Codeless State Machine

Thread Starter

danadak

Joined Mar 10, 2018
4,057
Thinking further using the state machine wizard I know I can generate
the addresses incrementing or decrementing based on a pin, and changing
state based on another pin. So that would be code-less.

What I am trying to figure out is if I can get a hold of the A/D results to permit
control of speed/direction, and thats where I don't think I can do it code-less.


Regards, Dana.
 
Last edited:

MisterBill2

Joined Jan 23, 2018
27,513
The issue I have with this is not about having better tools... it's about poor programming... Having spent a fair amount of time with some of these chips every time something gets between you and programming it seems like there's garbage being added... many times it's for safety because they can't know your possible intent, other times it repeats instructions because it does not know how the chip is set up... in any case an example: Arduino IDE version of looping a simple high and low output. It is easy to use but here are the results:

digitalwrite() for 16Mhz Arduino takes 3.3-3.5 uSec equating to about 148kHz This is pretty fast

using direct port manipulation for the same 16Mhz Arduino takes 125nS (2 clocks high) - 251nS (4 clocks low) equating to about 2.7Mhz.

Graphical interfaces are great but what I don't like is that we keep increasing the speed of the controller to get what we want instead of maximizing through good programming the capabilities of what you have. These little chips are pretty low level compared to what you can get and also incredible for what they are.

This also reminds me of Macromedia dreamweaver... it was great because you didn't have to learn html and allowed folks to make pages visually. But wow the lines of repeated commands... I'm sure it slowed down performance.
OK, I have been ranting and raving about how slow poor bloated programming is for quite a few years. My employer was selling an electrical tester with a 386 @33MHZ, blazing speed at that time, but the older testers running a 6809 processor at 2 Mhz were still much faster. Yes, the programs had to be compiled but that only had to be done once and then after loading the code into eproms the machines were fast. So the real problem, still today, is very poorly written bloatware. Consider the most bloated of all, and we get weekly updates and patches to fix the gofs after it has been a year or more.
 
Top