Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

The quiet resurgence of RF engineering

atempleton.bearblog.dev

Post reply on HN
Print view View original

61–70 of 160 posts

Re: The quiet resurgence of RF engineering

#61

Post by sitzkrieg »

girvo wrote: post #44
qazxcvbnmlp wrote: post #43

Earlier quoted context omitted.

I can assure you the same thing is coming to fw and ee as well. If your in hw because you like building stuff its going to be a fun time to be alive, but there will be large chunks of engineering work that go away.

Oh I'm well aware. It's lagging behind "pure" software, but is catching up. The only thing I can see that acts as a bulwark is liability, bascially. The FW work I was doing requires a human and a large amount of careful review before acceptance. The "throw slop at the wall" that my current job is okay with won't fly there. But there's _lots_ of FW jobs in consumer gear that is already filled with god awful slop, so m…

i’m sure all auditing and qa reasons for human in the loop firmware development will magically go poof soon as well

Re: The quiet resurgence of RF engineering

#62

Post by sitzkrieg »

rhave wrote: post #4

Going into the RF field myself, I've been troubled with the license costs of tools like HFSS and CST. After a brief test of the open source tool OpenEMS I've landed firmly on the newer open source tool EMerge ( https://github.com/FennisRobert/EMerge ). It's a little rough around the edges still as it was released in the fall. But I've already gotten good results from it designing my own RF hardware. Apart from that I…

not to mention the price of a remotely usable vector network analyzer will make any software look affordable

Re: The quiet resurgence of RF engineering

#63

Post by dublinstats »

WarmWash wrote: post #47
dublinstats wrote: post #23

Earlier quoted context omitted.

I think the real deciding factor is government policy. So far they have favored software and services companies, letting them eat the lunch of the hardware producers.

The reality is that software is valued like it is hardware, but has a teeny fraction of the input costs and running costs. The government didn't have to do anything, investors naturally ran to the software "copy+paste" money printer. Build it once with only labor costs and then copy for nothing infinite times. To build a $100M software company you need 5 capable friends and a cloud account. To build a $100M hardware…

The government heavily revised and reinterpreted patent law many times in favor of software companies starting in the 80s. Otherwise hardware companies would have the only real moat since, as you say, software is relatively cheap and fast to produce.

Re: The quiet resurgence of RF engineering

#64

Post by sitzkrieg »

commandlinefan wrote: post #24

Unfortunately for us software types, somebody with an EE degree can go into software and then pivot back into RF engineering. I doubt that somebody with a CS degree could (as in, I think they'd be intellectually capable of it, but they'd never get hired).

i have done firmware development, pcb design and light ee work and basic rf and antenna design professionally with zero degrees. like anything else it’s a matter of getting your foot in the door first, then outperforming the academic nerds with zero experience

Re: The quiet resurgence of RF engineering

#66

Post by jmarbach »

You're right, and Hubble Network (https://hubble.com) is proof. We make commodity Bluetooth LE chips talk to satellites. That required antenna design and link budget work no software abstraction would fix.

Hubble does the deep RF work so that firmware engineers building IoT devices never have to.

Re: The quiet resurgence of RF engineering

#67

Post by lifeisstillgood »

>>> The underlying physics (electromagnetics, thermodynamics, materials science, manufacturing tolerances) don't reduce to algorithms. You have to build intuition for it, and that's not something you can shortcut.

Just wondering how an LLM replaces that job …

Re: The quiet resurgence of RF engineering

#68

Post by jacquesm »

I've come full circle, and it is amazing how much has changed since I've last worked with anything in this field. Frequencies and a degree of precision and insight that you could have only dreamed of on a normal person's budget in the 80's are now easily attainable and combined with some knowledge of software there isn't a whole lot you can not do that you can think of, as long as it is physically possible and you have the time to spare to implement it. Still, it's hardware, and debugging is an order of magnitude harder than debugging software, so you have to prepare for that, as well as to make sure you get very close to being 100% right on your first try, respins in hardware are - unlike software - very expensive and can easily kill you. And in a way that's good, and it would be much better still if the software world was somehow forced to stop shipping halfbaked stuff.

Re: The quiet resurgence of RF engineering

#69

Post by ted_dunning »

rhave wrote: post #4

Going into the RF field myself, I've been troubled with the license costs of tools like HFSS and CST. After a brief test of the open source tool OpenEMS I've landed firmly on the newer open source tool EMerge ( https://github.com/FennisRobert/EMerge ). It's a little rough around the edges still as it was released in the fall. But I've already gotten good results from it designing my own RF hardware. Apart from that I…

Tidy3d works very well (not free).

I gave up on OpenEMS.

Post reply on HN