|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
72.52.96.59
In Reply to: RE: Do they? posted by morricab on January 29, 2015 at 05:32:01
"There was an intersting article from the early 70s regarding the degradation in sound quality of recordings that the authors had primarily traced to the replacement of tube gear with SS gear in many studios."
Yes, the solid state technology in the late 60s through the 70s was rather poor.
But it's not today.
Follow Ups:
I will cut to the chase with you, shall I? I have done extensive comparisons with what are considered to be the best SS amps on the market as well as with multiple Class D, tube, hybrids etc. SS is smoother and less grainy than from the past...but arguably even less interesting to listen to because the life, which was there occasionally in the older SS designs, is gone gone gone in most of the newest ones.
Which kinds of SS amps? McInstosh (newer ones not old ones), Musical Fidelity KW series, Krell (various models from old to newish), darTZeel, Soulution, Spectral, Moon, Halcro etc. Most sound pretty smooth, even a bit warm at times but none has the life I hear from the best tube amps. Ever heard a Halcro that was all the rage about 10 years ago?? Dreadful stuff but measured better than anything else on the planet...literally.
I currently have two amps: one is pure SET the other is a single ended hybrid with that uses a single big MOSFET on the output in Class A. They sound fairly different from each other but both have high resolution and great tonality. Where they differ is in the presentation of the stereo soundstage and image field and in bass quality. The hybrid has a distinct SS tight dry bass whereas the bass on the SET is actually tight but not as dry...it has very good output iron. It also has a bit more of that inner "life" maybe its just how the darn transistors deliver the signal...not sure but you can hear it...maybe it is the lack of output iron...but it doesn't sound like an OTL either (had two different models of those as well).
I suggest you do some reading about the effect design has on perception of sound rather than raw numbers, which only tell what something is but not the impact of what it does. I would argue that feedback doesn't so much suppress or eliminate distortion as it simply shifts it around (read the white paper from Nelson Pass and some of Norman Crowhurst's writings about signal modulated noise floors created by feedback).
Well as for reading I am a big fan of Douglass Self for solid state amps and Morgan Jones for tube amps.
Self has identified some very interesting distortions like from speaker protection relays. And he has documented proof as well. Not a big deal design wise but it shows how good the resolution of modern test gear as well as solid state amp design. I mean if you can separate out the minuscule distortion from relay contacts, the rest of the amp must be damn good in that respect.
"Self has identified some very interesting distortions like from speaker protection relays."
That's nice, now do you know if it has an actual sonic impact to a listener or is it just another number?
"I mean if you can separate out the minuscule distortion from relay contacts, the rest of the amp must be damn good in that respect.
"
Perhaps but it can still sound like crap to an experienced listener.
Seriously, you can throw all the engineering you want at me and I don't disagree with you in the sense that many of these things are measurable. Most in fact. BUT and it is the BUT, if you don't correlate it back to the listening experience then you have just performed an exercise in navel gazing.
I have the same issues in my work. I develop drug products for a major pharma company. We have specs for those drugs technical performance just like the specs for an amp. Mostly though, they are for quality control because nearly all of them have NOTHING to do with how that drug works or doesn't work in your body. That is why we have to do clinical trials because all the technical data in the world will not tell you if it is beneficial or not to the patient until you actually put it into the patient.
THEN you can start to see if you have correlations between the clinical effect, which is a VERY noisy signal and why so many people and trials must be conducted, and some technical data (like rate of dissolution) that is relatively controlled and has a high S/N. However, usually the technical test must be heavily modified to a particular system to get meaningful data in terms of correlation.
The point of my story is that all the technical data that you like to discuss and ask for proof is meaningless on it's own for determining sound quality. Its that simple and I can't help you if you don't understand this basic philosophical concept. It is useful on its own to tell if one unit out of a production line is faulty...QC in other words but that's it. Once you can correlate with what listeners find more realistic then you are getting somewhere and discussion of relay distortions or capacitor distortions or feedback distortions becomes interesting.
This is a SCIENTIFIC perspective not an engineering perspective. I and others want to make better SOUNDING gear...not better measuring gear per se (only if the decisions that get better measurements leads also to better sound). If it measures good and sounds bad then there is a flaw in the ENGINEERING concept chosen as a solution to eliminating distortion. The scientific perspective is to point out that the engineering solution does not lead to the desired output...good sound. Once engineering steers towards this concept then better sounding products will emerge (many are going this way even if it appears they are using "obsolete" techniques).
The medical industry still has vast unknowns. There have been great advancements but AFAIK, we still don't know much about how a living entity works. So of course medicine is a lot of trial and error.
But to say in 2014 that we don't know how a capacitor works and how to engineering around any shortcomings is ludicrous.
"The medical industry still has vast unknowns." Not when it comes to designing and making drugs...it is all well know chemistry and physics. What we don't know about behavior in the body is quite analogous to what we don't know about how people respond to sound when listening to music. So, quite apples to apples...I am disappointed in you that you seem to have so badly not got the point...I assumed you were a pretty smart guy.
"So of course medicine is a lot of trial and error."
And you think making good sound is any less trial and error...LOL!! At this stage it is no better than medicine...worse in fact because at least we have clinical trials that show if it really works or not. Since audio won't kill anybody if its wrong then those controls don't exist in audio.
"But to say in 2014 that we don't know how a capacitor works and how to engineering around any shortcomings is ludicrous."
Clearly you have missed the point. I am not comparing understanding of medicine in a patient to a capacitor. I am comparing how drugs behave in a body to how music affects a person. We know a TON about how to make drugs and how to measure them...just like audio measurements. What we don't know is how they will behave in a body...just like we don't really know how measurements will affect the listener's perception.
I can't really make it any clearer than that...either you don't get it because your focus is so narrow or you are willfully misunderstanding the point.
You are saying that listening to music is an experience of mental stimulation and no two people will hear exactly the same things. Is that right? And of course someone may like the sound of a tube amp versus the SS amp.
And you have to admit there are mental biases involved.
But if we mask the test to where the bias is removed. If we test something that is immeasurable by ear as far as we know. Such as replacing copper wire with silver hookup wire. I think you are saying there could be someone that can actually hear that miniscule level difference.
I can't disprove that but it just seems implausible considering the numbers involved, which by the way are not in dispute.
If the test subject has any idea, even a hint in some cases which is which, the test is no good. And while I don't know much about psychology at all I have read the people unknowingly very often give away cues with their body language.
Hell, I wouldn't even discount some form of unconscious ESP that is telling the test subject which is which. But based on what I know about countless human hearing studies, no person should be able to resolve a difference between copper and silver hookup wire. The difference while present and even measurable is just to minute. But I could be wrong too.
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: