I keep trying to ponder this measurement stuff, and this month
I keep trying to ponder this measurement stuff, and this month
So, if the Eico doesn't have a flat frequency resonse, which the graph looks like it doesn't (Page 149, figure 1), then it may have varying phase?
I find that review this month's most interesting because it resides in the realm of where we try to reconcile sound with measurement.
The reason I asked the square wave question was actually based on that review because that amp deviates so much from most other square wave graphs.
I also note, in figure 7, that the amp seems to have its lowest distortion components (below 0.3% or so) in the 100Hz to 6 or so KHz range, but has higher distortion above that - maybe there is such a thing as euphonic distortion, maybe like analog dither?
I really appreciate this feedback. I think we mislabel objectivism in Hi-Fi. Objectivism seems more like the pursuit of answers for why things sound the way they do, not an implication that things don't sound different.
Cheers, and thanks again.
If you look at the specs for a $200 stereo amplifier from Sony, Kenwood, etc, you will find that it has vanishingly low distortion over the entire audio frequency range. Does it sound good? Nope.
This specs are obtained by measuring one sine-wave frequency at a time, with a resitive amplifier load, which tells you very little about what the circuit will do when it has hundreds of different frequencies to amplify at once (music) and a load which varies in impedance and reactance as frequency changes (a speaker). This is an infinitely complex situation; impossible to simulate precisely in a lab.
I am reminded of the lesson I learned in about 1980, when I bought a Yamaha integrated amp with 125W per channel and those perfect specs. Neil Sinclair of Absolute Audio, upon learning that I had spent $800 on this piece of poorly engineered JUNK, said that he had an integrated amp for $200 that would blow it away. he suggested that I take the NAD 3020 home with me and do a direct comparison.
After about one hour of comparison, I took the Yamaha back and got a refund. The specs on the NAD were not all that good by comparison; the sound was excellent. The Yamaha sounded terrible (and the 125WPC Yamaha was no match for the 25W of the NAD!!). I used that NAD 3020 with great satisfaction for years.
So much for spec sheets.
The intermodulation distortion and phase distortion are next to impossible to measure in a meaningful way in an actual system setting; that is why the ears are the point where the rubber meets the road!
A Ferrari and a Chevrolet may measure exactly the same on a dynamometer, but there is not much resemblance when they are driven on real roads.
By the way; I had that same EICO HF-81 amplifier back in 1960 when I was in high school; it really wasn't a very good-sounding amplifier back then, without out the $2000 worth of state-of-the art 2006 components.
I replaced it with another EICO amp that had MUCH better (huge) ultra-linear output transformers (about 10 pounds of iron each...lol) and 6L6 tubes and 30W; it had far, far better sound than the HF-81.
If you spend $2000 to rebuild it, as was done for the HF-81 article, you no longer have anything that even faintly resembles the original amplifier or its performance; ridiculous! Just replacing the original capacitors with today's state-of-the-art caps produces such a sea change from anything the original amp could do that it is ridiculous to talk of the amp as having any meaningful relationship to the original.
This amp was outclassed in its day by any number of inexpensive amps; I was there, and my friends and I had very little money and good ears,and we LISTENED real hard before we spent it. This was a mediocre performer in its day; nothing more! We listened to it, and rejected it as second-rate.
It is as if you took a 1955 Chevy and put on modern radial tires, modern disc brakes, a modern 400 horsepower dual-overhead cam engine with a 5-speed transmision, rebuilt the suspension, and then concluded "hey, this 1955 Chevy was really a great car...see how it performs with just a little tuneup!
Idiotic; there is hardly anything left of that 2nd-rate EICO amp but the front panel; you kept the box and the schematic diagram, and built a whole new 2006 amp. As the measurements prove, this thing had little power, no dynamic range above the noise floor, and a cheap little output transformer; and this is AFTER putting $2000 into it. What a joke!
I could give you a list of at least 10 amplifers from 1960 that were better designs to start with and far more worthy of such an article and rebuild; why someone picked this mediocre mutt is beyond me (and shows a complete lack of knowledge about what was truly worthy and well-designed in 1960).
Maybe I'm confused? Your post starts with several paragraphs explaining how the specs don't matter...trust your ears. Yet you seem to sum up by saying they do matter, "what a joke" yet you have never heard the piece of equipment.
Have I misunderstood your post?
I'll pose a similiar comparison, regarding the McCormack UDP-1 Universal player, which measures poorly, won't resolve Hi-Rez data as one would expect from a player dubbed SACD/DVD-A capable... yet still retains a Class A rating in Stereophiles recommended components. Apparently, those who have heard it trust their ears.
Yes, you misunderstood.
I started by explaining why the DISTORTION specs from most manufacturers do not indicate the actual equipment DISTORTION performance under real-world conditions.
That was very specific.
How you translate that into a rejection of "the specs" (ALL specs or test results is what you apparently infer...) is beyond me.
"The Joke" that I refer to is the spending of $2000 on a mediocre $60 amplifier (to the point of replacing virtually every replaceable component with 2006 state-of-the-art components), and then inferring that it was a great amplifier to start with. It wasn't; I owned one years ago, and it was mediocre in 1960 compared to other products. It is a joke to pick this amplifer that has mediocre output transformers, a marginal power supply, a lousy signal-to-noise ratio (which IS easy to measure accurately), and then compare it favorably with the current state-of-the-art.
I KNOW the above things due to my personal experience with the thing when I OWNED one 40 years ago. It DOES have a marginal power supply, and it DOES have undersized cheap output transfromers, and in 1960 it sounded distinctly second-rate compared to other products my friends and I auditioned, built, purchased, and modified.
Measured parameters are not always the bottom line, but in this case the high noise levels and poor dynamic range that were measured tell me that this component cannot resolve low-level information in any satisfactory manner; too damn much noise is too damn much noise!
As for the McCormack UDP-1, I have compared this unit in my system to my 6-year-old SONY SCD-777ES, and it is markedly inferior in dynamics and resolution to the Sony. The SONY was rated Class A+ when it was in production, and I do not know whether the McCormack should or should not be rated Class A. I was not impressed, however. It would not get a Class A rating from me. My SONY is more dynamic and presents voices and instruments in a much more realistic, lifelike manner.
That IS what my ears tell me; just trusting my ears.