[HN Gopher] The Software Behind Silicon
___________________________________________________________________
The Software Behind Silicon
Author : hasheddan
Score : 62 points
Date : 2024-05-11 11:50 UTC (1 days ago)
(HTM) web link (www.acquired.fm)
(TXT) w3m dump (www.acquired.fm)
| resource_waste wrote:
| These are one of those things that make you wonder 'why does
| anyone (real) Engineer when you can be a Programmer?'
|
| Its more economical to automate the engineering work with
| software/programming.
|
| (Or if its not more economical, you wouldn't automate it)
|
| Its not like there is a lack of engineering work to automate.
| Until we reach some sort of assembly-line-steady-state like we
| have today, I don't really know why someone would be a (real)
| engineer when you can make more money as a programmer. Real
| engineering is significantly harder physically, mentally, timing-
| wise, etc... I don't see any advantage. I will never be going
| back. I just automate now.
| alephnerd wrote:
| Even Morris Chang has pointed that out [0]
|
| The whole reason by the semiconductor fabrication industry
| consolidated in East Asia was due to low operating costs
| (thanks to massive government subsidies) and relatively low
| wages for line engineers - you're at best earning $50-70k at
| TSMC as an IC, despite most likely having some graduate
| education.
|
| This is why so many Taiwanese engineers immigrate to the US or
| Mainland China, the former for higher wages and better quality
| of life and the latter for extremely high wages (no taxation on
| expats)
|
| > automate the engineering work with software/programming
|
| Depends on the problem space. Some domains in Engineering do
| not lend themselves well to automation (eg. You will always
| need a hands on civil or power engineer when building a grid,
| and Petroleum/Chemical Engineers in the refining space will
| have to be handy if they're working at the plant)
|
| [0] - https://www.politico.com/news/2023/02/14/taiwan-tech-
| king-pe...
| tejuis wrote:
| BTW that was a very interesting interview... I kind of agree.
| However, in order to automate something, you need to know the
| subject. For example, in digital circuit design, we use these
| EDA tools, but on top of that we create a lot of custom tools
| to automate all kinds of phases of the process. You definitely
| have to know the field you're working on.
| ip26 wrote:
| The advice goes, _Fall in love with the problem_. Well,
| sometimes you fall for a problem that isn't programming.
| d_sem wrote:
| This brings up a classic distinction between a "Software
| Engineer" and a "Programmer". Without getting into a diatribe
| about semantics: my industry selects for people with software
| engineering skills. The ability to simply write code is not as
| valuable as the ability to problem solve, make thoughtful trade
| offs with known constraints, and design effective high quality
| solutions.
|
| Now readers may understand "Programmers" to also have these
| qualities. Fine. I'm not going to argue.
|
| But the same arbitrary distinction need not be applied between
| "real" engineers and programmers. Many real engineering are
| programmers.
| RicoElectrico wrote:
| Those who have been at Synopsys for a while know that it kinda
| lost its soul when the old guard retired (Chi-foon, Trac, Jan,
| Aart). Those were much down-to-earth than current C-suite who
| spews corporate bullshit, does weird layoffs and in general is
| more cozy with wall street than its own workforce, drinking the
| kool-aid of unsustainably growing valuation.
| aworks wrote:
| I worked at Synopsys for 11 years until I retired. Very diverse
| types of software engineers. I was in the hardware IP group
| (versus say the EDA or security businesses). My team had embedded
| software engineers, Linux kernel hackers, compiler developers,
| machine learning/AI tool developers and Java GUI programmer, all
| with some knowledge of underlying processor hardware. T
|
| The profile of those engineers was different from the EDA
| software developers. And the Coverity/security teams were rather
| different from all of the above.
|
| I particularly enjoyed my senior architect helping our hardware
| IP developers improve their RTL develpment via teaching good
| sofware engineering techniques and by building a Javascript-based
| tool for helping to automate writing of Verilog before it was fed
| into the Synopsys Design Compiler.
| gravescale wrote:
| The last specific silicon design tool I used was Electric, nearly
| 2 decades ago. It was a usability Superfund site, but I've also
| had to deal with things like Cadence since and my general
| impression of these "super-pro" tools in specific spaces (ok,
| well Electric isnt that) is that usability and stability is
| completely dreadful. Like, I'd probably Maybe they're powerful
| when you know exactly how to use them, and their quirks are
| burned into seasoned engineers' muscle memory, but if you're not
| living and breathing them with a greybeard to explain their
| vicissitudes, you're going to have a frustrating experience.
|
| Once wonders what the software that drives the floor in places
| like TSMC (and if anything is worse than design tools, it's
| internal industrial control software) is like and how well it can
| be transplanted to a new workforce.
___________________________________________________________________
(page generated 2024-05-12 23:01 UTC)