"Simple is beautiful" is Mr. Marsh' stance. That's fine as a preference. However, the cost argument doesn't, to me, hold water. No DIYer is that poor (not even myself!) not to afford 50p for ten more transistors. In particular after the heatsinks, case, power transformer, big elcos, were already procured.
This is the well known subjectivist position - it reflects a preference without any means or intention to quantify. Asking what exactly is "better SQ" and how would you extrapolate that to my perception would already be a futile exercise. Rather than that, I would think of a mental comfort in approaching a simple design that, at least at the first glance, looks manageable given the existing skills set.
Aesthetic satisfactions and the sense of comfort. Makes sense for any hobby, I suppose.
Wali
Probably one of the most common misconceptions about negative feedback.
Negative feedback is good where there is no influence of the reaction speakers at her. A resistive load NFB works great, no doubt. But when working on real load causing major problems. Unfortunately, not everyone understands the mechanism of these distortions.
regards
Petr
Probably one of the most common misconceptions about negative feedback.
Negative feedback is good where there is no influence of the reaction speakers at her. A resistive load NFB works great, no doubt. But when working on real load causing major problems. Unfortunately, not everyone understands the mechanism of these distortions.
regards
Petr
It appears you are in good company with this point of view ;-)
“Simplicity is the ultimate sophistication.” Leonardo da Vinci
“Perfection is achieved, not when there is nothing left to add,
but when there is nothing left to remove." Antoine de Saint-Exupéry
a) "Simple sounds better" the same-o-same subjective stance that nobody was ever able to prove.
b) Logical fallacy. Assuming "simple sounds better" is always true, all you could tell is that "what sounds worse is not necessary complex". Your "complex doesn't sound better" has absolutely no justification based on your original assertion; it is neither proved true or wrong. As much as the original "simple sounds better" which also remains only wishful thinking.
The multi-tanh input stage is a good example of open loop linearizing. It falls under the Cherry "bias increasing" type, unfortunately it would be difficult to apply to discrete circuits.
ADA4898 uses this technique and achieves well under 0.0001% distortions at 20KHz (both HD2 and HD3). If something like the LME49810 would be implemented using this technique, that would beat any two gain-stage conventional design circuit topology, hands down. Not going to happen, unfortunately.
Last edited:
Each time (including in this thread a few weeks ago) I asked people claiming that NFB has problems on real loads, to provide real facts , there was strictly not the slightest return apart that I was told I am ignorant of the basics of loudspeaker behavior.
This is a mischaracterization of my position. Better SQ to me isn't merely a preference, and I do have an intention to quantify it. Just its taking me time to work out how best to do that.
As you wish then - I shan't share my speculations and hypotheses as to what better SQ actually is.
Hi Petr,
I am sorry, but I also view what you have said here as a misconception. Properly designed, NFB handles the complex speaker load fine. This misconception was yet another thing "discovered" and spread by Matti Otala when he coined the term Interface Intermodulation Distortion (IIM). IIM does exist and can be measured, and I have done it, but it is not made worse by negative feedback. See the paper debunking this on my web site at CordellAudio.com - Home. While you are at it, see also my paper debunking most of what Otala said about Phase Intermodulation Distortion (PIM). There is also a section in my book, Designing Audio Power Amplifiers, that discusses many of the misconceptions about negative feedback.
Cheers,
Bob
Once that a performance metric is well defined (call it, "requirements") then obviously the "simplest" (in terms of cost, effort, timelines) solution always wins.
The problem is, you and others are trying to put "simple" as a requirement, before defining and justifying SMART (as in Specific, Measurable, Achievable, Realistic and Time-bound) objectives.
I believe that most over-broad generalizations are NOT a sign of maturity. Simpler is better is such an overly-broad generalization. That might mean that you would use a resistive load on the IPS of a VFA instead of a current mirror - not a good idea, in my opinion. It might also mean that you would not help the VAS transistor with an emitter follower in front. It might also mean that you would use the old electrolytic bootstrap circuit for the VAS load instead of a current source. It might also mean you would use a double instead of a triple output stage. It might also mean that you would never use a capacitance multiplier to clean up the rails. On and on and on.....
I think it is admirable as an exercise to see how much you can get out of the fewest parts, but that does not necessarily support a generalized simpler is better position.
Would the simpler is better advocates dispense with airbags and anti-lock brakes in their cars?
Cheers,
Bob
You are barking up the wrong tree. Its just some people observation from their listening and other qualities which they appreciate. I respect that. The R and T part are up to the individual and is subjective. Your own SMART requirements are yours to use and keep as you see fit. And, I respect your priorities and values, too.
Thx-RNMarsh
Last edited:
I have stated many times through out this and else where what I mean by simplifying. Others have their own meaning. Pls don't take anything out of context and add meaning to it that isn't there... at least don't apply it to me.
If anyone wants to go deeper, we can go much much deeper. That's not my intention here.
Thx-RNMarsh
Last edited:
Not overly broad meaning in the quoted individuals nor mine useage. The word taken literally is Not what I nor they had in mind. You really do not know how they are using the Concept of simplification? Or, coding being made compact can be simpler but still powerful?? Really?
Maybe this will work - using cars; The disk brake is an improvement over the drum and shoes. It is simpler, lighter, easier and more sophisticated. If that doesn't work for some... I cant explain it to you.
BUT don't anyone say that I said VFA was not sophisticated... it can be, for sure.
At the same time, there is no need for anyone to put down the CFA topology for audio as it can be simple yet sophisticated or it also can be as sophisticated as one wants to pursue.
Thx-RNMarsh
Last edited:
So another vote for (almost) pure subjectivism. Engineers are not supposed to educate, but to give them what they want (or they think they want).
Given the responses, I'm no longer surprised that a piece of crap like the original VSSA sells well. It's all about subjective perception and, if a good dose of gregarious behaviorism is added to the mix, it is no longer difficult to understand why the high end audio business is today so full of it. And when the "it" is attempted to be sold as competent engineering practice, then it really starts to stink.
Not unexpected, but sad indeed. Thank you, I'm done.
Last edited:
Waly,
I can see your point of view. I spent the better part of my life in high level R&D and it was all about physics, electronics and facts and tests and measuremnets and calculations et al..... I know that world very well. But as I progressed into and thru Sr project management, I learned to have more respect for others opinions and let each person find their own path to their goals -- with some guidance and focus (and budget), of course.
You might be surprised to know I have now and always have been into test and measurement. I continue to buy and use and modify test equipment. And, I have used and own many software T&M and SIM (MicroCap 1-10). So, lets not pretend anyone knows all about what I think by a few words here in a DIY forum.
I respect and learn from all views.... and backgrounds. And, I travel a lot to do same with engineers, manufactures and just poor folks in remote villages. That's is why I take a broader view and am more accepting of others perceptions as input. There are many forms and you have to figure out what it can mean and what to do with it.
Enough with the lecturing.
Thx-RNMarsh
Thx-RNMarsh
I can see your point of view. I spent the better part of my life in high level R&D and it was all about physics, electronics and facts and tests and measuremnets and calculations et al..... I know that world very well. But as I progressed into and thru Sr project management, I learned to have more respect for others opinions and let each person find their own path to their goals -- with some guidance and focus (and budget), of course.
You might be surprised to know I have now and always have been into test and measurement. I continue to buy and use and modify test equipment. And, I have used and own many software T&M and SIM (MicroCap 1-10). So, lets not pretend anyone knows all about what I think by a few words here in a DIY forum.
I respect and learn from all views.... and backgrounds. And, I travel a lot to do same with engineers, manufactures and just poor folks in remote villages. That's is why I take a broader view and am more accepting of others perceptions as input. There are many forms and you have to figure out what it can mean and what to do with it.
Enough with the lecturing.
Thx-RNMarsh
Thx-RNMarsh
Last edited:
I doubt it, John, but I'll wait and see. Before you claim to have debunked it, make sure you read my paper again thoroughly and completely. Bear in mind, I have never said there is no such thing as PIM (anyone who knows about differential gain and phase should know about PIM).
BTW, who is "we"? I am well aware of Ron Quan's work. He is a nice, intelligent and courteous guy, but his work thus far does not debunk my work.
And please don't name-drop my old friend Barrie Gilbert. His paper also does not debunk my work and as near as I can tell was never intended to.
I enjoyed seeing your old cantankerous self at RMAF
.Cheers,
Bob