Everything sounds the same

[fas42]

9 hours ago, pkane2001 said:

 

Yes, because that’s the space which ‘suits their skills’.

 

 

Round and round the merry-go-round we go ... the "finest measurement instruments were never designed to recognize patterns", but humans are excellent at detecting them - which is why AI been the great Battle. If the pattern is an unpleasant one we don't like it, and we can't pretend it isn't there just because the "finest measurement instruments" aren't registering it.

[marce]

3 hours ago, gmgraves said:

Yes, of course. After all, unless one uses precision resistors (unlikely even in uber expensive components) resistors are going to measure their given value within 5%, 10% or even 20% of their marked value, with 10%ers being the most common. Also both resistance and capacitance values vary with temperature. But also keep in mind that until the 1970's, values of components used in electronic circuits (even audio components) were figured using a slide-rule and the values calculated were more closely associated with "guesstimates" than actual calculations. I was always amazed at how some older engineers could whip out a "slip-stick", play with it for a minute or so and announce: "We need a 47, 324.236 Ω resistor in that spot" (OK, I'm exaggerating, but you get the point). Often I didn't see how they even came up with 47K, much less the rest of it (other than the fact that 47K is a standard resistor size). But what it does show is that most audio circuits  hell, most circuits period are simply not that critical. But yes, unless one is using very  precision, temperature compensated parts, two amps of the same make and model are very likely to sound subtly different. But since there is no way for the listener to know which is more "correct", it doesn't really matter.

 

I once "resto-moded" a pair of Dynaco Mark III power amps (60W/ea) using the William Zane Johnson (Audio Research) modification. when I finished the two amps each sounded slightly different, but with each playing, I couldn't tell that was true.

People still use 5% resistors, 1% is pretty standard these days, and better in critical positions, and if you use SMD you can get 4 resistors in a package where they are temp matched and kept at the same temp as each other for critical feedback, I will dig out a data sheet when back at mu work comp.

[fas42]

4 hours ago, pkane2001 said:

 

 Lush^2 thread is hilarious, by the way. I love all the participants playing with all the thousands of possible setting combinations and finding that one particular setting helps with male voices, another with soundstage, another with string instruments, another with female voices, another with bass, etc., etc. If this doesn't deserve some questioning, I don't know what does.

 

 

All those who find it "hilarious" have just driven over the dead body in the road, with complete blindness to what was in front of them. If they were listening to a speaker, and someone played with the crossover components, they would nod their heads sagely as they heard variations - but because a cable is not a prescribed meaningful element, "it's absurd!" ... oh, dear ...

 

Ummm, people can hear variations in distortion ... I suspect even the most Objective can accept that ... . Altering the configuration of the cable is changing the electrical environment that the system is working in - and because audio systems are not robust the precise spectrum of the distortion alters. In some text books this is known as logic ... something that some people only pay lip sevice to, .

[marce]

26 minutes ago, fas42 said:

 

Round and round the merry-go-round we go ... the "finest measurement instruments were never designed to recognize patterns", but humans are excellent at detecting them - which is why AI been the great Battle. If the pattern is an unpleasant one we don't like it, and we can't pretend it isn't there just because the "finest measurement instruments" aren't registering it.

Thats why all these average speed camera's can pick out your number plate at 70 MPH or it can spot a bad spot on a crisp and throw it out, or recognise a face and zoom in...

[mansr]

1 minute ago, marce said:

Thats why all these average speed camera's can pick out your number plate at 70 MPH or it can spot a bad spot on a crisp and throw it out, or recognise a face and zoom in...

Also don't forget how often our brains mess up and "detect" things that don't exist. There's even a word for it: pareidolia.

[marce]

6 minutes ago, fas42 said:

 

 

- and because audio systems are not robust the precise spectrum of the distortion alters. In some text books this is known as logic ... something that some people only pay lip sevice to, .

Its the 21st centuary if you can tune your Hi-Fi be altering the EMC pick up then it's a sad day...

[marce]

1 minute ago, mansr said:

Also don't forget how often our brains mess up and "detect" things that don't exist. There's even a word for it: pareidolia.

I detect:

 

[marce]

3 minutes ago, mansr said:

Also don't forget how often our brains mess up and "detect" things that don't exist. There's even a word for it: pareidolia.

Like my wife getting her purse out once...?

[fas42]

2 minutes ago, marce said:

Thats why all these average speed camera's can pick out your number plate at 70 MPH or it can spot a bad spot on a crisp and throw it out, or recognise a face and zoom in...

 

And that is an excellent giveaway of where your thinking is - say the word "pattern", and you lock in to a very precise technical variant of that concept - something that technology can always be adjusted to deal with.

 

[sandyk]

1 hour ago, pkane2001 said:

I’m basing it on reports from the Lush^2 thread. Changes in FR is what seems to be reported the most.

 

 Less masking of low level HF detail and low level harmonics by wideband system/external noise may be a part of the answer ?

 The best USB cable appears to be NO USB cable, just a simple USB adaptor without the screen connected through in order to reduce possible earth loops, as many USB devices have 0 volts and shield internally connected.(even USB memory)

This is more important if the PC/Server is earthed.

 If the USB device will also work without the need for the noisy +5V SMPS being passed through the cable then that will also result in a further improvement. Some commercial USB adaptors even have tiny bit switches to isolate +5V and shield connections.

 However, it's usually not possible to use an adaptor instead of a USB cable which should be no longer than 1M or so if possible.

[fas42]

1 minute ago, marce said:

Its the 21st centuary if you can tune your Hi-Fi be altering the EMC pick up then it's a sad day...

 

Every system I've come across is sensitive to its electrical environment - this is a failure of the engineering crowd, and currently means that all sorts of band aids need to be applied, to rectify.

[marce]

4 minutes ago, sandyk said:

 

 Less masking of low level HF detail and low level harmonics by wideband system/external noise may be a part of the answer ?

 The best USB cable appears to be NO USB cable, just a simple USB adaptor without the screen connected through in order to reduce possible earth loops, as many USB devices have 0 volts and shield internally connected.(even USB memory)

This is more important if the PC/Server is earthed.

 If the USB device will also work without the need for the noisy +5V SMPS being passed through the cable then that will also result in a further improvement. Some commercial USB adaptors even have tiny bit switches to isolate +5V and shield connections.

 However, it's usually not possible to use an adaptor instead of a USB cable which should be no longer than 1M or so if possible.

Probably a moot point but USB is designed to be hot swapable....

[sandyk]

8 minutes ago, fas42 said:

16 minutes ago, marce said:

Its the 21st centuary if you can tune your Hi-Fi be altering the EMC pick up then it's a sad day...

 

Every system I've come across is sensitive to its electrical environment - this is a failure of the engineering crowd, and currently means that all sorts of band aids need to be applied, to rectify.

 

It's sad that so many (especially in the USA) feel the need to use boutique mains cables with improved screening ,Hospital Grade A.C. sockets ,as well as special Mains Isolation transformers with 5/8 of SFA capacitance between Primary and Secondary sides to even get the best out of amplifiers , DACs etc.

[marce]

6 minutes ago, fas42 said:

 

Every system I've come across is sensitive to its electrical environment - this is a failure of the engineering crowd, and currently means that all sorts of band aids need to be applied, to rectify.

Wow I am sure every audio designer is happy to know he's failed in basic EMC...

[marce]

Just now, sandyk said:

 

It's sad that so many (especially in the USA) need to use boutique mains cables with improved screening ,Hospital Grade A.C. sockets ,as well as special Mains Isolation transformers with 5/8 of SFA capacitance between Primary and Secondary sides to even get the best out of amplifiers , DACs etc.

How many don't though?

[sandyk]

1 minute ago, marce said:

How many don't though?

Perhaps you should ask that question in another area of the forum, perhaps even start a Poll ?

 

Our 230VAC 50HZ systems appear to be far less susceptible to such problems, although I note that even a few Au. members use Isolation Transformers.

[fas42]

3 hours ago, manisandher said:

 

I invited @mansr up to my place (paid for his train ticket, collected him from the station, fed him, etc) because I was sure I could demonstrate what I was hearing to him. I scored 9/10 in the blind ABX test. I sat him down in front of the system afterwards and pointed out exactly what I was hearing. He claimed he couldn't hear the differences I was hearing.

 

I've found that people can be oblivious to meaningful variations, because they concentrate so hard on matters that they feel are important. The local audio friend took quite a long time to "hear" what was going on - even though I was pointing to it right beside him - I would think that he had got it, only to find next time that he gone backwards again, badly.

 

If a person can't 'hear' these important variations then they have no hope of making progress - probably ideal is to get a rig in optimum state, and then deliberately cripple the sound by doing something trivial - go back and forth, back and forth, between 'good' and 'bad' ... until finally the penny drops ...

[fas42]

10 minutes ago, marce said:

Wow I am sure every audio designer is happy to know he's failed in basic EMC...

 

What he's failed in, is understanding that people can be sensitive to subtle, unpleasant artifacts; caused by lack of adequate EMC for the situation. If one reads a textbook on designing circuits for the highest precision instrumentation purposes, it has exactly all the hints on what needs to be considered.

[sandyk]

32 minutes ago, marce said:

Thats why all these average speed camera's can pick out your number plate at 70 MPH or it can spot a bad spot on a crisp and throw it out, or recognise a face and zoom in...

 

 Or even point a camera at streams of traffic from a high vantage point 100s of metres away, and tell if you are using a mobile phone or not wearing a seat belt ,and photograph the evidence, as well as relaying your number plate info to patrol cars in the general area.

 In NSW Au. they even recently caught a truckie using 2 mobiles at the same time while driving  and had a photo as evidence !

[pkane2001]

1 hour ago, fas42 said:

 

Round and round the merry-go-round we go ... the "finest measurement instruments were never designed to recognize patterns", but humans are excellent at detecting them - which is why AI been the great Battle. If the pattern is an unpleasant one we don't like it, and we can't pretend it isn't there just because the "finest measurement instruments" aren't registering it.

 

What is your point, Frank? Why is AI and pattern recognition all of a sudden a thing for audio? I thought we were talking about the finest measurement instruments not being able to detect the differences. If there are no detectable difference there is no pattern. Regardless of how good the AI is.

[jabbr]

43 minutes ago, sandyk said:

Perhaps you should ask that question in another area of the forum, perhaps even start a Poll ?

 

Our 230VAC 50HZ systems appear to be far less susceptible to such problems, although I note that even a few Au. members use Isolation Transformers.

 

All your files have problems with noise though that we don't have ?

[sandyk]

21 minutes ago, jabbr said:

 

All your files have problems with noise though that we don't have ?

 ???

 Are you referring to the photos taken from my TV screen ?

 You do not have any direct copies of my A or V files to make such a claim, and any that may have been downloaded by you WILL have added noise due to  TX and  RX degradation via the Internet as well as added noise from your own PC, which in any event, you claim to be able to correct anyway with your Optical methodology.

[jabbr]

1 minute ago, sandyk said:

WILL have added noise due to  TX and  RX degradation via the Internet as well as added noise from your own PC, which in any event, you claim to be able to correct anyway with your methodology.

 

Nope, all my internet downloads go fiber into my NAS. No degradation. Perhaps the Australian photons are all jittery? They probably spin backwards ?

[Ralf11]

you're on fire today!

[jabbr]

2 minutes ago, Ralf11 said:

you're on fire today!

Yeah, my Win10 NAA (Minix) only gets to DSD256 over 5G Wifi, and my Topping DX7s won't do direct DSD at all with my Espressobin/linux NAA ... arrggghhhh ... the Espressobin is smooth as butter at DSD512 with the iFi iDSD Micro though ...

 

I'm creating the highest EMI environment possible to test out a new USB isolation technique...