That's fine if bipolar and wanting to do 300 MPH so it has a good change of working at 150 MPH. I would liken MOSFET's to diesel. Same holes and banging bits, different outcomes.
That might be Moth to the flame aversion? MOS FET are a bit more flame proof. SOA is not plagued by nasty secondary breakdown. What you see is what you get , it says 8 amps and you can have 8 amps. Their naughty problem is gain doubling. That can be minimized in good old Dvv practice of over biasing. 100 mA is often quoted as ideal.
Have you ever dated a girl who would bring here friend along? Years later you realize you liked the friend more, all she needed was a haircut and some lippy. Same thing here. You were that close to finding out. Been there , done that.
FET-BJT OPS
Well..... there seems to be a universal agreement that fet OPS and bjt OPS sound different. IMO that difference is related to speed and current capabilities, respectfully. So, IF you used the compound-complimentary OPS (D.Self seems to like it also... see his book for details as to the why).... you could run the OPS first transistor as a MOSFET at high enough idle to provide the fast, low power speed with the second transistor as a BJT to provide the max current demands. This will go some ways to having the best of both worlds in one amplifier.
Try it.
THx-RNMarsh
Well..... there seems to be a universal agreement that fet OPS and bjt OPS sound different. IMO that difference is related to speed and current capabilities, respectfully. So, IF you used the compound-complimentary OPS (D.Self seems to like it also... see his book for details as to the why).... you could run the OPS first transistor as a MOSFET at high enough idle to provide the fast, low power speed with the second transistor as a BJT to provide the max current demands. This will go some ways to having the best of both worlds in one amplifier.
Try it.
THx-RNMarsh
Last edited:
Stereophile seemed pretty sure that their pick of SOTA tube and speakers sounded different with Carver's $600 SS amp - any remember how that shook out after he nulled the output response with some output Z padding, feedback tweaks to match the tube amp?
are tubes more similar to Carver's output stage than MOSFET are to BJT?
why should any having up on read psychoacoustics, sensory psychology testing experiment design give any credence to such generic assertions of broad class difference in "sound" correlating with XX component type when the most vociferous proponents of these sound differences actually refuse scientific controls
if you want to advance the thread topic "Sound quality vs Measurements" - the overwhelming current lack is valid listening tests, properly controlled, of "Sound Quality"
are tubes more similar to Carver's output stage than MOSFET are to BJT?
why should any having up on read psychoacoustics, sensory psychology testing experiment design give any credence to such generic assertions of broad class difference in "sound" correlating with XX component type when the most vociferous proponents of these sound differences actually refuse scientific controls
if you want to advance the thread topic "Sound quality vs Measurements" - the overwhelming current lack is valid listening tests, properly controlled, of "Sound Quality"
Last edited:
We refuse 'scientific controls' because they keep people from hearing differences between amps once a few MINOR details are matched. However, in typical listening situations, the difference between the amps is still there. Yes, they never went away, it is just that the 'test' obfuscated
the differences between the amps, and they are still there in the real world. Now, if Bob Carver is so concerned with Stereophile, look at what he is building and selling today. (tubes)
the differences between the amps, and they are still there in the real world. Now, if Bob Carver is so concerned with Stereophile, look at what he is building and selling today. (tubes)
Here is the scoop ---
I think many of us have read about the brain filling in missing data.... such as when producing a 2H only and the brain gives us the fundamental.... esp with bass. Similar effects occure in the brain with visual info.... we can add colors that are not there. When we say we hear a freq or see a color that isnt there, it cannot be physically measured, of course. This happens in listening evaluations as well;
When you listen to music, you add in things which are not actually there.... your personal experience with sounds gives us the fill-in. So, though we are each listening to the same recording in the same invirnment, we may not be actually experiencing it exactly the same way. But, when testing A and B back and forth, the brain also does a fill-in and you get a single picture or sound that is a fill-in composite of the two.
This is hard to hear such audio fill-in over email but you can get the sense of what I mean by audio fill-in by playing with visual images and see how your brain changes the 'reality (measurable).
When we go to measure what we perceive, because our brain supplied the sound/visual data, it isnt there to be found. Same with the DBT methodology when dealing with perceptions. DBLT are fine for drugs with measurable results -- like a GO or NO GO result. etc. Things which do not involve the perceptual aspect of our reality.
The Negative Photo Illusion in Psychology
The brain does similar but different things with what we hear and perceive and when combined into a DBT, you get a single composite result of the two.
If you do compare two items by listening, it is the first impression that is the most accurate. longer back and forth will meld into one composite perception. And, no difference thereafter will be perceived. Another approach is long term listening to one and then long term listening to the other. That prevents the perceptual composite and again, at first change back to the other is when the greatest difference is heard.
THx-Richard Marsh
I think many of us have read about the brain filling in missing data.... such as when producing a 2H only and the brain gives us the fundamental.... esp with bass. Similar effects occure in the brain with visual info.... we can add colors that are not there. When we say we hear a freq or see a color that isnt there, it cannot be physically measured, of course. This happens in listening evaluations as well;
When you listen to music, you add in things which are not actually there.... your personal experience with sounds gives us the fill-in. So, though we are each listening to the same recording in the same invirnment, we may not be actually experiencing it exactly the same way. But, when testing A and B back and forth, the brain also does a fill-in and you get a single picture or sound that is a fill-in composite of the two.
This is hard to hear such audio fill-in over email but you can get the sense of what I mean by audio fill-in by playing with visual images and see how your brain changes the 'reality (measurable).
When we go to measure what we perceive, because our brain supplied the sound/visual data, it isnt there to be found. Same with the DBT methodology when dealing with perceptions. DBLT are fine for drugs with measurable results -- like a GO or NO GO result. etc. Things which do not involve the perceptual aspect of our reality.
The Negative Photo Illusion in Psychology
The brain does similar but different things with what we hear and perceive and when combined into a DBT, you get a single composite result of the two.
If you do compare two items by listening, it is the first impression that is the most accurate. longer back and forth will meld into one composite perception. And, no difference thereafter will be perceived. Another approach is long term listening to one and then long term listening to the other. That prevents the perceptual composite and again, at first change back to the other is when the greatest difference is heard.
THx-Richard Marsh
Last edited:
Ultimately the difference leads us to our listening preference. And what is for one could be different to another all together.
I wouldn't say that we refuse 'measurements under control', but that we always search to perfect them even further, improve them, in a direct relation to our hearing and emotional impact. ...A very tough challenge, impossible to reach, ever. IMO
Yes, it is long term listening to components that is most effective in allowing differences to be heard. Skeptics should try it for themselves, with an open mind.
nothing in the psychoacoustics required controls prohibits long term listening - but level matching, blinding are critical
there has been at least 30 years for someone to do these long tem listening tests to show results since the debates in the JAES
as for Carver selling tube amps, I'm sure he recognized the stock investing truism "the market can remain irrational longer than you can remain solvent" has another practical interpretation
there has been at least 30 years for someone to do these long tem listening tests to show results since the debates in the JAES
as for Carver selling tube amps, I'm sure he recognized the stock investing truism "the market can remain irrational longer than you can remain solvent" has another practical interpretation
Yes, i know all about those things. I've been a IEEE and AES member for ever. [and many others]. I suppose nobody really cares, is all I can think of. I dont. I did the tests and other types and then moved on.
If you can grasp what I just said about the mind/brain adding and filling in based on memory of sound and experiences etc etc.... you to can 'get it.' re DB comparison tests... which involve unmeasurable (with electronic test equipment) qualities of perception.... causing both to sound similar.
Figure it out, people. I gave a huge clue.
THx-RNMarsh
Thanks for that very nice rundown, Richard - as you say, people don't "get it", and part of the problem is that they don't want to get it: it complicates the nicely sorted out picture they have developed of what the audio 'scene' is all about ...
The answer I have developed, for myself, which is very effective, is to use recordings which have aspects of the reproduction which are on the edge of being acceptable. An 'excellent' recording is completely useless for assessing differences - for the reasons you mention - a 'good' recording is almost as bad, but a 'bad' recording can make things as easy as pie - because the objectionable aspect stands out in such strong relief that a change in the nature of that aspect immediately registers, has strong impact. The latter characteristic is the only thing I listen to when assessing, the 'good bits' of the playback are completely irrelevant, I totally ignore that.
Notwithstaninding what Richard has pointed out, the simple fact of life is that audio in general is well past its heyday, the time it was considered an art form and was treated as such. It was not a pastime for the masses, it was pursued by relatively few.
Once it became a hobby of the masses, the general level of performance was allowed to slide down, or remain on more or less the same level, in the name of price.
Thus, the price, and the looks, are the main crietria today. It has to look like it costs a cool million, but it has to be cheap or at least affordable. And if in the process it should also actually play music, well, so much the better.
Just look how many components (FETs, transistors) which were well known and well regarded, are made no more. At some point, they became too rare and were dropped by their manufacturers. Not replaced with the bext generation, but dropped altogether. Not enough volume because not enough interest.
In my view, this is also a part of the whole the brain has to integrate into its conclusion. It looks good, it weighs a ton, it has a famous name and it costs a lot - therefore, it must be good.
The truth of course is quite different. But, who cares?
Once it became a hobby of the masses, the general level of performance was allowed to slide down, or remain on more or less the same level, in the name of price.
Thus, the price, and the looks, are the main crietria today. It has to look like it costs a cool million, but it has to be cheap or at least affordable. And if in the process it should also actually play music, well, so much the better.
Just look how many components (FETs, transistors) which were well known and well regarded, are made no more. At some point, they became too rare and were dropped by their manufacturers. Not replaced with the bext generation, but dropped altogether. Not enough volume because not enough interest.
In my view, this is also a part of the whole the brain has to integrate into its conclusion. It looks good, it weighs a ton, it has a famous name and it costs a lot - therefore, it must be good.
The truth of course is quite different. But, who cares?
Last edited:
One more step: ... and it DOES sound good, the listener/customer enjoys fantastic sound, in his perception! The best he/she ever heard!
No it isn't. If you get to the point with all the perception factors, as you mentioned, it DOES sound the best he/she has ever heard! That's the step that is often left out.
Jan
I read a 1950's piece about making fake bass from OB using the organ builders trick. I know this trick so was fascinated. The amplifier deliberately made the fake harmonics to fool the brain. What it won't do is give the kick of true bass. Did anyone ever try this? To be clear we substitute the second harmonic at high amplitude and the others to match. My friend John has done the organ stuff so I might ask him.
The writer says 6 x 6 foot OB is 50 Hz - 8dB. That would be incorrect as it us room dependent. I support the general thought behind it.
The writer says 6 x 6 foot OB is 50 Hz - 8dB. That would be incorrect as it us room dependent. I support the general thought behind it.
Jan, I understand what you are saying, but while most can be fooled that way, some will not fall for it.
Those who THINK they are hearing the best sound ever are not infrequently stunned back to reality when their latest acquisition is compared to say the Otala/Lohstroh amp. Some will be sad because they will understand and accept that they have been taken to the cleaners, others will refuse to believe what they are hearing in favor of the small amp.
However, this is a clear cut move from sound and reality into the waters of pure psychology, and sometime full time psycho therapy.
To me, the point is that today, more than ever before, so many discussions have nothing to do with the actual sound we are hearing, but are in fact about personal perception. If so, it makes the hard reality somewhat meaningless, if you snag your customer with external details.
Those who THINK they are hearing the best sound ever are not infrequently stunned back to reality when their latest acquisition is compared to say the Otala/Lohstroh amp. Some will be sad because they will understand and accept that they have been taken to the cleaners, others will refuse to believe what they are hearing in favor of the small amp.
However, this is a clear cut move from sound and reality into the waters of pure psychology, and sometime full time psycho therapy.
To me, the point is that today, more than ever before, so many discussions have nothing to do with the actual sound we are hearing, but are in fact about personal perception. If so, it makes the hard reality somewhat meaningless, if you snag your customer with external details.
That mixed with a touch of current dumping perhaps? The Fetilington with the not required feed forward resistor does look possible. If we accept Mr Dvv's high bias even that seems OK for Fetlington. The more I read the more I agree with that. It is not a recipe for the best book spec. As no one on the planet should be able to hear the difference except to have a nicer sub one watt we should be careful to accept a better version of wrong as a wise choice? If I understand correctly Darlingtons have feed-forward to help empty the bases of charge. The fact it does a tiny bit of work and smoothing of switching is by the by.
If you choose to clobber the gate drive problems is personal. I would say listen first before you do.
Bose has used this trick in several products where they synthesize harmonics to give the impression that the equipment delivers very low freqs.
Jan
Darlingtons.
Did anyone ever do an optimum bias Darlington? That is where the No1 does less and No2 more? The would involve the bias to No2 being slightly different to normal practice. My conjecture is the No1 or No 2 might be hogging and if nothing else the gain is less than predicted. Into 1R 2 x 2N3055 as Darlington have gain of 5 in No2. Very important that is known. 2N3055 makes a very good 1 R 50 watt amp if that and SOA respected. Early effect is also spoken of if hogging happens. When wrapped in feedback I doubt anyone knows? Hogging is noted in capacitance multipliers where expected ripple rejection isn't realized. Also lower harmonics more noticeable than should be so.
Did anyone ever do an optimum bias Darlington? That is where the No1 does less and No2 more? The would involve the bias to No2 being slightly different to normal practice. My conjecture is the No1 or No 2 might be hogging and if nothing else the gain is less than predicted. Into 1R 2 x 2N3055 as Darlington have gain of 5 in No2. Very important that is known. 2N3055 makes a very good 1 R 50 watt amp if that and SOA respected. Early effect is also spoken of if hogging happens. When wrapped in feedback I doubt anyone knows? Hogging is noted in capacitance multipliers where expected ripple rejection isn't realized. Also lower harmonics more noticeable than should be so.
- Status
- Not open for further replies.