Hi and thanks for the reply
From what i understand then distortion is an issue and listening tests are the only way to assess it
The problem i have with listening tests is just subjectivity
Only that.
Regards, gino
Then structure your listening test to be objective.
http://www.linearaudio.net/images/LA Vol 2 Yaniger(1).pdf
And even some of those recommendations are fraught with unproven assumptions and folly. The scientific process requires controls and validation of assumptions. Rarely is that done in all the tests I've read.
I read an interconnect comparison test that used a Y-adapter to split the output of a CDP using the two cables under comparison to independent line level inputs of a preamp. The cocky reply was that you would likely not hear a difference. When you measure what's actually going on, I'd say that is a certainty. The grounds combine the electrical characteristics of the two cables as measured at either end. You are not comparing cable A to cable B. You end up comparing the combined capacitance of cables A and B at one input to the combined capacitance of cables A and B at the other input. Hear any difference?
ABX boxes share a similar invalid assumption. The usual story is that the various contacts and switches possess only tiny amounts of resistance and capacitance so they should not have any effect on the results. Which completely misses the point. When comparing amplifiers switching both the input and speaker leads, Frank Van Alstine has measured a similar blending of characteristics - in this case the reflected feedback loops found at the inputs of the two amplifiers which are now shared. You are not comparing amplifier A to amplifier B. You are comparing amplifier A with the feedback correction from amplifier B to Amplifier B with the feedback correction of amplifier A.
Bogus science results in bogus results.
Inherent in testing is some diligence in setting up the test. If you think that the specific suggestions I made are "fraught with unproven assumptions and folly" for answering the questions they are designed to answer, you should write a Letter to the Editor- Jan always appreciates well-thought-out criticisms and intelligent discussion, and frankly, so do I.
Don't you find it amazing that no one who uses ABX boxes has attempted to measure what Frank Van Alstine has noted?
As for me, I really couldn't care less. Perhaps someone more interested would query him as to the controls he has used to measure the operation of the boxes or Y-adapters. The so called "Clever Hans" effect can be controlled in a single blind environment without resorting to the bogus boxes.
It matters to some people. There are so many distortion mechanisms they are not all discovered, codified, or measured. I truly believe hard rock and rap musicians LOVE to listen to distortion. I personally have my list of dislikes: ODD HARMONIC DISTORTION (3, 5,7,9th to infinitum). TIM (Transient Intermodulation Distortion) SID (Slew Induced Distortion). Personally IMHO even harmonic distortion more than ..5% behaves almost like an enhancement to music, a sound elixer of euphony. Poorly designed room cause harmonic ringing and standing wave pertubations. The list is endless. GRH
I refer to spending any of my time debugging ABX boxes.
While I enjoy a wide range of popular music, what most people consider "live" includes the horrible resolution of sound reinforcement systems using PA amps and speakers.
As for me, I enjoy hearing higher fidelity at home devoid of those distortions. I stopped going to rock concerts as a teenager.
That wasn't a report, it was an internet post giving absolutely zero details (like who designed the particular ABX box in question and how it was actually wired) by someone who has a financial stake in the outcome. His speculations about feedback don't make any sense to me for non-inverting amps (the vast majority).
I can't comment on the Y-adapter stuff you posted since I have never done DBT comparisons of amps or interconnects that way. The ABX RM-2 module switches both inputs and outputs, and that seems to have been used by several people. VA's objection that the amps are fed by a common source seems to be incorrect.
They weren't speculations - which is all I ever see with guys like Krueger, et. al. Feel free to disparage the word of a well respected manufacturer of audio gear as to what he's measured. I'm sure he would just smile.
Which would cause significant POPS! during switching if the grounds weren't connected. Do you ever switch interconnects to your power amp with gain up?
Interesting. I note Frank isn't against controlled testing per se but has reservations for the way it is conducted.
Unfortunately, he apparently doesn't understand that the feedback signal in a generic amplifier is returned to the inverting input and NOT to the input which is connected to the input jack. So the feedback signal of one amp is not impacting the other.
But at any rate it should be rather simple to switch both input and output of the amp which totally disconnects any common connection.
Connecting the two duts to two separate speakers appears smart but it will in practise be impossible to match speakers to the same level as the amps are matched. For instance, freq response matching between amps can easily be a fraction of a dB, while even a few dB matching between speakers - over the audio band! - is impossible.
Plus the fact that the two speakers are physically in a different position already invalidates any sensible comparison. That is the reason that Floyd Toole spend many 100's of 1000's of dollars to build the 'speaker shuffler' at Harman: - to make sure each dut is always at exactly the same physical position.
So I understand Frankie's concerns but this solution actually makes it less convincing. And, as noted before, sensible test setups ARE possible.
BTW I agree that the Y-adapter is not a very smart way to test interlinks!
Jan
Last edited:
I'm just reporting the facts. The ABX switching module switches inputs and outputs. VA's assertion to the contrary is incorrect, irrespective of whether or not he's "well respected." And he provides absolutely no supporting data, nor anything that will allow replication of the "tests" that he claims supports his assertions, nor of his assertions about feedback. It's a self-promoting internet post, nothing more, nothing less. If there's an actual report somewhere containing the basics (specifics of test setup, data of results), I'd certainly be interested in reading it.
edit: Jan and I crossposted about the, uhhhh, novel assertion about feedback.
edit: Jan and I crossposted about the, uhhhh, novel assertion about feedback.
Exactly!
How do your measurements differ from his?
And create significant transient pops.
Possible, perhaps. We'll wait until the day they are implemented!
Not with a decent design (like the RM-2) made by actual engineers. Likewise, my Bastard Box cap tester. It was quite easy to avoid pops- unless you have an amp with severe offset problems "designed" by an audio salesman.
I gather that you didn't read the article I linked.
I'd like to comment on that tired old argument that the switches of the box would 'hide' the differences. That's is an untenable position if you think it through.
Uncontrolled, ad-hoc tests are done on 'a' system that surely contains switches, RCA connectors, tacked-on experimental connections and so on. Very often when you visit your buddy there's all sorts of crocodile-clip stuff on the xover networks, to just mention one thing.
Now the assertion is that despite all those crappy contacts and wires, the exchange of a single (for instance) cap makes a huge difference.
But when you insert a switch box with carefully chosen as transparent as possible switches, the best mechanical contacts and good quality wire, all of a sudden that difference disappears.
Really?
Jan
Uncontrolled, ad-hoc tests are done on 'a' system that surely contains switches, RCA connectors, tacked-on experimental connections and so on. Very often when you visit your buddy there's all sorts of crocodile-clip stuff on the xover networks, to just mention one thing.
Now the assertion is that despite all those crappy contacts and wires, the exchange of a single (for instance) cap makes a huge difference.
But when you insert a switch box with carefully chosen as transparent as possible switches, the best mechanical contacts and good quality wire, all of a sudden that difference disappears.
Really?
Jan
Which measurements?
Come on! That has been solved ages ago. You should not believe everything you read on the 'net (Including here
).jan
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.