New ask Hacker News story: A Dozen Microcomputers
A Dozen Microcomputers
2 by daly | 2 comments on Hacker News.
I have a dozen microcomputers, microcontrollers, FPGA boards, etc. If it is out there I've probably bought it. I've been learning how to program each one. One major source of frustration is that they all seem to have their own way of doing things. For example, I have the Arduino which likes the arduino IDE. I have a DE1 which requires Quartus. The DE10-nano requires a blaster cable. (Surprisingly in my "box of a thousand cables" I didn't have one). It is painfully time consuming to get the BLINK program (the "hello world" of hardware) running on each one. It is bad enough having to learn how to set the dozen random registers just to generate a quadrature waveform. But that's due to hardware design tradeoffs. It would be much better if there were open source VERILOG submodules that were considered "standard". The software IDE layer ought to provide a common API (e.g. timers, AD/DA conversion, I2C, USB programmability, etc.) Learning all of this takes time and I'm willing to crawl over the intellectual broken glass to do it but I think it is really holding back the industry. Perhaps it is time for some industry standards applied at the IDE level.
2 by daly | 2 comments on Hacker News.
I have a dozen microcomputers, microcontrollers, FPGA boards, etc. If it is out there I've probably bought it. I've been learning how to program each one. One major source of frustration is that they all seem to have their own way of doing things. For example, I have the Arduino which likes the arduino IDE. I have a DE1 which requires Quartus. The DE10-nano requires a blaster cable. (Surprisingly in my "box of a thousand cables" I didn't have one). It is painfully time consuming to get the BLINK program (the "hello world" of hardware) running on each one. It is bad enough having to learn how to set the dozen random registers just to generate a quadrature waveform. But that's due to hardware design tradeoffs. It would be much better if there were open source VERILOG submodules that were considered "standard". The software IDE layer ought to provide a common API (e.g. timers, AD/DA conversion, I2C, USB programmability, etc.) Learning all of this takes time and I'm willing to crawl over the intellectual broken glass to do it but I think it is really holding back the industry. Perhaps it is time for some industry standards applied at the IDE level.
Comments
Post a Comment