Reminds me of a comment I saw on reddit:
"Verilog/SystemVerilog are not languages many people write about online. Reading Synopsis documentation and occasional seminars are the best of what you can get."
https://www.reddit.com/r/hardware/comments/2m58ci/hardware_e...
As someone trying to learn Verilog better, it's frustrating.
EDIT: But still ... the tooling is pretty stable. I don't know that I would turn to an HN-like community to help me put a (frankly, pretty vanilla) module on a CPU bus from the Broadcom family. I know how to do that. You can go down the stack (at which point everything is proprietary) or you can go up the stack (at which everything is software). No?
From that link:
DeepChip.com is a 20 year old clearinghouse where semiconductor chip designers contribute data-intensive papers and articles of first-hand evaluations and production benchmarks of commercial EDA tools.
John Cooley at or (508) 429-4357 edits the content of both ESNUG and DeepChip.
Knowing that warms my heart. Thank you.
> hardware developers have way more at stake than their software counterparts
This is related, but software is fungible so more of the customization happens there. How many iPhone designs are there vs apps in the app store?
I'm a software engineer with interest in hardware, but I'm interested in something more than news about interesting new hardware tech. HN has more for software engineers. As someone not in the field, where do I find discussion about best practices in the industry for testing and verifying digital hardware designs? I have some project ideas about using an FPGA as a PCIe device. That seems daunting. There are a few examples out there, but they hardly teach understanding. Suitable hardware can be hand for a few thousand, well in the range of a hobbyist. Where can I go find people doing this kind of work?
I would argue that this website covers both hardware/software in the natural proportion of hardware:software developers. If I had to guess then that would be 1:30.
Why is it 1:30? Because hardware developers have way more at stake than their software counterparts. One injection mold costs $8000, one PCB assembly run costs $12000, one PCB costs $900 and one week, one wafer costs $400,000 and six months. So there are just a lot less hardware developers than there are firmware/software ones.
Look at the distribution of posts on the HN two front pages 9 Software optimization (compilers, language features) 9 Business/ IP 6 Other 5 Cutting Edge software like AI 4 Information Security /Privacy 4 Show HN or similar (a product or dev tool) 4 Historical Computers 3 Non-Technology news 3 Design 2 Other Technology news 1 Ask HN
Only 17/50 of those are actually pure software posts. (Software optimization, Cutting Edge software like AI Design) The rest would likely be on a hardware website too.
Software changes much faster than hardware. Software can be acted on by individuals and posted on HN by individuals, not just companies/universities with $1m research labs. But when new hardware comes out that is intellectually interesting, like IBMs quantum computer, or an ESP8266, or Google's Tensor Computer Units, you'll bet you can find it on HN.