My simulta ( Hsimen ) is intended for debugging software, not to test the emerging cirtuitos hardware, hence the simulation is implemented at a functional level and not gates, buses and signals. However Hsimen helped me, as I said in my previous article, to model certain hardware solutions, but I repeat this is not its purpose.
Either way, this project is now on a break until September, so I had a chance to think calmly about this and other matters. The simulation circuit has returned to haunt my head since seeing that other colleagues have used successfully in similar projects to mine.
I have two references: Bill Buzbee's Magic-1 and Dawid's DIY . Both authors chose (separately) for writing a program in C to represent the gate-level circuits and signals, and then test the circuit logic in the soft environment.
No longer a tempting offer because software development is far more productive than the hardware. But even so, I fear it would not work for me. The reasons are these: 1 .- Simulation
incomplete:
The simulation never reach to be complete. Not long ago, for example, had problems with my circuits "debouncing" which detected while testing the circuit to hit switches. These problems can hardly be purely electric simulated with a program written by me.
This being so, spare me the purpose of tuning changes on a circuit whose design has been "certified" by the simulator is automatically defeated.
2 .- A complex software project in itself:
A gate-level simulator is not written with simple "if ~ THEN ~ else." It is a complicated project that is to design, code and debug over many weeks before that demonstrates something. And even after the end probably will be bugs to fix ... as in any software.
suspect the same time and effort are used more by mounting the real circuit in a breadboard ... and I think opt for the latter. CONCLUSION
it known that I have the opportunity to consider the alternative of circuit simulations, supported by its use in existing projects such as Magic-1. But once removed all accounts, I closed the case:
circuits not prove my software but the old fashioned way: on a breadboard. And it is not surprising since I am a man tradisional, therefore I am building a computer tradisional cut in this century where respect for the technological tradisiones almost sounds like a word of bad taste.
0 comments:
Post a Comment