Somehow, that have never scared me. I see it more like a chess game or maybe crosswords. If something doesn't work, it's an interesting problem to try to reduce the problem into smaller problems until error sources (there may be many) will start to show up.


Yes, I definitely recommend dev-kits. I just note that even $50 can be a very significant amount of money for some peope, and that what we see as basically "for free" is what other people dreams of having.


My first own designs had external EPROM, external SRAM, external UART, external PIO, ...

Todays components may have extremely fast rise times. But most of the time, that doesn't matter since you can do so much with a microcontroller with all critical parts inside. It's enough that you get the oscillator running - and most newer microcontrollers allows you to select between internal RC oscillator or external crystal. Bouncing from very fast inputs/outputs doesn't matter when you drive a relay or a LED - you can't see the bounces. And handling inputs is no different from handling debouncing of keyboards.

It's when people start to play with sequence-controlled external components (shift registers, counters, latches, ...) that the rise times may hurt. But for most hobbyists, the rise times will just make your device behave like a radio transmitter. And a hobbyist doesn't care too much about any noise emitted from their project. I didn't care about how wide my FM transmitters did transmit. If I could transmit music to my FM radio, I didn't care about if I affected my neighbours receivers.

Having just a single processor also means that most hobbyists can get by with a reasonably good VCC and GND to the processor. Many processors can work between 3 and 6V, giving a bit of safety margin from bouncing ground and missing or badly placed capacitors on VCC.

In many situations, a professional with a good scope may see things that would make him cry. But the device would still work well enough that the hobbyist would be happy.

But back to the issue. The important thing when discussing with hobbyists, is to try to tell them reasons why their life will be simpler with a dev-kit. We need to motivate them, because there may be a very significant economic barrier involved, that we don't think about. We just think "it's just a $25 cable" and think how it saves us time. If you can afford $10/week for your hobby then it's 2.5 weeks of savings needed. That changes the perspective. It's people looking from that perspective that we must try to motivate. Better they save a couple of weeks more, and get something running, than takes the cheaper route and gets a wreck at the back of their desk drawer.