That's because it's known as a convention and it does not matter, as long as it's consistent, at all.
Hi KSTR, yes we agree on the concept of setting a 'baseline' before seeking to discriminate smaller differences which is the point of the good intended help I was trying to give scottjoplin. Cable direction swap differences are indeed much less magnitude than speaker polarity swap differences but ime do manifest in a manner sounding a bit like wrong acoustic polarity does and cable direction is just another standard baseline item to get right, everybody's mileage may vary. I wish standard unbalanced cables were not ever directional but I find this is not the case in the many common type economical cables that I have auditioned.
Dan.
To be clear Dan, the only issue I have with what you said was listening to both speakers whilst one is inverted.
I believe this is just an attempt to get rid, or, at least, reduce the effect of the input filter by overdriving-it.
What interest us is the slew rate of the part of an amp INSIDE the feedback loop. So, we have to remove the passive input filter before to measure the slew-rate of an amp. Saturating-it by a large signal input signal is a way to be sure that you will have the full level rising time at the output. The slope can be non linear. Reason why this 10-90% of the full excursion measurement standard ?
The good practice is to have a slew rate lot faster than any audio signal slope, and to ensure, with a low pass input filter, that no parasitic component can never reach this speed. I like to overkill this aspect. And current feedback topology as a way to achieve this.
Attachments
Last one on the cable "audible" ITD phase shifts. Now I have compared the input and output of the cable. The amp is driven by a very good step (about 5ns rise time), measurement is made simultaneously on amp speaker terminals and behind the 20 feet of speaker zip cord, loaded with my speaker box. Blue - cable input, red - cable output. Sorry, I do not believe in cable induced "phase shifts" being audible. The wave properties do not play any game in audio frequencies.
Attachments
Sometimes audio involves many of our senses: sight, hearing and smell.
To touch it, it can be even more disagreeable, with high voltages or high temperatures ;-)
I am tying to help you, what is your your issue with "listening to both speakers whilst one is inverted." (for the duration of the testing session) ???.
I've told you, it messes up the polarity/phase causing distortions etc, muddies the water unnecessarily. Connect them both in phase and reverse the polarities together to compare.
The issue is that one speaker in correct polarity and 2nd speaker inverted makes out-of-phase acoustic field (subtractions). No localization and impossible to evaluate both mono and stereo by ears. Just another crazy idea (oh sorry, open minded idea).
Max, you make me think of Sphinx of Giza, who goes through the centuries without ever changing his point of view.
Except that, he, the sand winds slowly erode its outlines.
Well, I leave you, I have to work at the ship that I build in my garage to time travel.
Last edited:
My method has for a long time worked perfectly well for me, and that method is to listen to each speaker in relative isolation by standing in front of each speaker in turn in order to decide which sounds preferable. This method is in real time and removes delays in switching polarities.....the effective polarity switching is in moving your white a** from in front of one speaker to in front of the other speaker. You can also run just one speaker at a time (one in normal polarity, one in inverted polarity) and decide which sounds preferable. You two can also do whatever you like, wherever you like, however you like, with whomever you like and whenever you like, I am just saying what works for me maybe it can work for you, I am well past caring.
Dan.
Last edited:
You give a perfectly valid definition. That said, an analog slew rate can be defined as the rate of change, as in volts per microsecond.
Many amps can reproduce a hf signal if it is small amplitude, but raise the amplitude and the amp will not follow well.
Jn
And by modifying one's posture, it may be established how rigid this testing method is. Brilliant.
Well - this was true in ancient times. It is no problem to design an amp which has SR high enough. Please let's not create pseudo-problems. We are moving in circles and often arguing with 60 years old SOTA.
No.Max Headroom said:
One will be 'correct' - assuming the entire recording chain preserved phase for every instrument etc.
One will be the opposite of 'correct' - and for most people listening to most music this will be indistinguishable from 'correct'.
The other two will be obviously wrong and the stereo image will simple collapse, together with a lack of bass.
People often confuse slew rate with slew rate limit. Slew rate is a signal property, and it depends on amplitude. Slew rate limit is an amplifier property.syn08 said:
Is this that special day each year when the weakest pupil in the class gets to pretend to be the teacher, and shares his confusion with his friends?Max Headroom said:
Most people are unable to do this. I suspect that some who think they can do this actually cannot. I think it highly unlikely that someone who actually can distinguish acoustic absolute polarity would find it such a strong issue that it trumps incorrect interchannel polarity, yet this is what you appear to claim.
Look at your time base. Given that humans are sensitive to ITD at the 2 to 5 microsecond level, do you believe that your test setup is capable of resolving an impedance dependent settling time at the timeframe humans are capable of?
Try that using resistors from 2 to 2000 ohms. (I typically do the 2,5,10,20,50,100,200,500,1000,2000 thing as it graphs well on log "paper")you might want to go finer resistor resolution around the cable RF z.
What you will see is that the voltage at the load will rise the fastest at cable z, probably 120 to 150 ohms unless it's landscaping zip with really thick insulation.
If you plot time to say 80% vs load r, you will find a cusp at cable z.
If you then bracket that graph using the high and low impedance of your typical speaker (just in the audible frequency range of course) you will see the span of delay that zip cord will cause on that speaker.
If the span is below 1 microsecond, it is inaudible. If it is over 2 microseconds, audibility cannot be discounted.
If you parallel 4 zips, each with slightly different twists, the repeat all, you will find that the resultant delay span is heavily reduced. I estimate a 25 to 30 ohm cable is optimal, and keeps capacitance low enough to keep even fast amps happy.
Thank you for doing the test, you are the best.
Jn
- Status
- Not open for further replies.