5 ms·

I've never gotten the obsession with prescribing form factors to breakout boards. What do I lose buying a generic breakout board for half the price of a brande

I've never gotten the obsession with prescribing form factors to breakout boards. What do I lose buying a generic breakout board for half the price of a branded shield/hat/wing and hooking it up on a $2 piece of protoboard or with some jumpers?

dimatura 5y ago
Time and effort?
wdfx 5y ago
You haven't lost anything. It's about what others can gain. Meaning, those who don't (yet) know how to make their own boards or dont care how the interfaces work can still plug a couple of boards together and make a working system. The reductio ad adsurbum on this would compare these boards to PCI cards for your PC - you wouldn't expect the majority of PC users to have to hand assemble their GPU would you?
- skybrian 5y ago
  Actually there are more beginner-friendly standards in the hobbyist electronic space. Using an i2c plug standard like Sparkfun’s Qwiic QT / Adafruit’s STEMMA QT (which are mostly compatible) avoids soldering, but you’re limited to the i2c parts that they have breakout boards for. https://learn.adafruit.com/introducing-adafruit-stemma-qt/sparkfun-qwiic https://learn.adafruit.com/introducing-adafruit-stemma-qt/sp...
- BoorishBears 5y ago
  Sounds like you misunderstand what a breakout board is? Like it's literally the shield minus a set form factor. It takes 0 more knowledge. It's still plugging boards together, still runs the same code. Just not in a shape that will cover up every single pin you have. Which is something that makes things harder for beginners if anything.
  - dimatura 5y ago
    I don't think it takes "0" more knowledge. Maybe 2-5% more knowledge? (In some made-up subjective scale). If you buy a feather and a wing you know they'll work together (I haven't looked up every feather-wing combination, but the only exception I've seen is large OLEDs which technically work with low-powered feathers, but not very well - and they spell this out clearly in the product description). Not necessarily the case for a random MCU and breakout board, and figuring out needs at least a little bit of knowledge. Same for connecting the boards. There's pretty much one way to connect a feather and a wing so you don't have to think about 5V/3V, SPI pins, RX/TX, etc. Or weird small connectors. Again, not super hard to figure out with arbitrary breakouts, but requires more than 0 knowledge.
    - BoorishBears 5y ago
      Their form factor places literally no role in their compatibility. If with a Feather/Wing combo you don't need to worry about 5v/3v, if there wasn't an enforced shape that would not change in the slightest. > All Feathers and Featherwings use 3.3V logic - so do not make anything that requires 5.0V logic. You must accept and emit 3.3V logic on any Feather/Wing (except for Analog inputs) Like, the literal shape of the PCB has nothing to do with their compatibility, rules like do. Same with SPI pins, RX/TX, etc: > Bus Pins > To make it easy to create add-ons, we have fixed inter-chip bus pins: - The main change vs female headers is now you're forcing beginners to choose something that locks up their hardware. This isn't a new thing, beginners since Arduino days have been overwhelmed by how shields take over all their pins and often run into clashes. It's less of a burden if anything to have proper connectors or simply well marked female headers, which newbies can then easily follow between boards. I mean if even if you're experienced with electronics just try following which pins a board is actually using. You'll need to look up a PCB schematic. That's infinitely less beginner friendly than simple color coding and fixed connectors.
skybrian 5y ago
Form factors matter for electronics that need to fit into small spaces. It’s a tradeoff between size and ease of assembly. Also, vendors want to make their microcontroller boards compatible with their breakout boards, which means supporting MxN combinations without having to do testing and write documentation for all of them.
- BoorishBears 5y ago
  You're actually making my point Right now I can get the M0 mentioned in the article in any size from "a little bigger than my thumb" to the size of my hand. The breakouts are almost universally smaller than those common denominator shields since they don't have to cover up every pin on the board. But I don't follow your second sentence. You still need the same amount of testing and documentation with the above approach. All you've changed is the connection style from "jumpers into headers" to "pins into headers". The most you save is "plug wire X into pin Y", but in exchange you had to add 15 odd extra pins to the breakout to satisfy the packaging requirements. Not exactly saving effort there...

dimatura 5y ago

Time and effort?

wdfx 5y ago

You haven't lost anything. It's about what others can gain. Meaning, those who don't (yet) know how to make their own boards or dont care how the interfaces work can still plug a couple of boards together and make a working system. The reductio ad adsurbum on this would compare these boards to PCI cards for your PC - you wouldn't expect the majority of PC users to have to hand assemble their GPU would you?

skybrian 5y ago
Actually there are more beginner-friendly standards in the hobbyist electronic space. Using an i2c plug standard like Sparkfun’s Qwiic QT / Adafruit’s STEMMA QT (which are mostly compatible) avoids soldering, but you’re limited to the i2c parts that they have breakout boards for. https://learn.adafruit.com/introducing-adafruit-stemma-qt/sparkfun-qwiic https://learn.adafruit.com/introducing-adafruit-stemma-qt/sp...
BoorishBears 5y ago
Sounds like you misunderstand what a breakout board is? Like it's literally the shield minus a set form factor. It takes 0 more knowledge. It's still plugging boards together, still runs the same code. Just not in a shape that will cover up every single pin you have. Which is something that makes things harder for beginners if anything.
- dimatura 5y ago
  I don't think it takes "0" more knowledge. Maybe 2-5% more knowledge? (In some made-up subjective scale). If you buy a feather and a wing you know they'll work together (I haven't looked up every feather-wing combination, but the only exception I've seen is large OLEDs which technically work with low-powered feathers, but not very well - and they spell this out clearly in the product description). Not necessarily the case for a random MCU and breakout board, and figuring out needs at least a little bit of knowledge. Same for connecting the boards. There's pretty much one way to connect a feather and a wing so you don't have to think about 5V/3V, SPI pins, RX/TX, etc. Or weird small connectors. Again, not super hard to figure out with arbitrary breakouts, but requires more than 0 knowledge.
  - BoorishBears 5y ago
    Their form factor places literally no role in their compatibility. If with a Feather/Wing combo you don't need to worry about 5v/3v, if there wasn't an enforced shape that would not change in the slightest. > All Feathers and Featherwings use 3.3V logic - so do not make anything that requires 5.0V logic. You must accept and emit 3.3V logic on any Feather/Wing (except for Analog inputs) Like, the literal shape of the PCB has nothing to do with their compatibility, rules like do. Same with SPI pins, RX/TX, etc: > Bus Pins > To make it easy to create add-ons, we have fixed inter-chip bus pins: - The main change vs female headers is now you're forcing beginners to choose something that locks up their hardware. This isn't a new thing, beginners since Arduino days have been overwhelmed by how shields take over all their pins and often run into clashes. It's less of a burden if anything to have proper connectors or simply well marked female headers, which newbies can then easily follow between boards. I mean if even if you're experienced with electronics just try following which pins a board is actually using. You'll need to look up a PCB schematic. That's infinitely less beginner friendly than simple color coding and fixed connectors.

skybrian 5y ago

Form factors matter for electronics that need to fit into small spaces. It’s a tradeoff between size and ease of assembly. Also, vendors want to make their microcontroller boards compatible with their breakout boards, which means supporting MxN combinations without having to do testing and write documentation for all of them.

BoorishBears 5y ago
You're actually making my point Right now I can get the M0 mentioned in the article in any size from "a little bigger than my thumb" to the size of my hand. The breakouts are almost universally smaller than those common denominator shields since they don't have to cover up every pin on the board. But I don't follow your second sentence. You still need the same amount of testing and documentation with the above approach. All you've changed is the connection style from "jumpers into headers" to "pins into headers". The most you save is "plug wire X into pin Y", but in exchange you had to add 15 odd extra pins to the breakout to satisfy the packaging requirements. Not exactly saving effort there...