GaryGaulin
Posts: 5385
Joined: Oct. 2012
|
| Quote (Quack @ July 22 2013,15:14) |
OMG, what did I do? LOL. |
I'm not sure yet but I never saw this before!
After using the line of code that gets RAM Address 1 cycle behind optimum (instead of no perceptible delay) the confidence lines on the chart showed one lobe dominating another, that I found coming from small imbalance between the two in how left/right spin sensor was calculated when there were many digits in the number. That took out the imbalance there, then the dominant lobe went to the other side. I then added a line of code to be 1/2 cycle behind in timing, and the lobes worked together just fine and was a faster learner.
One thing for sure, experimenting with timing is helping find the annoying math errors from some numbers getting big. Seems like when timing is way slow the least little thing out of place between the two lobes causes one to become dominant over another, or is in how the loop switches lobe but I'm still testing sensory code, all because of you! But it was a good time to go over the new sensory code, in this much detail, so I don't mind.
--------------
The theory of intelligent design holds that certain features of the universe and of living things are best explained by an intelligent cause, not an undirected process such as natural selection.
|
