5 ms·

I’m an EE doing asic design. Everything I do for a living I learned outside of school. Aside from the basics like reading/maths etc.

throwaway613834 9y ago
Huh, this is hard to swallow. What digital EE courses did you take in school?
- ktta 9y ago
  I've had the same thing said to me by my professor about FPGAs. You can't really find much about current EE practices because companies just don't give them away or blog like the software guys do. Textbooks tend to be very general (also old) and don't focus on the tooling used which is the bulk of the work you have to do.
  - egwor 9y ago
    Tooling changes (very) regularly. Learning concepts and techniques is far better for the long term since many of those techniques/concepts can be applied to new situations.
  - throwaway613834 9y ago
    Also skeptical of this one. Tooling seems kind of beside the point? It's like arguing that your school taught you Python and that didn't prepare you for Java (or whatever).
    - ktta 9y ago
      Bulk of the work is done in simulation and verification with FPGAs.
      - DigitalJack 9y ago
        This is true, but those tools doesn’t change that often at a given company. VCS and questa probably cover 99% of the fpga market.
        ktta 9y ago
        That's exactly my point. You don't get access to those tools in school unless you are in an excellent graduate program AFAIK if that. The simulation I can do in Vivado is apparently nothing compared to the scale those tools take it to. I've talked to someone in the industry that they told me Vivado's simulation is only okay for beginner projects.
    - DigitalJack 9y ago
      Tooling is most definitely not beside the point. And if your focus is backend, that is what you spend most of your time on.
      - throwaway613834 9y ago
        > Tooling is most definitely not beside the point. And if your focus is backend, that is what you spend most of your time on. The question isn't what you spend your time on though. It's what is worth spending time teaching/learning in school. You also spend all your time using a mouse and keyboard but (hopefully) you don't take courses to learn how to use the mouse and the keyboard in school. If most people could pick it up in a short period of time on the job then it doesn't seem a worthwhile use of time to teach in a course.
- DigitalJack 9y ago
  Fair point, I had one class called logic design. It was my favorite, although I hated breadboarding 7400 logic. The class was mainly Boolean algebra with some flip flops thrown in at some point. If I recall correctly that was my sophomore year. There was next to no digital in my courses. Heavy on math and analog circuits. I think that one class on digital logic was it.
  - PascLeRasc 9y ago
    Digital logic was also one of my favorite classes. Really fun to think about some of the problems.
  - throwaway613834 9y ago
    > Fair point, I had one class called logic design. The class was mainly Boolean algebra with some flip flops thrown in at some point. Yeah that explains it. :-) So nothing about writing Verilog, feedback pipelining, adder design, clock skew, and stuff like that? Because those are the kinds of things I'd expect in a digital logic course (if one were available) and I would be shocked if you didn't need to know those concepts for a job in ASIC design.
    - DigitalJack 9y ago
      Nope. Well, I’m sure we learned simple adders, again...Boolean algebra. Everything else I learned on the job. If I had gone into board design, I’m sure a lot more would have been relevant. Many places the chip designer is also the board designer, but sadly that is not the case here.
      - psykotic 9y ago
        CE/EE departments sometimes offer an undergraduate VLSI intro course where you learn the theory from a textbook like Weste and Harris and get to design and tape-out your own chip via the Cadence or Synopsys tools and then fab it through an MPW service like MOSIS or CMP. So, it's certainly possible to learn some hands-on chip design fundamentals in school, but if your department doesn't offer it, you're pretty much out of luck. You could do your own extracurricular MPW run of a small design for maybe a thousand dollars, but you'll be stuck with SCMOS cell libraries and free tools like Magic or Electric, which aren't representative of industry.
Draco_Au 9y ago
Including solving differential equations? You needed Calculus for that? Did you teach that on your own? Let’s step up, what about Graph Topology, Parallel Algorithms, and Complexity.

throwaway613834 9y ago

Huh, this is hard to swallow. What digital EE courses did you take in school?

ktta 9y ago
I've had the same thing said to me by my professor about FPGAs. You can't really find much about current EE practices because companies just don't give them away or blog like the software guys do. Textbooks tend to be very general (also old) and don't focus on the tooling used which is the bulk of the work you have to do.
- egwor 9y ago
  Tooling changes (very) regularly. Learning concepts and techniques is far better for the long term since many of those techniques/concepts can be applied to new situations.
- throwaway613834 9y ago
  Also skeptical of this one. Tooling seems kind of beside the point? It's like arguing that your school taught you Python and that didn't prepare you for Java (or whatever).
  - ktta 9y ago
    Bulk of the work is done in simulation and verification with FPGAs.
    - DigitalJack 9y ago
      This is true, but those tools doesn’t change that often at a given company. VCS and questa probably cover 99% of the fpga market.
      - ktta 9y ago
        That's exactly my point. You don't get access to those tools in school unless you are in an excellent graduate program AFAIK if that. The simulation I can do in Vivado is apparently nothing compared to the scale those tools take it to. I've talked to someone in the industry that they told me Vivado's simulation is only okay for beginner projects.
  - DigitalJack 9y ago
    Tooling is most definitely not beside the point. And if your focus is backend, that is what you spend most of your time on.
    - throwaway613834 9y ago
      > Tooling is most definitely not beside the point. And if your focus is backend, that is what you spend most of your time on. The question isn't what you spend your time on though. It's what is worth spending time teaching/learning in school. You also spend all your time using a mouse and keyboard but (hopefully) you don't take courses to learn how to use the mouse and the keyboard in school. If most people could pick it up in a short period of time on the job then it doesn't seem a worthwhile use of time to teach in a course.
DigitalJack 9y ago
Fair point, I had one class called logic design. It was my favorite, although I hated breadboarding 7400 logic. The class was mainly Boolean algebra with some flip flops thrown in at some point. If I recall correctly that was my sophomore year. There was next to no digital in my courses. Heavy on math and analog circuits. I think that one class on digital logic was it.
- PascLeRasc 9y ago
  Digital logic was also one of my favorite classes. Really fun to think about some of the problems.
- throwaway613834 9y ago
  > Fair point, I had one class called logic design. The class was mainly Boolean algebra with some flip flops thrown in at some point. Yeah that explains it. :-) So nothing about writing Verilog, feedback pipelining, adder design, clock skew, and stuff like that? Because those are the kinds of things I'd expect in a digital logic course (if one were available) and I would be shocked if you didn't need to know those concepts for a job in ASIC design.
  - DigitalJack 9y ago
    Nope. Well, I’m sure we learned simple adders, again...Boolean algebra. Everything else I learned on the job. If I had gone into board design, I’m sure a lot more would have been relevant. Many places the chip designer is also the board designer, but sadly that is not the case here.
    - psykotic 9y ago
      CE/EE departments sometimes offer an undergraduate VLSI intro course where you learn the theory from a textbook like Weste and Harris and get to design and tape-out your own chip via the Cadence or Synopsys tools and then fab it through an MPW service like MOSIS or CMP. So, it's certainly possible to learn some hands-on chip design fundamentals in school, but if your department doesn't offer it, you're pretty much out of luck. You could do your own extracurricular MPW run of a small design for maybe a thousand dollars, but you'll be stuck with SCMOS cell libraries and free tools like Magic or Electric, which aren't representative of industry.

Draco_Au 9y ago

Including solving differential equations? You needed Calculus for that? Did you teach that on your own? Let’s step up, what about Graph Topology, Parallel Algorithms, and Complexity.