On Sun, 10 Jan 2010 10:49:43 GMT, "Brian Gaff"
wrote:

Well, I suppose the misleading part of all this is actually calling them
square waves in th first place, as has been noted these do not exist.
There are other forms that can be used of course, the one sided square, with
a leading slope but a fast drop, and the inverse of this.
Then there are triangular waves of course, which sound almost as bad as
square waves.

If we limit the requirement to "within the meaning of the act", then
square waves are doable - to the same extent and for the same reasons
as triangular waves.

Of course the other big issue with square wave testing is
differentiation. It is quite easy to run out of voltage drive if the
peak to peak level has been doubled by a highpass filter.

I think the listening tests do prove some things though. Firstly, nobody is
really sure what 'right' actually is, so an amp may sound bad, but it might
be actually more accurate than one you like.

It is interesting sometimes to put a scope on the psu when an amp is running
too, often some frequencies exist at quite high levels, despite all the
decoupling in the world being there, indeed, decoupling itself can make an
amp sound leaden.


In a decent amp, it doesn't matter much what is going on at the supply
rails. A good PSRR will eliminate everything. I really don't know what
you mean by leaden though - care to elaborate?

In the complexity of the audio and all the bits in the chain to how the
brain perceives the whole is still not well understood, I feel.

However, I do not think that most speaker wire tests are actually testing
the wire...


They certainly weren't when Ben Duncan did them...

Brian

d