Lies about vinyl vs digital

[esldude]

4 hours ago, sandyk said:

Dennis

 In your post 1380 you said that you didn't hear a difference after level matching. .

 Those who did hear differences would most certainly have done relatively close  level matching.

 

Alex

It isn't my experience in general Alex.  I find few people bother with level matching.  A surprising percentage will argue it isn't needed.  And most who do match levels are referring to simply matching by ear which is not good enough.  I think assuming this was done is going way too far. 

[Ralf11]

assuming this was "inane"

[sandyk]

8 minutes ago, esldude said:

It isn't my experience in general Alex.  I find few people bother with level matching.  A surprising percentage will argue it isn't needed.  And most who do are referring to simply matching by ear which is not good enough.  I think assuming this was done is going way too far. 

Dennis

 Most Audiophiles don't need to have ACCURATE level matching to hear obvious differences. They just need them to be fairly close.

It IS good enough doing it by ear unless the differences are only minor, and if the files have different dynamics you may be doing more harm than good to the results if trying to match them to fine limits.

 

Alex

[esldude]

3 hours ago, Rexp said:

No Paul, we are referring to your original 24/192 aiff v esitudes 16/44 downsample of the same. Frank confused matters. Your recording sounds great, the other sounds off. 

So assuming you read this post. 

There is no difference above the 24 bit noise floor below 20 khz.  The material above 20 khz is very low in level. If you perceived a difference it would necessarily have to be a very small minor difference.  Not one where it sounds obviously off. 

 

Unless, you are using a DAC that has leaky, aliasing minimum phase filtering which alters the frequency response.  Then it might sound off, but I'd expect even that to be minor rather than much off.  

 

You didn't say obviously off.  Just off.  So how does that fit with the graphed differences?

[esldude]

5 minutes ago, sandyk said:

Dennis

 Most people don't need to have ACCURATE level matching to hear obvious differences. They just need them to be fairly close.

It IS good enough doing it by ear unless the differences are only minor, and if the files have different dynamics you may be doing more harm than good to the results if trying to match them to fine limits.

 

Alex

I fully and completely disagree.  🧐

 

The level at which level matching isn't needed for obvious differences are truly huge and tremendous differences.  HUGE I TELL YOU HUGE!

 

So why even do it that way when good level matching puts you into a place where pretty small differences can be heard and slight level mismatches completely swamp those?

[sandyk]

4 minutes ago, esldude said:

I fully and completely disagree.  🧐

 

 Of course you do .

If you are doing definitive DBTs etc. it is a different matter.

 If members were expected to go to the Nth degree to do this , almost nobody would ever bother participate in stuff like this, or any of the comparisons that you post for that matter. This is a Hobby , not an Engineering Society report.

[esldude]

1 minute ago, sandyk said:

 

 Of course you do .

If you are doing definitive DBTs etc. it is a different matter.

Nope.  If you are doing casual comparisons sighted it is still highly important.  Unless you just want the shit and giggles factor for some cheap meaningless entertainment.  

[sandyk]

12 minutes ago, esldude said:

Nope.  If you are doing casual comparisons sighted it is still highly important.  Unless you just want the shit and giggles factor for some cheap meaningless entertainment.  

 You are taking this stuff far too seriously. This is a Hobby , not an Engineering Society report.

If you only want stuff like Frank provided in the threads ,then that's a different matter, but don't expect too many members to participate and go to the extremes that you are demanding.

It's too damn time consuming, and how many members have the equipment to calibrate the differences to such a fine degree ?

What do you want people to do, include a test tone with each example so that you can read the level at the speaker terminals ?

[esldude]

1 minute ago, sandyk said:

 You are taking this stuff far too seriously. This is a Hobby , not an Engineering Society report.

If you only want stuff like Frank provided in the threads ,then that's a different matter, but don't expect too many members to participate and go to the extremes that you are demanding.

It's too damn time consuming, and how many members have the equipment to calibrate the differences to such a fine degree ?

You need a multimeter.  You need the ability to read voltages.  And determine percentages.  

Is this extreme?  Considering the things done and discussed on this site I'd say having a multimeter and reading voltages is not at all extreme.  Much less thinking about the hard headed opinions held by people who can't be assed about using a multimeter to set voltages.  

 

Come on, don't be ridiculous.  

 

You advocate unusual practices and the need for special power supplies to even copy digital files to even do meaningful comparisons at all, but having a multimeter to read speaker voltages is extreme?????  🤪🤔🙄🤯

[Rexp]

20 minutes ago, esldude said:

So assuming you read this post. 

There is no difference above the 24 bit noise floor below 20 khz.  The material above 20 khz is very low in level. If you perceived a difference it would necessarily have to be a very small minor difference.  Not one where it sounds obviously off. 

 

Unless, you are using a DAC that has leaky, aliasing minimum phase filtering which alters the frequency response.  Then it might sound off, but I'd expect even that to be minor rather than much off.  

 

You didn't say obviously off.  Just off.  So how does that fit with the graphed differences?

It tells me something I already knew: measurements don't mean squat. 

Maybe as this is OT the comparison should be posted in a new thread. 

[esldude]

3 minutes ago, Rexp said:

It tells me something I already knew: measurements don't mean squat. 

Maybe as this is OT the comparison should be posted in a new thread. 

Well you are living in a make believe fantasy world.  Fantasies are fun as far as they go.  Enjoy.  

 

[sandyk]

4 minutes ago, esldude said:

You need a multimeter.  You need the ability to read voltages.  And determine percentages.  

Is this extreme?  Considering the things done and discussed on this site I'd say having a multimeter and reading voltages is not at all extreme.  Much less thinking about the hard headed opinions held by people who can't be assed about using a multimeter to set voltages.  

 

Come on, don't be ridiculous.   

 

You advocate unusual practices and the need for special power supplies to even copy digital files to even do meaningful comparisons at all, but having a multimeter to read speaker voltages is extreme?????  🤪🤔🙄🤯 

 Dennis

 You are being ridiculous. Most members are not going to stuff around like that  to do simple listening comparisons, and I doubt that even  20% of the members have a DMM and know how to do things like this.

[esldude]

1 minute ago, sandyk said:

 Dennis

 You are being ridiculous. Most members are not going to stuff around like that  to do simple listening comparisons, and I doubt that even  20% of the members have a DMM and know how to do things like this.

Then maybe they shouldn't post their opinions as being meaningful.  Or maybe they should learn these simple skills.  A 10 year old can do it pretty easily.  

[esldude]

Here is a new poll on multimeters.  Please vote in it. 

 

 

 

[sandyk]

13 minutes ago, Rexp said:

It tells me something I already knew: measurements don't mean squat. 

 

Unfortunately, to a small select group of members they mean everything, with some looking at Audio files with S/W before even listening to them to help with their decisions.

[sandyk]

4 minutes ago, esldude said:

Here is a new poll on multimeters.  Please vote in it. 

 

 

 

Dennis

 The bulk of the members never even post regularly, and currently there is only a relatively small core group that contribute to the forum,  and those that do post regularly are more likely to have a DMM.

The results from such a poll will be skewed and will be meaningless.

[esldude]

6 minutes ago, sandyk said:

Dennis

 The bulk of the members never even post regularly, and currently there is only a relatively small core group that contribute to the forum,  and those that do post regularly are more likely to have a DMM.

The results from such a poll will be skewed and will be meaningless.

So you know the pulse of the readership.  And any actual data from polls are skewed and we can just trust you that 20% have multimeters?

 

Okay, then a worthy goal would be to educate those who are lurkers in the benefits and ease of using very inexpensive multimeters to match levels and improve the efficacy of their efforts.  

 

Beats banging on about bit perfect files sounding different depending upon how they have been treated by a factor of about 1 trillion to one or perhaps more correctly by an infinite ratio.  I am talking about something that costs $20 to implement and anyone could learn in a few minutes.  While you are talking about something that costs hundreds, is incredibly inconvenient, and has no theoretical basis for being real nor any actual basis for being real.  It is such an ephemeral result (according to you) that any mis-step of the smallest proportions swamps it.  What I am talking about is a top level immediate difference that is irrefutable.  

[Paul R]

56 minutes ago, esldude said:

I fully and completely disagree.  🧐

 

The level at which level matching isn't needed for obvious differences are truly huge and tremendous differences.  HUGE I TELL YOU HUGE!

 

So why even do it that way when good level matching puts you into a place where pretty small differences can be heard and slight level mismatches completely swamp those?

 

I have to admit, I do agree with Dennis here. It's easy to put something in a tape loop and have one volume, the tape loop or the direct path,  just slightly louder than the other. The louder one will definitely sound better, every time.  

 

I put the five copies of the two unprocessed (but level matched)  recordings in a loop this evening, then had my wife delete two of them at random.  I listened sighted to the files twice through, then went away for 30 minutes or so. I then did five listening sessions, separated by about 10 minutes each. 

 

The pattern that was left in the playlist was 44.1K, 192K, 44.1K, but I did not know that until after the testing was complete.

 

It was fairly easy to tell the difference for me, but much harder to decide which was which.

 

The differences I believe I heard are that the 192K version sounds much more balanced to me, a little thinner perhaps, but much cleaner and more precise. The 44.1K version had more slam, which may be a deciding factor for some people. There seemed to be more detail and separation in the 192K version as well.  All of which is subjective, but the differences I made my guesses below on.

 

I think I need to remove the clicks and pops and have Karen setup a long playlist in random order. Then I will try again, going through part of the playlist in each listening session.

 

-Paul 

 

 

 

Guess             Actual         

192K.             192K 

44.1K             44.1K

44.1K             44.1K

 

192K              192K

192K              44.1k

192K              44.1K

 

192K              192K

192K              44.1K

44.1K             44.1K

 

44.1K             44.1K

192K              192K

44.1K             44.1K 

 

44.1K.            44.1K

192K              44.1K 

192K              192K  

 

[sandyk]

39 minutes ago, esldude said:

Beats banging on about bit perfect files sounding different depending upon how they have been treated by a factor of about 1 trillion to one or perhaps more correctly by an infinite ratio.

 

 A well qualified E.E. from Grafton NSW verified my reports with the " eye candy" video today.

 Have you checked out the more recent comparison yet Paul ?

[sandyk]

16 minutes ago, Paul R said:

It's easy to put something in a tape loop and have one volume, the tape loop or the direct path,  just slightly louder than the other

Very few members these days are likely to have a tape loop these days, as very few are still likely to have a Tape or cassette deck these days. Besides which, most well implemented Tape Loops use a Buffer amplifier at more than likely a totally different output level.

[SoundAndMotion]

55 minutes ago, sandyk said:

The results from such a poll will be skewed and will be meaningless.

 

Alex,

You are absolutely right! I can think of other reasons such a poll is skewed. But is Dennis not merely saying the same thing?

 

You say: “The results from such a poll will be skewed and will be meaningless.” 

 

and Dennis says (paraphrased): “The results of such* a listening test will be skewed and will be meaningless”
such* = lacking an objective level match. A simple objective method is important; otherwise, allowing a louder, better-sounding level for the one you know should sound better is a risk.

 

1 hour ago, esldude said:

I fully and completely disagree.  🧐

 

The level at which level matching isn't needed for obvious differences are truly huge and tremendous differences.  HUGE I TELL YOU HUGE!

 

So why even do it that way when good level matching puts you into a place where pretty small differences can be heard and slight level mismatches completely swamp those?

 

Let me ask you a simple hypothetical: let’s say there’s a thread asking which sounds better to you, 128kbps mp3 or 128kbps aac? Being a clever fellow, I check the frequency spectra and find they differ. I believe this is “unfair” so I “correct” the spectrum of one of them and ask you to compare. Would you find that reasonable?

 

Although various measures of equal loudness curves (Fletcher-Munson, ISO226:1987, ISO226:2003) differ in absolute numbers, they all agree that perceptual frequency response is level-dependent.

[sandyk]

2 minutes ago, SoundAndMotion said:

Being a clever fellow, I check the frequency spectra and find they differ. I believe this is “unfair” so I “correct” the spectrum of one of them and ask you to compare. Would you find that reasonable?

In a word, NO !

 You are assuming that the extra processing will make absolutely no difference in the S/N or Jitter areas.

 That is NOT what I find, where even opening and saving again a high res file in Sound Forge 9 results in the new file sounding slightly degraded. In fact, a past member, "thesurfingalien"  created for me a simple software to make some DVD-A files universally playable, and his S/W resulted in a file that sounded closer to the original than the SF9 version which only removed an incorrect size header chunk..

[SoundAndMotion]

8 minutes ago, sandyk said:

In a word, NO !

 You are assuming that the extra processing will make absolutely no difference in the S/N or Jitter areas.

 That is NOT what I find, where even opening and saving again a high res file in Sound Forge 9 results in the new file sounding slightly degraded. In fact, a past member, "thesurfingalien"  created for me a simple software to make some DVD-A files universally playable, and his S/W resulted in a file that sounded closer to the original than the SF9 version which only removed an incorrect size header chunk..

My apologies! My post was unclear and therefore you misunderstood my point. My entire post was in support of objective level-matching for listening tests that one intends to report to others.

 

An incorrect level match will skew the results of the listening test.

 

Subjective level matching is a risky, unreliable method. Although not 100% of people have multimeters, buying one and using it correctly is a small price to improve the quality of the listening tests... that you intend to share. For your own purposes at home, there are no restrictions and everyone can obviously choose their own methods.

[Paul R]

24 minutes ago, sandyk said:

Very few members these days are likely to have a tape loop these days, as very few are still likely to have a Tape or cassette deck these days. Besides which, most well implemented Tape Loops use a Buffer amplifier at more than likely a totally different output level.

 

Probably everybody here has a tape loop or similar capability.

 

 

[Rexp]

1 hour ago, esldude said:

Well you are living in a make believe fantasy world.  Fantasies are fun as far as they go.  Enjoy.  

 

Lol, thanks for the file and graphs though, very interesting for those that can hear a clear difference.