Whatever the author, i believe, using a bad example, that you draw the wrong conclusion. I provided a link to one of my pages, where you will find a LT spice sheet of a real CFB amp. and its VFB version.
Use-it to explore deeper if you are more interested by the subject than arguing in forums,
read the Analog Device datasheets of video OPAMP and guess why the faster ones (1000V/µs, >100mhz) are CFB..
I remember a blind test we made, decades ago, in my recording studio, with other sound engineers, to try to evaluate the quality of the ATRAC algorithm. Comparing instant ( synchronized ) an original digital mix with an ATRAC copy of it.
The result was ATRAC was mainly preferred to the original
I don't agree that our ears are better than measurements instruments. Instruments are objective, and reliable. And you can see things that you don't even heard. (Yes we can hear things that we don't know how to measure)
Our feelings vary with time, humor, culture, on what we focus our attention, and even with what we dream to get. Our musical memory is nothing we can rely on.
The only problem with measurements, on my point of view, is to know what and how to measure and what kind of effects the results of measurements we see (levels, distortions, noise, bandwidth etc.) produce in our feelings.
Last, what is the purpose of "blind" ? Remove our bad faith ? Did-you lie to yourself ?
If something is obvious, you do not need blind tests. If a difference between two parts is so thin that you are not sure what is what, where is the interest of spending time to make a classification between them ?
But that is OOT.
CFB is not a religion. Only it appeared to me that all my preferred amps, since the 80s where Current feedback. Even the Studer i I had chosen for one of my studios, ignoring it.
EOT for me, chose your poison.
IMO, CFB for audio is inferior comparing to VFB from every angle when you've not built one properly and didn't give it a good listen. And the story changes when you do.
From the seat of the listener, I don't give a damn about their names and how many poles they remove and and what marketing hype they are contaminated with and how fast they are compared to the speed of light. All I seek is a sonic experience feeling close to reality. And in this criteria only my properly built CFB amps pass. End of story.
From the seat of the listener, I don't give a damn about their names and how many poles they remove and and what marketing hype they are contaminated with and how fast they are compared to the speed of light. All I seek is a sonic experience feeling close to reality. And in this criteria only my properly built CFB amps pass. End of story.
Elvee the groupies have arrived, youre done for.
Maybe its time you showed what modern VFB can do, Im sure youre familiar with VFB designed around CFB principles. With this more modern topology CFB has no chance in either spec sheet or one could argue sound quality as one or 2 high end manufacturers use the toplogie in their top amps. Cant argue bandwith, slewrate or zeroes or poles or anything else for that matter.
Maybe its time you showed what modern VFB can do, Im sure youre familiar with VFB designed around CFB principles. With this more modern topology CFB has no chance in either spec sheet or one could argue sound quality as one or 2 high end manufacturers use the toplogie in their top amps. Cant argue bandwith, slewrate or zeroes or poles or anything else for that matter.
It seems that the distinction between blind and double blind has not been made clear.
If you do a blind test it is similar to getting a group of people to measure something with a finely graduated ruler that is too short, averaging the results gives you a more accurate estimate, the trouble is it is wrong because the ruler is too short.
If however you take the sample sample size with another ruler as well , but this time it is about as much too long as the other is short, the average converges on the true mean because the measurements are randomly distributed either side of it instead of being biased to one side of it, this is a double blind test.
The unfortunate thing about these is that they tend to produce results that for instance indicate that people cannot hear the difference between VFB and CFB and so on, very inconvenient for anyone with a bias towards, and / or a commercial interest in one or the other.
rcw
If you do a blind test it is similar to getting a group of people to measure something with a finely graduated ruler that is too short, averaging the results gives you a more accurate estimate, the trouble is it is wrong because the ruler is too short.
If however you take the sample sample size with another ruler as well , but this time it is about as much too long as the other is short, the average converges on the true mean because the measurements are randomly distributed either side of it instead of being biased to one side of it, this is a double blind test.
The unfortunate thing about these is that they tend to produce results that for instance indicate that people cannot hear the difference between VFB and CFB and so on, very inconvenient for anyone with a bias towards, and / or a commercial interest in one or the other.
rcw
Last edited:
In fact, i do not need any blind test, about current feedback. The question was: did people can listen the difference. My answer was: Yes, i can. And my friends and family too.
With the clever feedback topology used by L.C.in his SSA, it is easy to move from VFB to CFB on most of the existing amps and compare. So everybody able to try-it can make his own opinion.
Of course it is the same landscape, and, if your VFB was transparent, the difference is not obvious. It is like looking to a landscape across a transparent glass. CFB is more transparent.
And it is correlated by measurements.
It is not a religion, on my side, it is just an [COLOR=#0]observation[/COLOR], based on numerous experiences during decades.
Of course, because Current feedback topology introduce a negative effect, a reduced PSSR, you need good power supply rails.
But you can let the inverting stage and use-it for symmetrical input (always a good thing). And, so, avoid this negative effect. Read the SSA thread for more about this.
Wow what flames!
Very nice explanation.
When that awkward moment arrives that you can't hear the difference in a double blind test between two amps but a day-long visioned test puts any of them above the other, which would you choose to be with? Will you toss a coin? Or use both amps, one for the right channel and the other for left? Or go for the winner of the one-day visioned challenge?
I did a double blind test with the help of my bro and I was unable to differentiate two normally apparently sonically different amplifiers. Now, which of them should I keep? Both? But then I'll be forced to listen to the one that sounds less-elegant than the other in a long listening trial. The double blind test has made its point clear - both amps are sonically same. But my brain disagree, it says "This one is always natural, and that one is not". Which option do I choose? To believe the DBT or my day-long expereince?
Of course there isn't an answer to this question. Totally personal preference. I believe my ears, I'm happy. Someone else believes the DBT, he is happy too.
And Biased are the ones who dismisses an amp for anybody other than himself/herself, depending on either DBT or visioned test or any kind of tests for that matter and saying things like "this should sound better, if not then you have tin ears" or "both should sound the same, if not then you are biased".
I would like to repeat what Christophe said - "Choose your poison", and will like to add - "Don't choose for others".
VFB is a "superior" choice for many. And that's perfectly cool.
From the first post:http://www.diyaudio.com/forums/members/elvee.html "I would like to investigate about the actual/perceived/imagined superiority of current feedback, and possibly to pinpoint what exactly makes it superior (if indeed it actually is)."
I don't understand why YOU are rude and disagreeable. Some bad experience with CFA ?
Or no experience with it at all ?
A lot of well known audio designers, Mr Nelson Pass, Mr Richard N. Marsh ( http://www.diyaudio.com/forums/anal...owtorch-preamplifier-part-ii.html#post3204326 ), Mr Mark Alexander, D Self, Mr Malcolm Hawksford... (and me are all enthusiasts about CFA.
A lot of top noted amps (near audiophiles) like QUAD, Accuphase, Marentz, Studer, Sansui etc... where current feedback based.
So what the hell with you ? Member of some king of VFB sect ?
are all enthusiasts about CFA. A lot of top noted amps (near audiophiles) like QUAD, Accuphase, Marentz, Studer, Sansui etc... where current feedback based.
So what the hell with you ? Member of some king of VFB sect ?
The amp that i had in my recording studio, chosen between all the prestigious amps at that time, for its transparency, free dynamic, instruments separation and easy listening with no fatigue. (And no ventilators !!!)
Powered 24/24 and high level listening (studio) during years with no failure.
Don't worry about the given distortion, it is professional limit to indicate the power capability. Distortion was at the limit of my measuring instruments.
Even if it is 1 000 000 times better than 1% (and I mean 10e6), it is still well short of what can be achieved with a thoughtful design.
Last edited:
Revox was the consumer products line of Studer.
http://www.studerundrevox.de/
http://www.studer.ch/news/willistuder.aspx
Last edited:
I think it is a good summary, and it falls in with my own conclusions.
let us stick to technical postings please....- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.