The A-rating is simply what the reviewers liked best among the products they've listened to, probably with some consultation with other reviewers and people whose ears they trust who are outside of the profession. It is certainly not the last word, simply because there is not enough time to review everything.
I have a very good friend who has been an audiophile for a very long time (since he bought the then newly introduced Ortophon SPU at the age of seventeen). He has put together a system with great effort over a period of several years that I am here to tell you is the best I have ever heard. The retail value of this system is in excess of $300,000, although he aquired it mostly through horse-trading (there is a lot of that in the business). My point is that NOTHING in this system is or has been listed as A-rated. But, if you told any reviewer what the make-up of that system was he would tell you that it probably is an outstanding system.
John
MikeBettinger said:
Hi Mike,
I understand what you mean. I’ve seen some reviews like that, but I really don’t think that is the case with the Halcro DM-58 review and the other numerous very positive Halcro reviews in Stereophile. These guys were going way beyond polite in their praise for the Halcro. All of those listed below uniformly and strongly praised the Halcro in various Stereophile reviews. Also, take a look at the review of the DM58’s successor, the DM-88, in the August 06 issue. No one was holding a gun to their head in declaring the DM-88 their Amplifier of the Year.
Paul Bolin
John Atkinson
Brian Damkroger
Sam Tellig
Art Dudley
It also brings to mind some of the performance review directions we used to have to follow as supervisors: even for your very top guy in your group, you HAD to find at least one “Area for Improvement” that you had to say about that person.
Cheers,
Bob
Good design tends to distinguish recording quality, poor design tends to mask it and cover in Global Sameness 😉
Do you know for a fact whether the designer at Halcro "voiced" his amps through critical listening or he simply designed them at the bench? The crux of this debate is (again) subjectivity versus objectivity. The only amp that I know for sure that was designed strictly with test equipment only was the Quad II, and you talk about mediocrity (reliability, noise, AND sound) foisted on the unwary public through technical hype.
John
In fifty years or so, when autonomous robots are commonplace, I'm certain that endless rants about Halcro's .00000000001% distortion will be meaningful...to the robots. In the meantime, I've not heard that tests have revealed that humans are able to perceive the difference between .001% and .0009% distortion. The specs-uber-alles crowd's insistence that it is meaningful is tantamount to declaring that people can taste the difference between 13.5% alcohol in a wine and 13.49%--and that it is the definitive descriptor of wine quality. It simply ain't so and focusing on the alcohol content to the exclusion of the thousand other chemical compounds in wine is laughable, or would be if it weren't so sadly deluded.
I find the assumption that the Halcro amps' favorable reviews are due to their distortion specifications rather amusing. Not once do the people who champion the Halcros consider the possibility that Halcro might have done something else that influences their sound quality. Perish the thought. The Halcro circuits have low distortion, ergo it's the low distortion that makes them sound so good to the reviewers at Stereophile.
Pardon me, but that line of reasoning is pathetically weak.
And if the reviewer's writing style doesn't suit you, find another reviewer. To me that merits a big "Duh!" There are scads of reviewers out there. Honestly, I don't care for a lot of them either, but you won't find me getting unduly exercised over their descriptive styles.
But then again, some people are just born literalists, forever doomed to live in a two-dimensional world, ranting and spraying spittle at those who declare the existence of a three-dimensional world forever beyond their ken.
Next stop...Flatland.
Grey
I find the assumption that the Halcro amps' favorable reviews are due to their distortion specifications rather amusing. Not once do the people who champion the Halcros consider the possibility that Halcro might have done something else that influences their sound quality. Perish the thought. The Halcro circuits have low distortion, ergo it's the low distortion that makes them sound so good to the reviewers at Stereophile.
Pardon me, but that line of reasoning is pathetically weak.
And if the reviewer's writing style doesn't suit you, find another reviewer. To me that merits a big "Duh!" There are scads of reviewers out there. Honestly, I don't care for a lot of them either, but you won't find me getting unduly exercised over their descriptive styles.
But then again, some people are just born literalists, forever doomed to live in a two-dimensional world, ranting and spraying spittle at those who declare the existence of a three-dimensional world forever beyond their ken.
Next stop...Flatland.
Grey
Seems to me some audio folk need to evolve to the point where they can accept that different technically solid designs are equally valid even though they produce different subjective results.
G.Kleinschmidt said:
Well, either your memory is faulty or your ability of reason is.
The reason I was pointing out the low distortion of my latest amplifier design is to make the point that it is not necessary to use feedback to attain good performance on the test bench. (That point is completely consistent with the letter I wrote to Stereophile.)
The very first iteration of the prototype amplifier measured just as good as the production version. However, it had quite a number of audible sonic deficiencies. I spent four months fine-tuning the circuit solely on the basis of listening tests. At the end of that time the measurements hadn't changed, but the sonic performance was leagues better.
I think a better way to summarize my experience with regards to measurements would be something like this:
"If all else is equal, a more linear circuit may sound better than a less linear circuit. But achieving linearity by applying band-aids or after-the-fact corrections to a circuit does not necessarily lead to better sound, and in most cases actually degrades the sound."
Charles Hansen said:
I agree on this. In 2003, when I developed my preamp for Luxtone (W.A.), I spent several months by listening tests and tuning (mostly signal filtration at different stages). The THD before and then was the same 0.0007%. The sound difference absolutely clear.
Regards,
Pavel
PMA said:
Pavel, thank you for sharing. I know that you and I don't always agree on everything, but this was the point of my letter to Stereophile. While there is nothing inherently wrong with good measurements, they are far from a guarantee of good sound.
And to me that is what makes audio such an interesting field. To use your preamp as an example, what is it that can make such a difference in the sound? I think we all agree that there must be some difference in the signal. But what is that difference? And can that difference be measured? (Clearly not by traditional THD+N tests!)
Jlsem has it right. We know that we have to 'impress' the reviewers with something. It might be 'the hard way' by making the component sound as good or better than most other similar components. Or it might be remarkable marketing campaign that awes the reviewers that your component is really better for technical reasons. Over the years, buying a reviewer and expensive dinner seemed to be part of the deal. It depends.
I try to win a class 'A' rating 'the hard way'. I like to make my stuff sound better than much of its competition. That takes 'tweaking' and design skill. One is NOT enough. I found that out over the decades.
What many of you don't understand is that Charles, Nelson, and I started just like many of you.
I personally welcomed negative feedback in the 1960's. I made current drive amplifiers as well as voltage drive amplifiers, and everything in between.
I loved the concept of the op amp and the IC. In 1970, I helped design an all IC mixing board for the Grateful Dead. That is why I was first hired. I had been professionally evaluating linear IC's since 1966, both at Friden and then at Ampex. I used 741's by the dozen and thankful for them, in my advanced servo motor drive designs.
For the Grateful Dead, I selected the then 'exotic' dielectrically isolated +/- 24V capable, super slew rate (+5/-2.5V/us) IC from Harris Assoc. then called Radiation Inc. These things cost $5 ea in 100's and we only made $5/hour as engineers at the time. Big bucks! Still, the Grateful Dead after long evaluation went back to TUBES!
I scratched my head, and finally went on to other designs.
Finally, a few years later, I was again called up to build analog gain blocks for the mixing console. This time I did it discrete, linear, and fast (100V/us). It cost $50 a hybrid style module, but well worth it! This time the GD loved it, and went with it.
Mark Levinson built these modules as a subcontractor for us. He switched over to the hybrid modules and the JC-2 was designed with it. We were incredibly successful with the JC-2. Yet the LMP-2, and even more expensive preamp, never got very far. Think about it.
I try to win a class 'A' rating 'the hard way'. I like to make my stuff sound better than much of its competition. That takes 'tweaking' and design skill. One is NOT enough. I found that out over the decades.
What many of you don't understand is that Charles, Nelson, and I started just like many of you.
I personally welcomed negative feedback in the 1960's. I made current drive amplifiers as well as voltage drive amplifiers, and everything in between.
I loved the concept of the op amp and the IC. In 1970, I helped design an all IC mixing board for the Grateful Dead. That is why I was first hired. I had been professionally evaluating linear IC's since 1966, both at Friden and then at Ampex. I used 741's by the dozen and thankful for them, in my advanced servo motor drive designs.
For the Grateful Dead, I selected the then 'exotic' dielectrically isolated +/- 24V capable, super slew rate (+5/-2.5V/us) IC from Harris Assoc. then called Radiation Inc. These things cost $5 ea in 100's and we only made $5/hour as engineers at the time. Big bucks! Still, the Grateful Dead after long evaluation went back to TUBES!
I scratched my head, and finally went on to other designs.
Finally, a few years later, I was again called up to build analog gain blocks for the mixing console. This time I did it discrete, linear, and fast (100V/us). It cost $50 a hybrid style module, but well worth it! This time the GD loved it, and went with it.
Mark Levinson built these modules as a subcontractor for us. He switched over to the hybrid modules and the JC-2 was designed with it. We were incredibly successful with the JC-2. Yet the LMP-2, and even more expensive preamp, never got very far. Think about it.
Charles Hansen said:
Charles,
yes, I agree that for this reason audio design (and listening evaluation) is that exciting. I would be glad to find a measuring method explaining the differences we hear! 😀 😀 . I tend to belive in a kind of difference measurement (out - in), though it has pitfall in mixing of linear and nonlinear errors together. The problem is that spectral analysis is useless for transient signals.
Regarding "To use your preamp as an example, what is it that can make such a difference in the sound?", it was the signal filtration at every stage that changed te sound (RF, who knows). It was not invasive filtering, BW (-3dB) remained at some 130kHz. But it had to be done in every stage of the signal chain of the preamp.
Regards,
Pavel
Maybe THD as read on a meter didn't change, but something did, and there's no question that it could be measured. I don't understand why people keep coming back to gross THD measurements, and then declaring the uselessness of all measurements. If the signal changed, it can be measured. That's the first step to understanding the sonic changes. What other path will ever lead to some coherent set of design rules and requirements to produce decent equipment? The idea that signals can be altered in ways too subtle to measure is rubbish.
Conrad,
this is a misunderstanding. I do not measure a THD number, but spectra. I believe that spectra are saying something, and it corresponds to class A listening results. One must be able to read in them.
For me, low level (or better unmeasurable) of high order harmonics is a must. I also like to have low 2nd and 3rd, but can tolerate something like -90dB for lower power levels and more for higher levels. Then, low power must be spectrally pure and with no high order harmonics, even not -120dB (here is the pitfall of class B/AB). In case this is fulfilled, the amp is stable and fast, and has low output impedance, it MAY sound very well.
For me, low THD (with regards of spectral content) is a necessary, not satisfactory condition of a good sounding amp.
Regards,
Pavel
this is a misunderstanding. I do not measure a THD number, but spectra. I believe that spectra are saying something, and it corresponds to class A listening results. One must be able to read in them.
For me, low level (or better unmeasurable) of high order harmonics is a must. I also like to have low 2nd and 3rd, but can tolerate something like -90dB for lower power levels and more for higher levels. Then, low power must be spectrally pure and with no high order harmonics, even not -120dB (here is the pitfall of class B/AB). In case this is fulfilled, the amp is stable and fast, and has low output impedance, it MAY sound very well.
For me, low THD (with regards of spectral content) is a necessary, not satisfactory condition of a good sounding amp.
Regards,
Pavel
G.Kleinschmidt said:
You're not that obtuse, though you certainly give the impression of thinking so of others. The point was top experts in their fields, at least in my experience, rarely speak in the language of the near-religiously self-assured proclamations displayed here. Cordell, Curl and Hansen may have violent disagreements but they have the mileage and experience to admit what they don't know and leave the opinions of their peers open to possibility.
An side suggestion, the 'all measurements make no difference' strawman should be set aside. I don't think a single competent person here holds the position +-10db freq response coupled with 25% IMD is inaudible, I take the meaning as 'after a certain neccessary point measurements alone are no longer sufficient.'
x-pro said:
Very well stated, Alex.
Both subjective testing and objective testing have their problems.
Darn, the problem with subjective testing is that it is so subjective!
Another problem with subjective testing is there is usually virtually no testing discipline, but, understandably, when one introduces some discipline into listening evaluations there are concerns about the Heizenberg uncertainty principle effect, or the like (yes, I have probably spelled it wrong).
So, undisciplined lestening evaluations, disciplined listening evaluations, and bench testing all have very serious limitations. I think that we all just need to avoid throwing the baby out with the bathwater on any of these.
Cheers,
Bob
Bob Cordell said:
Also very well stated, Bob. That is also one of the reasons why audio is so much fun. I'm doing it for 33 years and I still can not stop completely 🙂 , thought I am not doing it for living anymore...
Cheers
Alex
P.S. - sorry for misprint
jlsem said:
Hi John,
No, I really don't know how Candy went about his design. I do understand that he took details and performance very seriously, and would not be surprized if he spent a lot of time listening to his prototypes during their development.
While this debate has inevitably wandered into the subjective vs objective battlefield, I think the major origin of this most recent portion of the dabate was Charles' assertion that feedback was bad based on a quote from a Stereophile Review. When it surfaced that the review was of the Halcro DM-58, it became evident that the quote was plainly out of context. When the review was looked at in its entirety it appeared that it was actually an extremely positive review of the Halcro. Things then lauched off into people questioning the review itself (or the interpretation of it).
Cheers,
Bob
GRollins said:
Actually, you make a very good point. However, I have never said that the extremely low distortion of the Halcros is solely responsible for their good reviews, and neither has John Atkinson.
I honestly don't think anyone can hear 0.0001% THD. However, I do suspect that the process of properly designing for the highest linearity in order to achieve very low distortion can lead to a better design, perhaps in performance metrics that we don't actually measure. Put another way, if spectrally-measured THD and IM are extremely low, there is likely less "wiggle room" for other distortions that are not as well understood to get through.
I'm sure that Halcro had to do a lot of other things right to achieve the sonic performance that they apparently did.
It is probably fair to say that extremely low distortion is neither necessary nor sufficient to result in an extremely sonically good product, but perhaps the process one goes through in achieving that may increase the odds of success.
Cheers,
Bob
I guess that is self-evident. If there is an audible difference, that means there is a difference in air vibrations, which means there is a difference in speaker vibrations, which means there is a difference in amp output signal, which means it is - in principle - measureable. It's that " - in principle - " that's eluding us, so far.
Jan Didden
- Status
- Not open for further replies.