Aspie think.

[Ronald Zeeman]

I love stories like this. We are not like NT's Here I am an Aspies, picking out fellow Aspies 70 years later. I can spot us.
hiding in plain sight. That was fourth video I've seen so far all were obviously one of us.

 

[Ronald Zeeman]

Here is another one. The person making these war videos does not see the pattern, I do so obvious.

 

[Ronald Zeeman]

Really liked this one. Yes society needs us.

 

[Ronald Zeeman]

The one common factor is the confidence these Aspies had which I share do not second guess. Being right is being right no room for debate. They may not see it, you do.

 

[Ronald Zeeman]

This is why I fell in love with quality assurance.

 

[Tuffsy]

You hit the nail on the head, we are indispensable! As an IT support person I could always see behind the problem a person was describing, which usually assumed (incorrectly) a reason for the problem which was totally off the mark.

 

[Ronald Zeeman]

Was this guy one of us, visual thinker.

 

[Ronald Zeeman]

more examples

 

[Ronald Zeeman]

 

[Ronald Zeeman]

My brothers latest Email

• 've finally solved all the problems with measuring low voltages and come up with a very simple circuit and even simpler algorithm to control it.
  I'm using the current version for use with a load cell and the resolution is only down to the millionths of a Volt, but it's easy to extend it down to the billionths of a Volt range.
  I realize that the majority of people reading this email will have no interest, but just in case someone is curious, I've attached all the design files and other information.
  Conventional ultra low voltage measuring meters used choppers to convert low level DC to AC then amplify the AC convert it back to DC and then feed it back to the input. This sort of works, but there are all kinds of artifacts caused by the conversion and the signal being fed back contains a lot of noise which limits the ultimate sensitivity.
  In my circuit, everything is DC and digitally controlled which eliminates almost all feedback artifacts and as a bonus gives a direct digital reading of the input signal. Also, no precision parts are required except for the digital to analog converter used for measuring.
  The feedback algorithm is stupid simple, although it took me a long time to get to that point. It's like stock market investing, it's hard to discover the simple rules underlying a good investment strategy because of all the obfuscation. But once understood, incredibly simple.
  All the algorithm does is sample the null detector output and if the feedback is too low, increase it by one count. If the feedback is too high, decrease it by one count. Keep doing this and eventually the feedback will bounce around the correct value. The only trick is to make sure that the feedback doesn't change faster than the null detector output can change. If the change is too fast, the feedback will overshoot and undershoot the input voltage causing oscillation.
  The feedback voltage itself has to be in the millionths or billionths of a volt range also and this was a major sticking point. I solved this by creating coarse and fine voltage controls. Both the coarse and fine control only have 256 steps each and are constructed using eight bit DACs. The trick is that the fine control interpolates a few steps of the coarse control. If the fine control interpolates over 10 steps of the coarse control, the final step size is 256 / 10 * 256 = 6,554 steps. Additional interpolation can be added for even higher resolution.
  Adjusting the controls is done by setting both controls to mid range and then adjusting only the fine control. When the fine control goes out of range, it is set back to its midpoint and then the coarse control is adjusted by one step. The fine control is then adjusted, and if necessary, the coarse control as above. After a short time, the coarse control won't require adjustment and only the fine control will have small adjustments to maintain the null.
  The resistor networks required for combining the DAC outputs are very difficult to design using conventional techniques, but relatively easy using the quantum annealing algorithm. That's another stupid simple algorithm that works incredibly well and it's also based on a very simple principle.
  Every problem exists in a solution space. Solving that problem involves searching the solution space until the solution is found. If the solution space is continuous, there is effectively an infinite number of possible solutions and searching for a solution can take an infinite amount of time.
  But generally, a solution doesn't need infinite precision. When choosing a stock to buy, a company is effectively a large chunk of the solution space. The solution space is quantized into discrete companies and there are a limited number of companies. It's the same with the resistors required for the DAC summing network. There are a limited number of available resistor values so even though there are potentially an infinite amount of summing network solutions with unconstrained resistor values, there are only a few solutions when the resistor values are fixed or quantized.
  To choose a company to invest in, I set a few simple constraints, it must have low debt, it must pay a dividend, the yield should be above 10% and the range of industries is also constrained. The number of companies meeting these constraints is very small and a simple search can quickly find the right companies.
  It's the same with the resistor values. There are constraints on the maximum and minimum resistor values. There are constraints on the voltages going into the network because each DAC has only eight bits of output. There are also constraints on the network output voltage for each DAC setting.
  All of these constraints chop up the problem space into big chunks requiring only a relatively small amount of searching to find a solution among the chunks.
  I suspect our brains work the same way by categorizing information into generalized chunks and then searching among those chunks. The most egregious example of this is the division of politics into left and right. This gives only two chunks to choose from and greatly reduces search time. Various characteristics are then assigned to either the left or right with the expectation that have if have an opinion on some subject that falls on the left side of the political spectrum than all of your other views must also be on the left side.
  This applies to a lot of other characteristics and is so strongly held that it must be something that we are born with. From what I've seen with my quantum annealing program, this is a very powerful way of dealing with complex problems and evolution would strongly favor it. Having to do multi dimensional fine grained searches for a solution when faced with a predator would take you out of the gene pool very quickly. It's far better to focus on something like if it's got big teeth, run like hell, than to spend time debating the creature's oral hygiene and whether brown eyes are more dangerous than hazel eyes, etc.
  I won't be surprised that in the very near future that someone will discover that the brain uses very simple tricks and all of those megawatt eating AI computers are suddenly going to be obsolete. The early steam engines were used to pump water out of the coal mines and they were so inefficient that the joke was it required a coal mine to fuel them. It didn't take that long before we had tiny powerful turbines.
  Maximum efficiencies of engines and turbines, 1700-2000
  As much as I love technology, I won't buy any AI stocks. The disparity between current AI technology and our brains is just too large and the chances of a breakthrough are rising exponentially as AI is used to further refine AI.
  This also applies to my own research, I've been working on the microvoltmeter problem for over 10 years, only to see it collapse into a very simple solution over a very short time. All it takes is a little bit of the right insight.
  That insight also shows how to solve a lot of other problems. My hope is that others can see what I see and build on it as that is how society advances .
  
  
  
  
  

  ---