[time-nuts] Advantages & Disadvantages of the TPLL Method

[Bob Camp]

Hi

Schematics of each of the various implementations tired would be nice. That would allow others to implement duplicates and see what happened. Without schematics only approximations can be implemented and analyzed. That generally opens up more issues than it addresses. The nature of the conversation so far is not one that encourages participation or shared experimentation. 

Details of the various experiments attempted would be helpful. The "rabbit out o the hat" approach to data gets pretty old pretty fast. It's apparent from various oblique references that far more than the run off against the Symmetricom box has been done. 

A more rigorous  approach to the math is needed if this is ever going to be accepted. This is after all an area where hair splitting issues over the math have been a very big deal for at least 40 years. 

Bob


On Jun 20, 2010, at 7:17 PM, Magnus Danielson wrote:

> On 06/21/2010 12:49 AM, Chuck Harris wrote:
>> WarrenS wrote:
>>> 
>>> GandalfG8 Posted:
>>> > "Snake Oil anyone?"
>>> 
>>> I nice short response, but it shows missed the MAJOR difference. You
>>> need to see:
>>> < http://www.thegleam.com/ke5fx/tpll.htm >
>>> ws
>> 
>> No disrespect to you or John intended, but that is a data set of one
>> experiment.
>> We used to call that a High School Proof.
>> 
>> It takes an infinite number of experiments of that sort to "prove" a
>> conjecture.
> 
> It doesn't prove it, just makes it fairly likely correct.
> 
> But then again, "correct" is not available here. Sufficiently accurate within some bounds is. These bounds can be of a myriad of types.
> 
> As for the Warren style TPLL, a particular implementation needs to be referred to, both in form of analogue design and digital processing.
> 
> From that limiting properties may be established. Some of these may be compensated for enhanced performance.
> 
> Cheers,
> Magnus
> 
> _______________________________________________
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.
>