An interesting question.
Below is Mooly's post.
I share this thinking, and have thought about it myself.
So, the question initially breaks down to a 50-50 relationship.
A two part system, stage A and stage B. The "interconnect" between them is not a cable, but a recording medium. So actually it is a Three Part system. Stage A (the microphone & recording electronics); Stage B (the recording medium and all it entails); Stage C (the playback electronics and speakers)
Now the conjecture suggested is that given that the usual case is that the Stage A side utilizes non-optimum parts (opamps in particular, but also caps and resistors, etc...), Why, how could an "optimization" of Stage C lead to an "improvement" in the perceived sound via being more linear, lower THD, etc??
IMPORTANTLY, can conversely, IF an experimental setup was created where Stage A was the highly "optimized" one, and Stage C was the "non-optimum" one, would the sonic results be the same??
How about if both Stage A and C were "optimized"?
For those who feel that it all makes no difference, and that it's indistinguishable, then please if you must say so, just say it simply, and let others discuss the concepts. Thanks.
_-_-
Below is Mooly's post.
I share this thinking, and have thought about it myself.
So, the question initially breaks down to a 50-50 relationship.
A two part system, stage A and stage B. The "interconnect" between them is not a cable, but a recording medium. So actually it is a Three Part system. Stage A (the microphone & recording electronics); Stage B (the recording medium and all it entails); Stage C (the playback electronics and speakers)
Now the conjecture suggested is that given that the usual case is that the Stage A side utilizes non-optimum parts (opamps in particular, but also caps and resistors, etc...), Why, how could an "optimization" of Stage C lead to an "improvement" in the perceived sound via being more linear, lower THD, etc??
IMPORTANTLY, can conversely, IF an experimental setup was created where Stage A was the highly "optimized" one, and Stage C was the "non-optimum" one, would the sonic results be the same??
How about if both Stage A and C were "optimized"?
For those who feel that it all makes no difference, and that it's indistinguishable, then please if you must say so, just say it simply, and let others discuss the concepts. Thanks.
_-_-
Last edited:
Mooly is spot on again, if we could achieve true wire with gain replay most folks probably wouldn't like the result . We could talk at length about optimized systems but that's already been done, repeatedly and to distraction by Frank.
People would be "rolling wire" to try to improve the sound.
Oh, wait, they already do that.
No, no, no.
Look at what I said, please.
Try to focus on the main thesis... as stated above.
The relationships between the 3 parts, and what might happen when you address them in the manner suggested??
Aka, perhaps this will illuminate reasons that somethings "sound better".
the ideas do not revolve around how people do things.
Look at what I said, please.
Try to focus on the main thesis... as stated above.
The relationships between the 3 parts, and what might happen when you address them in the manner suggested??
Aka, perhaps this will illuminate reasons that somethings "sound better".
the ideas do not revolve around how people do things.
Some people believe in an idea that I will call Euphonic Coloration, viz., it is possible for a less-than-perfectly-neutral playback system to produce greater listening enjoyment than a perfectly neutral playback system.
Such people would be happy to include this type of non-neutrality (distortion) all along the recording, storing, and playback chain.
One possible observation of Euphonic Coloration may have been the Stereophile Carver Challenge link. Or it may not; opinions vary.
Such people would be happy to include this type of non-neutrality (distortion) all along the recording, storing, and playback chain.
One possible observation of Euphonic Coloration may have been the Stereophile Carver Challenge link. Or it may not; opinions vary.
Since the goal's transparency then consider each stage as a pane of glass. The total transparency is the three panes of glass stacked - a change in transparency of pane C is going to be noticeable even when C is less opaque than A.
Ok, so if the light source is behind side A, does it make any difference what the opacity or distortion in the glass is in either position, IF the end result (the viewer) is similar or identical?
Assuming "A" is a given, and less than "transparent" to use your term, how much less than full "transparency" can be tolerated in subsequent layers before it becomes:
-detectable
-tolerable
-objectionable
??
I think it is more useful to think of it as the way 200 yo glass looks, which is non-flat and distorted like a bad lens.
(it's an imperfect analogy, but useful for conceptual investigation)
Assuming "A" is a given, and less than "transparent" to use your term, how much less than full "transparency" can be tolerated in subsequent layers before it becomes:
-detectable
-tolerable
-objectionable
??
I think it is more useful to think of it as the way 200 yo glass looks, which is non-flat and distorted like a bad lens.
(it's an imperfect analogy, but useful for conceptual investigation)
Last edited:
One complexity with your model is that with 'modern' multitrack techniques, A and B are welded together, and the result, the delivered signal, is monitored in the studio with their own version of C. The mix engineer will listen to A->B->C and make that sound ideal, and then the home listener will use their 'D', making their chain A->B->D
So, the home version D compared to the studio version C will be the essential difference.
Over the years, both C and D have become more similar and generally closer to neutral, but there are a number of odd exceptions. For example, music made for discos where their D has a big but somewhat predictable LF boost. Or, where D is the home playback used for TV and video.
Anyway, just thought I'd toss that into the mix, since simple, pure recordings done with mike arrays in a nice recording space are not all that we listen to!
So, the home version D compared to the studio version C will be the essential difference.
Over the years, both C and D have become more similar and generally closer to neutral, but there are a number of odd exceptions. For example, music made for discos where their D has a big but somewhat predictable LF boost. Or, where D is the home playback used for TV and video.
Anyway, just thought I'd toss that into the mix, since simple, pure recordings done with mike arrays in a nice recording space are not all that we listen to!
Sure, but not terribly relevant to the "investigation" since it is useful to examine the 3 part system before introducing any additional variables.
Clearly "pure recordings" form the baseline, and for all intents and purposes whatever preceded the receipt of the media by the end user (us) can be seen as "A"... we're mostly interested in first considering the orginal conjecture which revolves around the issue of "op-amps" and their effects or lack of...
Clearly "pure recordings" form the baseline, and for all intents and purposes whatever preceded the receipt of the media by the end user (us) can be seen as "A"... we're mostly interested in first considering the orginal conjecture which revolves around the issue of "op-amps" and their effects or lack of...
In the case of glass, I'm not sure but would think it doesn't.
In the case of electronics I'm sure that there's going to be 'distortion of distortion' hence I reckon the opacity is best left to the end (nearest the listener). Keeping the opacity at the back-end of the chain means nothing distorts the distortion of the final stage.
Yep, like all analogies has its limits but is useful within those limits.
Hi Bear,
The original signal has passed through tape, outboard gear, mixing console to the master. Then it's over to the replication process. Some of the outboard equipment will see the signal more than once through, and that equipment could easily be tube (for it's sound) or even discreet transistors with varying amounts of distortion. Those 1950 ~ 1970 designs can be pretty horrible, and some of that stuff hasn't been serviced.
Really, I honestly believe that beyond a certain point, the playback system in a home isn't going to make a difference unless the quality falls below a certain level. That will vary by studio and engineer - producer. Now the real crying shame is that most equipment does fall short these days. So OP Amps? Using any modern ones that are good will be transparent as far as the total path is concerned.
Once again, it is the execution of a design that can matter more than the parts used. Would you agree with that bear?
-Chris
The original signal has passed through tape, outboard gear, mixing console to the master. Then it's over to the replication process. Some of the outboard equipment will see the signal more than once through, and that equipment could easily be tube (for it's sound) or even discreet transistors with varying amounts of distortion. Those 1950 ~ 1970 designs can be pretty horrible, and some of that stuff hasn't been serviced.
Really, I honestly believe that beyond a certain point, the playback system in a home isn't going to make a difference unless the quality falls below a certain level. That will vary by studio and engineer - producer. Now the real crying shame is that most equipment does fall short these days. So OP Amps? Using any modern ones that are good will be transparent as far as the total path is concerned.
Once again, it is the execution of a design that can matter more than the parts used. Would you agree with that bear?
-Chris
Good post Chris. I'd also point out that most musicians are not audiophiles and neither are most recording technicians so sound quality is often neglected. Obsessing about replay will often lead to frustration as the source material has all it's flaws highlighted by a transparent system.
Years ago i had a tube mini amp/headphone amp, the sound through my cans was amazing in many ways, dynamic, exciting, bass had real punch and slam. I just couldn't live with the mains hum during quiet moments. It also lacked fine detail compared to virtually any decent sand amp but it was captivating in a way i hadn't heard before or since. Wish i'd kept it now, tube savvy people on here could probably have helped me mod it to reduce hum.
Years ago i had a tube mini amp/headphone amp, the sound through my cans was amazing in many ways, dynamic, exciting, bass had real punch and slam. I just couldn't live with the mains hum during quiet moments. It also lacked fine detail compared to virtually any decent sand amp but it was captivating in a way i hadn't heard before or since. Wish i'd kept it now, tube savvy people on here could probably have helped me mod it to reduce hum.
The question of the "quality" of the source really isn't so important to the discussion, except to note that there is a range of qualities possible, to simplify:
-- low, as in inherently flawed
-- normal, as in better than low and not as good as:
-- optimal, the best possible with SOTA methods
Clearly, if optimal A and optimal C are used, and negating (for the moment) effects of the medium (B), then this would be the best possible and some sort of control or standard by which to adjudge.
So the question that may reveal something interesting is where one compares non-optimal A + optimal C with optimal A + non-optimal C. Will they be equivalent?
And no anatek, my personal experience is that I keep hearing fairly clear "variations" in the perceived sonics between opamps and other components... execution certainly plays a role too. That's why I'm trying to talk about these things, hoping to vector in on a way to characterize this via some sort of viable means/test(s).
One idea is that for the "C" part one needs to accomplish either not adding more of the existing artifacts in the same type and manner AND/OR doing something that may include ameliorating said artifacts (aka "reverse filtering").
Of course if you think there are no artifacts produced or present back in the "A" section (or even in the "B" section) then there is nothing to talk about.
--------
davym, the question that needs to be resolved is IF the "sand amp" actually has "fine detail" or if that is actually A) something added, or (interestingly) something lost?? (clearly the tonal balance seems to be different)
-- low, as in inherently flawed
-- normal, as in better than low and not as good as:
-- optimal, the best possible with SOTA methods
Clearly, if optimal A and optimal C are used, and negating (for the moment) effects of the medium (B), then this would be the best possible and some sort of control or standard by which to adjudge.
So the question that may reveal something interesting is where one compares non-optimal A + optimal C with optimal A + non-optimal C. Will they be equivalent?
And no anatek, my personal experience is that I keep hearing fairly clear "variations" in the perceived sonics between opamps and other components... execution certainly plays a role too. That's why I'm trying to talk about these things, hoping to vector in on a way to characterize this via some sort of viable means/test(s).
One idea is that for the "C" part one needs to accomplish either not adding more of the existing artifacts in the same type and manner AND/OR doing something that may include ameliorating said artifacts (aka "reverse filtering").
Of course if you think there are no artifacts produced or present back in the "A" section (or even in the "B" section) then there is nothing to talk about.
--------
davym, the question that needs to be resolved is IF the "sand amp" actually has "fine detail" or if that is actually A) something added, or (interestingly) something lost?? (clearly the tonal balance seems to be different)
Nothing added just the fine macro detail you get from solid state done well but with a flatter, less dynamic sound. I wish i could get the best of both (tube & SS) without having to pay off someones mortgage for them.
We all would prefer that the 'best' in solid state or tube designs was easily affordable. I can only 'afford' most of my designs, because I am given a working unit, by previous agreement, before I design it. I still don't have a Constellation product in my hi fi system, they are just 'too much' for my living room. The reason that we make these expensive, exotic products is to see where our BEST efforts will get us, that is the satisfaction we get from designing the stuff.
Hi bear,
We don't disagree on op amps at all. I can hear differences between them. It's just that with average quality program material, using an NE5532 or better isn't going to damage anything much more than it has already. The TL07x and TL08x are particularly nasty in what they can do to the sound quality. I'm calling those out because I see lot's of them in use as an "upgraded" part.
I think that unless the stages A & B are optimized, you are going to find the differences between op amps are less obvious. They can be heard, especially if you can switch them in and out quickly. But, you still need a clean source in order to judge step C effectively. Sadly, situation C is the only one we have control over. All we can do is make it the best it can be and deal with what we are given to reproduce.
-Chris
We don't disagree on op amps at all. I can hear differences between them. It's just that with average quality program material, using an NE5532 or better isn't going to damage anything much more than it has already. The TL07x and TL08x are particularly nasty in what they can do to the sound quality. I'm calling those out because I see lot's of them in use as an "upgraded" part.
I think that unless the stages A & B are optimized, you are going to find the differences between op amps are less obvious. They can be heard, especially if you can switch them in and out quickly. But, you still need a clean source in order to judge step C effectively. Sadly, situation C is the only one we have control over. All we can do is make it the best it can be and deal with what we are given to reproduce.
-Chris
Thanks for the honest appraisal John, i still have a JC2 clone knocking about and it is impressive for something that cost around the price of a steak diner in my local pub. I've never been able to afford true high end kit but have spent far too much in the past on mid range commercial gear which for the most part is garbage in a decent enclosure. I just wish that after enough time has passed (say 15yrs) for designer and manufacturer to recover costs and hopefully make a decent profit, that they would make their high end designs available at a more affordable price range. This should open up a new and much larger market after all.
No need to wish, you can just build your own.
Ought to be pretty simple to arrive a a very fine amp for your phones without significant cost at least after a few iterations, wherein you change/modify a few things...
But this is a different topic.
Chris, this is exactly the issue - having control over "C".
BUT, we do have *some* control over "A" if we record our own things, or purchase media ("B") that is known (by some means) to us. I have examples of both.
The entire situation is confounding, that is IF you do hear some opamps as sounding "better" than others.
Here's why: for me I find that the high performance opamps (AD797, etc...) do in fact sound rather a lot better in key areas than do 5532/4... I am perhaps very critical of the way that the 5-10kHz (nominal) range is reproduced. I'm looking for sound that is "natural" to the ear and where there isn't excess sibilence - especially if it manifests as a type of "smear/excess" beyond what a voice ought to.
So, now, IF I place the only opamp in my system between the DAC and the passive attenuator (no opamps in the JFET/Mosfet SE Class A amps) AND when I switch to the technically higher performance opamp (and then choose between the high performance ones for the "best sound") what does that say about what is going on before the opamp? That being the source, the DAC??
The result being that 1) the soundfield is clearer, easier to "see" and 2) the excess sibilence and other HF artifacts are substantially reduced and "resolved" toward a "natural" sound. Can it possibly be said that the higher performance opamps are in effect "lying"??
For the case that was just presented is the difference heard the result of adding the A + B distortions to more "in kind" distortions in "C"? (thus tipping over some inherent auditory threshold - at least in my ears?)
I can say that some recordings (CDs in the main) do sound noticeably "better", cleaner, exhibit far less HF artifacts (sibilence/cymbal splash and/or "single tone quality"), while some are quite frankly unlistenable in this DAC (or my older DAC which is still quite good) when using "typical opamps".
In short too much HF hash/trash/artifact or energy and I do not want to listen.
It's just not natural to my ears.
BUT, we do have *some* control over "A" if we record our own things, or purchase media ("B") that is known (by some means) to us. I have examples of both.
The entire situation is confounding, that is IF you do hear some opamps as sounding "better" than others.
Here's why: for me I find that the high performance opamps (AD797, etc...) do in fact sound rather a lot better in key areas than do 5532/4... I am perhaps very critical of the way that the 5-10kHz (nominal) range is reproduced. I'm looking for sound that is "natural" to the ear and where there isn't excess sibilence - especially if it manifests as a type of "smear/excess" beyond what a voice ought to.
So, now, IF I place the only opamp in my system between the DAC and the passive attenuator (no opamps in the JFET/Mosfet SE Class A amps) AND when I switch to the technically higher performance opamp (and then choose between the high performance ones for the "best sound") what does that say about what is going on before the opamp? That being the source, the DAC??
The result being that 1) the soundfield is clearer, easier to "see" and 2) the excess sibilence and other HF artifacts are substantially reduced and "resolved" toward a "natural" sound. Can it possibly be said that the higher performance opamps are in effect "lying"??
For the case that was just presented is the difference heard the result of adding the A + B distortions to more "in kind" distortions in "C"? (thus tipping over some inherent auditory threshold - at least in my ears?)
I can say that some recordings (CDs in the main) do sound noticeably "better", cleaner, exhibit far less HF artifacts (sibilence/cymbal splash and/or "single tone quality"), while some are quite frankly unlistenable in this DAC (or my older DAC which is still quite good) when using "typical opamps".
In short too much HF hash/trash/artifact or energy and I do not want to listen.
It's just not natural to my ears.
Last edited:
- Status
- Not open for further replies.