If "Tube Sound" Is a Myth, Why Tubes?

[levimax]

Thanks Blumlein 88. That's a generous bunch of information.

However, I'm the furthest from a DIY guy you will find. If it's not essentially easy for me to do the test it won't happen, and the above is getting too far down a rabbit hole for my comfort level. I may even have sprung for the switchbox I linked to earlier, but with my industry hit hard by COVID money is tight.

Hi Matt:
I am into DIY and am going to modify a switcher I have to be "tube friendly" per Blumlein 88 and SIY and run some tests. When I am done (and have proven the modified switcher works right) I can send it onto you.... as long as you promise to send it back when done . I will see if I can make it as easy as possible. I am thinking a mono speaker test as my switcher is 2 sources to 1 output so 2 amp channels and one speaker. Do you have the ability to "downmix" stereo to mono or have true mono sources you can output to two channels .... preferably not mono LP's as people will claim that is an issue.

 

[MattHooper]

Hi Matt:
I am into DIY and am going to modify a switcher I have to be "tube friendly" per Blumlein 88 and SIY and run some tests. When I am done (and have proven the modified switcher works right) I can send it onto you.... as long as you promise to send it back when done . I will see if I can make it as easy as possible. I am thinking a mono speaker test as my switcher is 2 sources to 1 output so 2 amp channels and one speaker. Do you have the ability to "downmix" stereo to mono or have true mono sources you can output to two channels .... preferably not mono LP's as people will claim that is an issue.

Thanks very much for the offer!

Yes, if you manage to devise this switcher I'd likely want to try it out. Very kind of you.

Though the mono thing is throwing me a bit. I'm aware of the Harman Kardon mono tests, but I'm used to listening critically in stereo and one of the stronger aspects I detect is in imaging/soundstaging along with the other aspects I mentioned. It would be a bummer to remove the ability to detect those changes in a test.

 

[Inner Space]

@MattHooper - again, genuinely interested, and not trying to give you a hard time ... but why do you need a switcher? First (and yes, this might be a personal hobbyhorse) there is actually no research that says you can't remember SQ impressions for a long time. And in your case, your honest and heartfelt report indicated you could remember your CJs' SQ for at least two days: "After a couple of days I got my CJ amp back ... Re-listened to a bunch of tracks on the Bryston, then threw in the CJ amps and .... Aaaaahh. There it was! That gorgeous, relaxed, richer, rounder, bigger sound I was used to."

So wouldn't it be enough to have a family member switch connections, maybe on a daily basis? You could listen all evening, and note down "Bryston" or "CJs". Then at the end of the week (or month, or whatever) you could check your score. Wouldn't that work? You wouldn't even need to level match either - I assume you didn't just prior to your "Aaaaahh" moment? Yet the difference was immediately apparent.

 

[MattHooper]

@MattHooper - again, genuinely interested, and not trying to give you a hard time ... but why do you need a switcher? First (and yes, this might be a personal hobbyhorse) there is actually no research that says you can't remember SQ impressions for a long time. And in your case, your honest and heartfelt report indicated you could remember your CJs' SQ for at least two days: "After a couple of days I got my CJ amp back ... Re-listened to a bunch of tracks on the Bryston, then threw in the CJ amps and .... Aaaaahh. There it was! That gorgeous, relaxed, richer, rounder, bigger sound I was used to."

So wouldn't it be enough to have a family member switch connections, maybe on a daily basis? You could listen all evening, and note down "Bryston" or "CJs". Then at the end of the week (or month, or whatever) you could check your score. Wouldn't that work? You wouldn't even need to level match either - I assume you didn't just prior to your "Aaaaahh" moment? Yet the difference was immediately apparent.

I get your point. But if I really want to do a blind test I'd want to do it right. Matching output levels, randomized switching, immediate switching.
I think the above would be unlikely to satisfy either me or anyone here, given the looseness of that method.

 

[SIY]

Immediate switching is indeed far more sensitive, which is another reason my skepticism pops out when we're talking about delays of days for comparisons.

Delighted to see @levimax and @MattHooper heading down this path. May I suggest one more potential alternative? If your experience is that good ADC/DACs are audibly transparent (this is very much my experience), then record the signals at the speaker terminals, level match using software (I use Adobe Audition), then ABX those files.

 

[Blumlein 88]

@MattHooper - again, genuinely interested, and not trying to give you a hard time ... but why do you need a switcher? First (and yes, this might be a personal hobbyhorse) there is actually no research that says you can't remember SQ impressions for a long time. And in your case, your honest and heartfelt report indicated you could remember your CJs' SQ for at least two days: "After a couple of days I got my CJ amp back ... Re-listened to a bunch of tracks on the Bryston, then threw in the CJ amps and .... Aaaaahh. There it was! That gorgeous, relaxed, richer, rounder, bigger sound I was used to."

So wouldn't it be enough to have a family member switch connections, maybe on a daily basis? You could listen all evening, and note down "Bryston" or "CJs". Then at the end of the week (or month, or whatever) you could check your score. Wouldn't that work? You wouldn't even need to level match either - I assume you didn't just prior to your "Aaaaahh" moment? Yet the difference was immediately apparent.

Actually there are tests of hearing differences short vs long term. Short is much, much more discriminating.

 

[levimax]

Immediate switching is indeed far more sensitive, which is another reason my skepticism pops out when we're talking about delays of days for comparisons.

Delighted to see @levimax and @MattHooper heading down this path. May I suggest one more potential alternative? If your experience is that good ADC/DACs are audibly transparent (this is very much my experience), then record the signals at the speaker terminals, level match using software (I use Adobe Audition), then ABX those files.

I have an ADC and Audition... my ADC only has line level or MIC level inputs.... How would I hook up to speaker level inputs ?

 

[Inner Space]

Actually there are tests of hearing differences short vs long term. Short is much, much more discriminating.

The tests I have seen - and I have looked hard - are from cognitive psychology, not audio reproduction. The areas of research are two different things. I think it's a mistake to assume one speaks for the other. That short-term switching is more discriminating is commonsense, but once learned, certain sound signatures can be remembered and recognized for years.

 

[andreasmaaan]

The tests I have seen - and I have looked hard - are from cognitive psychology, not audio reproduction. The areas of research are two different things. I think it's a mistake to assume one speaks for the other. That short-term switching is more discriminating is commonsense, but once learned, certain sound signatures can be remembered and recognized for years.

Not saying it’s wrong necessarily, but what do you base that on?

 

[Inner Space]

Not saying it’s wrong necessarily, but what do you base that on?

The examples usually cited are e.g. a violin - if you heard one for the first time in ten years, you would still know it was a violin. Or your mother's voice on the phone, and so on. In our world, some loudspeakers are like that. Suppose your formative years had been spent listening to original L100s ... you would recognize that sound even decades later.

 

[richard12511]

@MattHooper - again, genuinely interested, and not trying to give you a hard time ... but why do you need a switcher? First (and yes, this might be a personal hobbyhorse) there is actually no research that says you can't remember SQ impressions for a long time. And in your case, your honest and heartfelt report indicated you could remember your CJs' SQ for at least two days: "After a couple of days I got my CJ amp back ... Re-listened to a bunch of tracks on the Bryston, then threw in the CJ amps and .... Aaaaahh. There it was! That gorgeous, relaxed, richer, rounder, bigger sound I was used to."

So wouldn't it be enough to have a family member switch connections, maybe on a daily basis? You could listen all evening, and note down "Bryston" or "CJs". Then at the end of the week (or month, or whatever) you could check your score. Wouldn't that work? You wouldn't even need to level match either - I assume you didn't just prior to your "Aaaaahh" moment? Yet the difference was immediately apparent.

Why wouldn't you need to level match? Wouldn't different levels make them sound different?

 

[richard12511]

The examples usually cited are e.g. a violin - if you heard one for the first time in ten years, you would still know it was a violin. Or your mother's voice on the phone, and so on. In our world, some loudspeakers are like that. Suppose your formative years had been spent listening to original L100s ... you would recognize that sound even decades later.

Those are big differences, though, which won't be the case in this test(most likely). If, decades after listening to the L100s you heard an L100 that had its 62Hz frequency attenuated by 0.1dB, do you think you'd be able to correctly identify that? My guess is no, you'd probably - incorrectly - think it's the same speaker.

 

[Blumlein 88]

https://www.audiosciencereview.com/...ity-and-reliability-of-abx-blind-testing.186/

This describes comparing long term vs short term perception of 2.5% distortion.

 

[Inner Space]

If, decades after listening to the L100s you heard an L100 that had its 62Hz frequency attenuated by 0.1dB, do you think you'd be able to correctly identify that? My guess is no, you'd probably - incorrectly - think it's the same speaker.

I'm certain you're right - I would. Just like I would know it was my mother on the phone, even if she had a heavy cold. Recognizing a "signature" is clearly a complex thing. But the issue for me is the incorrectly received conventional wisdom that we have really short auditory memories. I think we're good for much longer than was wrongly extrapolated from a test of something else entirely.

 

[andreasmaaan]

The examples usually cited are e.g. a violin - if you heard one for the first time in ten years, you would still know it was a violin. Or your mother's voice on the phone, and so on. In our world, some loudspeakers are like that. Suppose your formative years had been spent listening to original L100s ... you would recognize that sound even decades later.

Ok, I understand you now. Perhaps we understand the "conventional wisdom" differently, but my understanding is not that our ability to discriminate drops inevitably to zero as time passes, but rather, the more conservative claim that it simply drops (to some lower level that may not be zero). I think that's fairly uncontroversial, right?

I haven't heard it claimed that we completely lose our ability to discriminate between all stimuli once sufficient time has passed (death notwithstanding).

 

[Inner Space]

https://www.audiosciencereview.com/...ity-and-reliability-of-abx-blind-testing.186/

This describes comparing long term vs short term perception of 2.5% distortion.

Thanks, and understood, but my narrower point was - this needn't be conducted as an ABX between two close variables. That's total overkill. This is about a guy recognizing the sound of the CJ amps he's lived with and loved for years. He's already confident he's good with a two-day gap. Why does he need instantaneous switching? With cooperation from a family member, all he needs to do is listen all evening and check one of two boxes: CJs or Not CJs. My wider point was, if this guy can remember the sound of his amps after two days, why do we pretend our auditory memory is four seconds long?

 

[Inner Space]

Ok, I understand you now. Perhaps we understand the "conventional wisdom" differently, but my understanding is not that our ability to discriminate drops inevitably to zero as time passes, but rather, the more conservative claim that it simply drops (to some lower level that may not be zero). I think that's fairly uncontroversial, right?

I think that's exactly right. But the question is implied, isn't it - at which "lower level that may not be zero" does our ability to discriminate become no longer useful, in a practical sense? After four seconds have passed? Four minutes? Hours, days, years?

 

[SIY]

I have an ADC and Audition... my ADC only has line level or MIC level inputs.... How would I hook up to speaker level inputs ?

Line level. And you'll want to do a 5 or 10:1 attenuation unless you're listening quietly or have efficient speakers; most ADCs tolerate 4V max before reacting in ill humor. The value isn't critical as long as the resistor ratios are well-matched between channels. If your ADC has input volume controls that can knock down the measured voltage, you're gold.

If you need the attenuator, the cheap and easy way is to buy a couple of 47-50k resistors (matched to 0.1% or so), get RCA cables with connectors that can be unscrewed, unscrew them on the ADC end, disconnect the "hot" lead from the center pin of the RCA plug, then solder the resistors in series with the "hot" lead and the center pin of the RCA plug.

If your ADC is balanced, it can get slightly trickier.

 

[Blumlein 88]

Thanks, and understood, but my narrower point was - this needn't be conducted as an ABX between two close variables. That's total overkill. This is about a guy recognizing the sound of the CJ amps he's lived with and loved for years. He's already confident he's good with a two-day gap. Why does he need instantaneous switching? With cooperation from a family member, all he needs to do is listen all evening and check one of two boxes: CJs or Not CJs. My wider point was, if this guy can remember the sound of his amps after two days, why do we pretend our auditory memory is four seconds long?

He believes he hears a difference after 2 days. He hasn't proven that he has. So we go with the quick switching to make it as easy as possible to hear a difference. If a surprise result is he can't hear a difference, then it is unlikely he actually hears a difference after 2 days though that could be tested next.

 

[andreasmaaan]

I think that's exactly right. But the question is implied, isn't it - at which "lower level that may not be zero" does our ability to discriminate become no longer useful, in a practical sense? After four seconds have passed? Four minutes? Hours, days, years?

I guess that depends on the stimulus, right? I mean, you kind of suggested the answer in your earlier post I think

So, I would guess that I'd be better at discerning a recording of my mother's voice from recordings of other similar voices if I'd just spent the week at my mum's place vs if I hadn't heard her voice for two decades, but I would still probably do a reasonable job at discerning her voice in the latter case (or at least, a better job than someone else who had only heard her voice once). So in that sense my ability to discriminate might always remain somewhat useful. But if I were being asked to discriminate between two amplifiers powering a particular pair of speakers, the "useful" extent of my auditory memory might only be a few minutes, or even a few seconds (if indeed it were any use at all).