The SoC builder you describe might be the MiSTer project for console hardware emulation. But even then, the number of people who develop verilog for that project is rather tiny.

However, it's the first platform that has gotten me, a long time software developer, to really look into how it works. As in, I've been working through verilog tutorials, etc. and formulating a project to build on it. Compare to when I first bought a Xilinx FPGA dev board some 20 years ago, I found that I had no idea what to do with it, and there were not many resources to learn from compared to today. I don't think it's going to be mainstream soon, but there's definitely been some progress.

Meh. HDL isn't that hard and there are good books on it. The main reason FPGA programming is unpopular is because the tools are pretty hideous. Icestorm makes things a lot better but it supports like 5 parts and they're not very powerful ones.

How cheap are the devkits, can I easily source <10 unit quantities for <$ouch, and what are the limitations?

- Fundamental questions I imagine a semi-interested tinkerer would be asking

devkits compatible with icestorm? I have an icestick that I've messed with, at 1x it costs $20 (at least when I bought it off amazon.)

> Since 2000, mainstream FPGA programming has been right around the corner.

As an avid technologist I dabble in everything from electronics to assembler. But, every time I try dipping my toe into FPGA programming I quickly retreat!

The tools are too complex and the whole process has a learning curve that far out weigh using a stock CPU/SoC.

> The tools are too complex and the whole process has a learning curve that far out weigh using a stock CPU/SoC.

Sadly this is true, although the open source tools are gaining ground pretty quickly from what I understand. https://symbiflow.github.io/getting-started.html

Accessibility of FPGAs has made massive strides since 2000. You can nowadays get people build a really basic FPGA demo in an hour, starting with a dev board and their unprepared laptop, and even I remember times where you'd struggle to get the tooling installed in that time.

(Not going to go into "mainstream", because that's a) incredibly vague (is microcontroller programming "mainstream"?) and b) IMHO not that interesting - it's fine for specialist things to be accessible but not mainstream)

Personally I think "programming" RTL is dead simple. It's so simple, in fact, that it's not even interesting. But I've been doing it forever so my perspective is skewed.

I think you are thinking of different types of tasks. A lot of RTL is brainless, but if you are actually trying to compute with it, it can get tough.

A mediocre C++ programmer can write a (bad) hash table in C++ in a few hours. How fast can you give me one in Verilog?

You're not writing a hash table in verilog, you're writing a hardware configuration that can hold and manipulate a hash table. It's a different ball game.

I have to make a lot of assumptions, but a couple hours doesn't seem unreasonable for the core function. Among the many assumptions are the hash function, what is actually being hashed, what processor interface is, desired performance, etc.