexing" would mean to Google
"tinyavr vs pic24" or even "tinyavr vs" and see what google suggests is a competitor.
My guess is that it's a new site? Only a few articles, and Open Site Explorer reports a Domain Authority of 18, which is approximately equal to zero :)
However, if the author keeps writing these articles at this level of quality, chances are in a year it'll dominate relevant SERPs.
An ordinary journalist could not write this article. It would take hundreds or thousands of hours to gain the domain knowledge required to understand what needs to be written up and described - it demonstrates a deep understanding of topics like ISR latencies, peripheral compatibility, memory and instruction requirements, and other stuff that really matters to engineers but is typically beyond the reach of a casual reader. As a result, news in the field is very, very rarely unbiased, informative reviews - there are a lot of 'new product' press releases unsubtly promoted as 'articles'.
An embedded systems engineer at some company building a product using these microcontrollers would not write up this article. Not only is the comprehensive knowledge required uncommon even among skilled engineers, it would take dozens of non-billable hours and would make it easier for competitors. And we typically have very narrow domain expertise, so the ability to create a website and rich blog experience like this would be unusual. (The database issues the site had under load this morning are excused!) Open-source and information sharing are not commonplace within embedded systems development.
And you won't be able to easily generate this comparison yourself without a lot of work. You might think that the numbers described would be front-and-center on the first page of each datasheet for these microcontrollers. But you'd be wrong, or at least naive. Each manufacturer's marketing department exaggerates their numbers in slightly different ways, so the first page of the datasheet is inevitably worthless bull shit.
> As an embedded design consultant
I know that for my own part, a lot of what I blog when I blog tends to be motivated either from increasing my professional visibility or because it's something I spend time doing anyway and where I figure I it'll be useful to someone (I know I've done it right whenever I find my own blog posts in search results a couple of years later and they answer my question)
I know that this article has me completely convinced that Jay could be extremely effective assisting with any of my projects. Much more so than any Ivy League diplomas, years of experience or lack thereof, glossy brochures, testimonials, or other conventional marketing tools. And it's significantly more effective than any FizzBuzz quiz I could administer to potential candidates.
For all the foibles of business politics, expert opinion from within the organization is extremely persuasive when it's available (though an organization that really needs Jay's services because they have no internal experts would have a hard time using this article).
> Jay is available for engineering services on a contract basis
I'm not saying that's the number one goal here, I think this very thorough and informative piece goes way beyond a personal advert. But it must help (and rightly so).
So, Google needs to train its AI to recognize this instead of relying on PageRank.
We still need page rank to measure authority, though. Otherwise other AIs will soon mass produce content which looks like it "requires in-depth knowledge of a large number of microcontrollers, and the time, desire, and ability to verify, collate, and summarize all that information" but is actually nonsense. Alternatively, we trick spam producer into doing valuable work.
A search engine that only gives you the best result wouldn't even need the search results page. And that's where Google likes to sell ads.
A starting point could even be the corpus of HN articles and comments, perhaps augmented with entries from the Common Crawl archives and traditional crawling.
I tend to be interested in novel and creative work, and it seems to me that most of the sites I find most rewarding are not commercially driven.
Over the last 1.5 decade I've been seriously on-line I developed this heuristic: quality of content is inversely proportional to advertising, and also inversely proportional to page bloat.
"Does anyone remember websites?"
https://news.ycombinator.com/item?id=15631565
"Search engines like Google are indispensable, able to find answers to all of your technical questions; but along the way, the fun of web surfing was lost. In the early days of the web, pages were made primarily by hobbyists, academics, and computer savvy people about subjects they were interested in. Later on, the web became saturated with commercial pages that overcrowded everything else. All the personalized websites are hidden among a pile of commercial pages. Google isn't great at finding those gems, its focus is on finding answers to technical questions, and it works well. But finding things you didn't know you wanted to know, which was the real joy of web surfing, no longer happens. In addition, many pages today are created using bloated scripts that add slick cosmetic features in order to mask the lack of content available on them. Those pages contribute to the blandness of today's web.
The wiby search engine is building a web of pages as it was in the earlier days of the internet."
The paragraph you quote pretty much sums up my feelings, though I'm not particularly concerned with finding old school sites.
Google used to do a good job of answering technical questions, as well as finding the little gems of originality (and lots of lovely mediocrity too, of course).
I don't think the two need be mutually exclusive. It's just that what I consider the good stuff is lost in the attention seeking nonsense that the web has become. Filter that out, and I have a feeling that lots of great stuff will become a lot more accessible.
wiby is wonderful, I had a play around and found pangs of nostalgia, and stuff that made me laugh out loud. But I think the criteria of lightness of a site, for example, the amount of JavaScript used, means it will be missing many great contemporary sites that happen to use the technology of today.
Sad since it turns everything into tasteless unoriginal yogurt. But true.
There's also the tried and true distribution method of newsletters and mailing lists specific to a certain industry.
If you've been in embedded a long time, you probably know about Jack Ganssle and his "Embedded Muse" newsletter, which has been going on for about 20 years now.
As someone who runs a similar blog and is first page for hundreds of search terms, Google is fantastic at discovering niche content. You can't even being to fathom how good it is.
The article is already the 4th google result if you look for "best micro controllers under a dollar".
If anything, the title should be improved by the author. "The Amazing $1 Microcontroller" is a very poor description of the content and noone is going to google that.
That worked because you're mentioning the "$1" from the article's title. Google is not interpreting "under a dollar" as "cheap". As proof, I searched for "best microcontrollers under £1", "best microcontrollers under 50 cents", and "best microcontrollers under $2". Jay Carlson's article wasn't found in any of those cases.
A more commonsense search of "best cheap microcontrollers" doesn't find the article even in the top 100 results.
> "The Amazing $1 Microcontroller" is a very poor description of the content and noone is going to google that.
I thought the title intriguing and pithy. You don't want a web where every title is SEO-optimzed, do you?
I find the title intriguing too... but on HN and with 100 upvotes already. It's pretty terrible in a search context. If I am looking for a comparison, I'm not clicking on something called "the amazing microcontroller".
Google is really good for general information, but really bad for domain specific information. Piglet (in contrast) is really good at domain specific, but can be poor in the general case (although I haven't seen that yet).
For instance, this article appears right at the top of the search results: https://imgur.com/a/AkUTb
I recommend reaching out to me if you're interested - we are currently doing a private beta. Spcifically, the application is for investing (we are building an AI financial advisor), but I personally use it as a search engine as well.
...
Consequently, the megaAVR remains the most open-source
8-bit microcontroller on the market — by a long shot.
...
In addition, places like Adafruit [1] and Sparkfun [2] build their own boards and make most of their code free/libre/open-source [3] [4], with a large focus on Arduino.
It'll be interesting to see how the electronics market evolves when the current and subsequent generations are trained on AVRs.
I think the major point is that "ease of use, accessibility, community and the tools" are exactly where the open source software model thrives over a walled garden one. It's not impossible to do this with proprietary software but for the open source projects that are successful, they tend to be successful because the freedom they offer encourages each of those points (imo).
But until you can spin your own silicon with the ease of something like a 3D printer or buy custom low cost/quantity silicon, it doesn't matter what your open source HW design does if you don't have the money or a business to create and distribute it. That's one thing that is very hard to do with something physical and why opensource has worked so well in the software community.
There does seem to be some stuff being done here using FPGAs... But they are still a rather expensive way to go just to drop a custom architecture on for now.
In the meantime, there is still plenty of software work to do to make a truly useful and professional grade opensouce stack that you can develop on the chips we have.
But even as popular as Atmel is among hobbyists, Atmel has largely stayed out of this
space directly. Instead, they’ve secured small-volume AVR sales by relying on the open-
source community to build their own tools for themselves: turning out a slew of
hardware and software used to program the megaAVR devices.
While I applaud the efforts of these developers, these tools are inferior to Atmel’s.
Their programming speeds are terrible, they don’t support the new tinyAVR 1-Series
devices, and they have absolutely no debug capability.
> Consequently, the megaAVR remains the most open-source 8-bit microcontroller on the market — by a long shot.
and your quote:
> But even as popular as Atmel is among hobbyists, Atmel has largely stayed out of this space directly. Instead, they’ve secured small-volume AVR sales by relying on the open-source community to build their own tools for themselves: turning out a slew of hardware and software used to program the megaAVR devices.
> While I applaud the efforts of these developers, these tools are inferior to Atmel’s. Their programming speeds are terrible, they don’t support the new tinyAVR 1-Series devices, and they have absolutely no debug capability.
Don't use leading spaces to quote text, that's for unformatted text. Do this instead:
> *Text.*Ecosystem is an interesting area. It seems that everyone has converged on Eclipse, which is probably better than trying to write their own environment but can still be infuriatingly clunky at times.
I'm impressed with the breadth of this review and the commitment to including everything that fits in the $1 parameter - which is now quite a lot, I hadn't expected to see Cypress PSOC there.
Another novelty of small/cheap controllers is when I was playing with the defunct 10F2xx family I didn't own a meter sensitive enough to measure the current draw at 32 khz clock. I do now, but it was an expensive meter! In the old days cheap EE would wire up a capacitor and run the ckt off the cap and time how long the voltage drop took and use that to create an imaginary resistance and use the measured V and theoretical R to calculate the theoretical current. Of course my cheap meter drew more current than the microcontroller which leads to weird games like sample voltage for a second. Also leakage currents in most caps is larger than the uC current draw oh so much fun fun fun. One trick is to drive a square wave generator thru a diode to a low leakage ceramic disk cap, assuming your reset circuitry is faster than the square wave and you trust the DC response of your scope. Another thing I tried was creating a transfer function where a 100 watt lightbulb and a tiny solar panel generate say, 1 mA, so a 99.9% optical attenuator in front of the cell means 1 uA, right? Of course sleep current is a tiny fraction of that, so you stack a SECOND optical attenuator on the solar cell, say merely 99% blocking, and that should be 10 nA which is about half sleep current, right? You can also play games when your CPU costs 30 cents like hooking up 100 in parallel for $30 to measure 100x average current. Low current is fun if you're ever bored.
(also the article has reservations about PIC, pointing out its strangenesses and describing PIC24 as underwhelming)
This is an ESP8266 breakout/programmer for about ~$3: https://www.aliexpress.com/item/ESP8266-ESP12-ESP-12-WeMos-D...
Absolutely incredible stuff.
Ordered this one coming soon: https://www.aliexpress.com/item/WEMOS-WiFi-Bluetooth-Battery...
And in the realm of awesome breakouts that I'm excited for: https://www.aliexpress.com/item/2pcs-TTGO-LORA-SX1278-ESP32-...
There is some really incredible stuff coming out based on the ESP8266 and ESP32 right now. IMO completely blows all of the normal arduino stuff out of the water.
It looks like the ESP32 is on the back. Sparkfun also sells an ESP32 that comes in that package: https://www.sparkfun.com/products/13907
What's really interesting going on right now is that adafruit/sparkfun have become radioshack, and aliexpress has become my go to place to find interesting new stuff.
I will say that whenever I'm doing a commercial project[2], however, I still use the more expensive adafruit/fry's/genuine arduino boards.
[1]: I help run a hackerspace in Phoenix, and I do all sorts of maker workshops all over the world, so having a bunch of weird dev boards and parts and stuff is pretty important to me. If those don't end up working, it was worth the $40 to find out.
[2]: Part of my life is helping create one-off experiential advertising installations. Usually these things have a stack of arduinos in them controlling lights, sensors, motors, etc.
I've tried a few breakout boards and my favourite is the WEMOS D1 Mini. All you need to supply is a MicroUSB cable. You can get them shipped from China for under $3: https://wiki.wemos.cc/products:d1:d1_mini
Big warning I'd make, though, is that the D1 has two revisions - the R1 (rarely described as such of course), and the R2.
You definitely want the R2. However, many ebay vendors are still selling R1's.
Yes and no. ESP8266 would typically survive 5V applied on any pin, and sometimes even sort-of work with 5V power applied (although with much higher power consumption and no WiFI).
But there is a catch. Pretty much all CMOS chips (ESP8266 inclided) have body diodes from each IO pin to power/ground. This means - you can't have voltage higher than VCC+0.6v or lower than GND-0.6v on any pin. So what happens when you have ESP8266 powered from 3.3v and connected to 5V I/O is - ESP will get (partially) powered via I/O pin. Depending on what 5V part is connected to that I/O, that might be bad for that part (excess of I/O sink current). And in extreme cases (5V I/O pin that can deliver high current, or ESP is in deep sleep), this will supply current into 3.3V VCC rail, bringing it up as high as 5 - 0.6 = 4.4v potentially. Which could be bad for anything else on that rail.
As a practical example - let's say we have ESP8266 powered by a switching supply with MP2307DN [1] chip delivering 3.3v, and connected to a 5v AVR. In normal operation, ESP takes ~70mA, which exceeds 40mA driving capability of AVR, so it's all good [2]. However once ESP8266 goes to deep sleep, all 40mA from AVR's I/O go via ESP's body diode back to MP2307DN switch, which usually would fail catastrophically and end up shorting input to output, supplying 12 (or whatever the input is) volts to ESP's VCC. Which is likely to fail as well, sending 12v further and frying the whole thing.
[1] MP2307DN is widely and cheaply available on "Mini360 DC-DC buck convertor" boards and it really doesn't like current being back-fed into its output
[2] Except I/O drive transistor in AVR is overloaded, which might or might not lead to its failure eventually.
The actual smallest I've seen is one I made: https://github.com/skorokithakis/tiny-ESP8266-breakout
My second favorite ESP8266 hardware are the "Sonoff" brand WLAN mains power switches. I especially like the Sonoff S20 for its form factor, and the Sonoff POW for it's power metering capability. They contain an ESP8266 and can be reflashed with different firmware that does not need the vendors cloud services to work.
I'm pretty sure that its the work done to make these into $3 wifi arduinos that is what made them popular. That's why wemos literally sells one that is shaped like, and has the same header spacing as, an arduino uno.
So now my kids have 2 wifi-tablet-controlled trucks with very custom programming (for N-point turns, playing music through the motor windings, headlight blinking, etc.) and the ability to update the program over the air at any time. It's basically magic. I spent only about $8 per truck.
Yeah, the WEMOS D1 is the real deal. I love it so much that I bought a pile of them before I've even decided what I want to use them for.
> A new series that explores 21 different microcontrollers — all less than $1
Fom "the rules":
>While some projects that come across my desk are complex enough to require a hundreds-of-MHz microcontroller with all the bells and whistles, it’s amazing how many projects work great using nothing more than a $1 chip — so this is the only rule I established for the shoot-out.
>I wanted to explore the $1 pricing zone specifically because it’s the least amount of money you can spend on an MCU that’s still general-purpose enough to be widely useful in a diverse array of projects.
>Any cheaper, and you end up with 6- or 8-pin parts with only a few dozen bytes of RAM, no ADC, nor any peripherals other than a single timer and some GPIO.
>Any more expensive, and the field completely opens up to an overwhelming number of parts — all with heavily-specialized peripherals and connectivity options.
Of course, the ESP8266 is precisely this "hundreds of megahertz" chip (160 Mhz) with "heavily specialized connectivity options" and peripherals (wifi and, in the case of ESP8265, a full megabyte of flash memory). While all this is interesting, at the $2+ price there are "an overwhelming number of parts."
So the author chose $1 as a very hard cut-off.
I’m currently using Cypress’s PSoC line (4200M boards) for a couple of personal projects. The reconfigurable hardware and analogue parts actually kill a lot of external hardware. It’s pretty amazing and ridiculously cheap. Visual Studio is the IDE for this.
https://uk.rs-online.com/mobile/p/processor-microcontroller-...
Note: don’t just jump into these if you think it’s just a better Arduino as the learning curve is extreme. You really have to know your stuff before you open the box.
But that £8 includes next day courier delivery (which is free), a phone number you can actually call and talk to someone instantly, materials and tutorials on their web site and product support.
They do a slightly lesser dev board for £3 as well.
Hell I can order a £0.20 resistor and get it next day!
I think what we have here is a server or a web app poorly configured.
I appreciate the kind words from everyone so far; it's been a huge learning opportunity, and I hope others can get inspired to grab one of these parts (especially one of the weirder ones) and do something cool with it.
https://pixady.com/image/0jab/
Even though all the text content is present (this image is disabling CSS and scrolling down a little):
Requiring javascript for rendering what basic HTML already renders is not a very inviting or inclusive way to go with the blog.
I have to chime in. What's the point of using the UNO form factor if you don't support the shields, libraries or the environment? It makes completely no sense.
Instead of working out all the fine details of I2C, I can call a well developed function that takes care of all that for me. Instead of spending days writing a graphics library, I can find an off the shelf GFX library that has more features/is faster/ etc
The ability to move quickly with well documented, prebuilt libraries is invaluable when you want to get something up and running ASAP.
Once you're ready to develop it, then you can go back, strip out everything that's not needed, rewrite things, etc.
That's the beauty of Arduino.
If you buy them in bulk, you can get it down to less than $1 per MC. [2]
You can even install Arduino libs on it and the Arduino interface has quite good support for it as well.
Some of my projects with it include a DSLR camera slider [3], wireless home automation (turn off lights when I'm not in the room, communicate with my air-conditioner to maintain the room temperature using IR, etc.)
[1] http://www.microchip.com/wwwproducts/en/ATtiny85
[2] https://www.ebay.com/itm/5PCS-ATTINY85-20PU-IC-MCU-8BIT-8KB-...
[3] https://www.instagram.com/p/1AscllHDQ0/?taken-by=discovery.d...
http://webcache.googleusercontent.com/search?q=cache:https:/...
There's even lots of sub-dollar MCUs out there too. As the author notes, picking the best one depends a lot on what you need to do.
If answering "it depends on what you need to do" requires a lot of domain expertise to evaluate, there's definitely a place for summaries like this!
[1]: https://www.digikey.com/products/en/integrated-circuits-ics/...
I mean raspberry is cool, cheap and everything, but I'd prefer something all integrated, as small as the raspberry pi zero, less fragile (meaning I can put in my pocket or bag, and still expect it to work), not necessarily with pins.
I don't know about what the demand is for micro-controllers, but as a consumer and a developer, having a wifi-capable portable micro computer would really be cool.
Does it already exist? Under $50?
What would you want by way of input on such a device? I've been working on something like this based on TI OMAP CPUs, and the hardest thing (after a decent semi-open SoC) is getting a decent keyboard.
I think the N800 was probably the pinnacle of handheld Linux devices (or the Psion 5 series if you step outside of Linux). You could look at the Raspberry Pi handheld game console options out there, but ultimately there's little that can compete with a Pi Zero, battery pack and ILI9341 screen on price.
If you remove the Linux part from your wishlist, you can consider some "NodeMCU" boards on Chinese websites, eg: https://www.banggood.com/Wemos-NODEMCU-ESP8266-12F-WiFi-1_30... https://www.banggood.com/Wemos-TTGO-WiFi-Bluetooth-Battery-E...
I used to play a little bit with the nerdkits (http://www.nerdkits.com) and really enjoyed their approach but they seem to have gone out of business and their content has not been updated for some years..
Any other suggestions?
Edit: typos
Sparkfun has pretty much the same thing with a "learn" page (learn.sparkfun.com) and all sorts of tutorials. E.g. here's a list of tutorials/projects dealing with ATtiny variants, a microcontroller discussed in the original artical: https://learn.sparkfun.com/tutorials/tags/attiny
It's really important to have a specific goal in mind regardless of whether you prefer to learn how to solve it with structured or unstructured materials, and to have an initial idea of what level of detail you want to go in order to accomplish the goal. The lower level you want to learn the less up-to-date your resources have to be, our understanding of electricity hasn't fundamentally changed since Maxwell's equations. :) Something like "I want to turn on a smart light of this type when there is motion in this area, and turn it off after 5 minutes" is a nice goal. And you can accomplish it at many different levels of hackery, and learn very deeply if you want about each component of your solution (for instance there's the entire field of RF if you want to know how the antenna works).
I've only used the 3rd edition, but I found it very accessible and well-written. I think the 4th edition has expanded content for microcontrollers and plan on getting it when I have the time to play around with electronics again.
What I liked about the book is that it provided information in a practical way so you could "just build something" without knowing all of the theory, but the theory was available when you wanted to increase your depth of understanding.
I admit the title is cheesy, but I could see this book being used to teach an introductory class on electronics. It also serves as a good reference entry point for topics/electronics where you're completely clueless.
There's a person (fairly active on reddit, I believe) who put together a youtube channel reviewing such kits. Sadly he stopped posting new videos years ago (bored and moved on?) Think he's an EE or something comparable for work, but, in either case take a look at some of his offerings: https://www.youtube.com/user/SUBLOOP/videos
The video that I got started with and caught my attention was his "Active Speakers" kit, part 2. They're entertaining, his soldering is just hypnotic, and it covered some troubleshooting you'd run into with electronics.
I plan on trying out this $12 ESP8266-based device which ships configured as a USB serial port -- I plan on using it as a serial console. It would be awesome if I could convince it to somehow act as a USB VGA-enough and see both the boot/BIOS screens and the OS GUI.
As for VGA i think you are out of luck, or need a faster/specialized co-processor. VGA requires 3 analog inputs of at least ~30MHz, and an ESP8266 has one analog input in the low (I think I've seen 20) kHz range, so more than 1000 times too slow.
The cheapest analog-digital-converter chip I found is the TVP7002. Dealing with the digital pixel output is still way way too much data, but now it might be possible to scan the picture a few pixels each loop. The reasonable approach would be to process it in an FPGA or specialized chip and only deal with a compressed and scaled image in the ESP8266.
* does it have an LLVM backend?
* is it listed on the forge page?
* is it listed as having std support?
The farther you go down this list, the simpler it gets to use Rust for it.
Latest work has been on ARM boards and a ton of work on AVR, which brings 8/16 but support.
The best part: You can use many of Rust's high level zero-cost abstractions. Just take a look at this example code which compiles to very efficient assembly: https://github.com/japaric/discovery/blob/116fe76491d661b5ee...
The other advantage is that its I2C slave core operates autonomously- it provides a shared register space. It means that a master can write a bunch of stuff to the shared memory without waking up the CPU during each transaction (which usually involves clock stretching).
I know this is off topic but would love HN to show a page size besides each link so we know whether the link is even worth opening.
The STC8 and N76 parts are 8051, the other two are their own design. The HT-66 looks very much like a PIC16 part, and IDE and compiler are totally free.
The STM8 is probably the best-performing part in that price range, and has a free IDE and compiler.
My review includes pretty extensive discussion on the main page, plus separate reviews for all these parts — check it out and let me know if I need to clarify anything!
I made a repository for the toolchain necessary to get STM boards working here: https://github.com/abraithwaite/STM32
Although it's outdated, the links at the bottom appear to link to updated tools which are presumably better :-)
I ended up ordering the Attiny13A as that was the cheapest option in aliexpress/ebay with freeshipping in small quantities (around USD $0.3 a piece).
I would have loved to find something even cheaper!
You can do MCU based PCB designs. It gets harder when it comes to the mentioned Cortex-A. It will be harder to sample, say, 40 pieces of them. You're very much supposed (and better off) to buy boards like OrangePi or similar.
ESP designs, especially ESP32, are somewhat in between. Started off as a little documented blackbox with a RTOS interface it is now a much better (but absolutely not on par with Cortex-M designs) documented platform. However, part of it is still a proprietary code blob. They are beefy beasts, however, and when you need just that - well, it's a no-brainer. I just wish they were documented like STM32s plus ARM manuals and had a little more toolchain support. The price tag keeps me from complaining any louder.
When you don't really have hard constraints and are fine with a 5 second bootup, chose whatever fits your bill. When exact timing is an issue (and you don't want to go bare metal on the Cortex-A), when fast bootup is an issue, well - better look at MCUs.
But it still blows my mind that the equivalent (a vast superset, if you consider the GPU and specialized coprocessors) of my workstation from the late 90s now costs less than $5.
And probably no realtime clock, or well-documented ADC / DAC.
Sometimes, all you gotta do is:
1. Measure a voltage from something (external temperature sensor: such as a Thermistor: a resistor whose resistance changes based on the temperature. Apply a constant current to a Thermistor, and you'll get voltage that varies with temperature)
2. Change the voltage somewhere else. (IE: ramp a motor up or down based on the temperature).
3. Do this from ~500 to 10,000 updates per second (ie: 500Hz to 10kHz / An update every 2ms to every 100 us).
Don't be fooled by CPU specs. They're actually rather unimportant in the microcontroller world. Good features to have are like... integrated OpAmps, ADCs, DACs, Timers, and other features that allow you to do job #1 and #2 better. Even an Arduino/ATMega runs at 20MHz, giving you 2000-cyles per 100-microseconds. And that's why Arduinos are so popular, because they have all the ADCs / Timers needed to do the job, and 20MHz is way faster than what most people need.
A more "premium" chip would have OpAmps, which will offer finer voltage control and more customization for the hardware engineer. I guarantee you: if you're doing hardware (aka: reading voltages and setting voltages), you will NEED a bunch of OpAmps on the board somewhere. So chips that come with integrated OpAmps can really cut down on external parts.
Most of these cheap parts go up to 100MHz (ie: 10,000 cycles per 100-microseconds), but really don't have a need to go beyond that. Besides, higher MHz means more power usage... and a lot of these uCs are designed to run on button-batteries for literally months (assuming you get the right amount of sleep). So power-efficiency severely trumps processor speed in their typical use case.
--------------------
Anyway, consider this. If you have a 1GHz chip but your ADC converter only runs at 50kHz, then you practically can't do more than 50k updates per second. Because at best, your system is only able to read the voltage level once every 50kHz.
It takes a bit of time for voltages to settle down and make a decent measurement. I'd imagine that for most of these chips and systems, they're limited by peripherals and not by CPU power.
Oh, they're well-documented alright, over these 4000 pages of PDFs from our various portals and manuals and peripheral standards and ARM standards and SDK code and...probably a few other places. So thoroughly documented you could never finish reading it all before the chip is obsolete!