 Paul S wrote: |
| Well, I thought there was a lot of this going on all along, but that we just haven't come up with much that is "universal" | | Yes, my thought is that there is NOTHING universal, but that these choices must be observed in CONTEXT of topology to be useful. An example:
| Paul S wrote: |
| As an offhand example, I wonder if the Cu PIO coupler would "work just as well" in the same circuit but with a 300B instead of a 2A3? | | Yes, this is a good example. Maybe it DOESN'T, and we have that in our "library." Then suddenly we have something POWERFUL! We can think: what is the physical and electrical difference between the 2A3 and 300B and what is the physical and electrical difference between the PIO and the Teflon cap. From this we can begin to extract some useful ideas about the needs of topology for proper function.
I am going back to the car analogy: the Honda Civic and the Ford F-150. Initially we just classified them by the properties: wheels (yes/no), and engine (yes/no). So they seemed the same for a drive to the store. Then we added a new topology: a long distance trip, which only the Civic could do, so we discover a new property: gas mileage (high/low). And we formulate a hypothesis: for long distance trips, a high gas mileage car should be used.
In audio, we might say something like: (just an example) in DHT circuits, a capacitor with lower inductance should be used. Then we can begin to think why this would be the case.
| Paul S wrote: |
| And maybe this is why forums like this tend to fragment into "sections" that address more-specific (and putatively better-defined) | | I think it is actually because it is done in a non-systematic way, without much thought given to figure out the scientific basis of observations.
| Paul S wrote: |
| As somewhat of a romantic, I also wonder if what we need is already "out there", spread out here and among the various OCD sites, like some wild, gigantic "Encyclopedia of Hi-Fi"? Shades of Jorge Borges! | | I am sure it is, but it can only be used if it is compiled coherently in one place.
| Paul S wrote: |
| So, again, any thoughts on just how better to compile and use such a "raw" database? | | Well, just as I said above. I would hope we can get some examples like:
Component: capacitors
Types: Teflon, PIO
Application: coupling between 6SN7 and 2A3.
Topology: DHT power amplifier stages
Observation: PIO is better
This would be a positive example (where we hear a difference) BUT also important would be NEGATIVE examples, to borrow from Romy's note on the Sansui thread:
Component: diodes
Types: FRED, Schottky
Application: DC power supply
Observation: no difference
That is what I mean.
Adrian
|
|