Random preamp noise. Noise is not distortion. I'm not comparing to anything. And I'm not talking about surface noise, or rumble, and certainly not about bandwidth linearity, whatever it is you mean by that expression.
Unbalanced vestibular folks over ~60 may, in some cases, be better off with a noisy preamp
My question was just about coherency.
If you are in the "believe only in measurement's numbers" camp, digital is just like a formula one compared to a bicycle.
Now, if you can take some pleasure listening to old vinyls, why not ? But in this case, I cannot understand how the same guys (not specially you, he ;- ) can refer to some moral reasons to impeach-us to evaluate the sonic differences between OPAs. "Op-amp rolling"
The non-existent straw men than "only believe in measurements" because they would like to see ears only no peeking data. You can find dozens of op-amp rolling threads on any number of fora and every possible sonic enhancement/degradation is attributed to simple socket pulls, then read enough of them and every op-amp comes out on top or near the top at sometime. Conclusive, right? Op-amp A oscillates in op-amp B's circuit, who cares you can't hear 60MHz anyway (true story).
Can even sound better: oscillation polishes off any residual X-over distortion
(also a true story)
When I pointed out a fairly-low-level few-MHz oscillation in an integrated class-AB NXP power amp, the head of the group looked me straight in the eye and said SO?? They had no intention of fixing it. Again, as Anna Russell used to say, I am not making this up.
^ Agreed. Rolling opamps is a shotgun method and haphazard mess, and, if the original designer did his/her homework, there's likely little to be gained and much to be lost by swapping opamps. If a pathologically bad opamp is selected for the position, there's probably warrant for a redesign, as a modern opamp will also benefit from proper decoupling, layout improvements, etc. (Or, expect aforementioned oscillations)
In the absence of carefully controlled experiments (I do not consider the plural of anecdote data, although it often provides a good hypothesis to test), why not understand the demands of the system and match the appropriate solution to the demands?
Lastly, Esperado, this link may help understand where our discussion is going WRT noise: http://www.analog.com/media/en/training-seminars/tutorials/MT-047.pdf
In the absence of carefully controlled experiments (I do not consider the plural of anecdote data, although it often provides a good hypothesis to test), why not understand the demands of the system and match the appropriate solution to the demands?
Lastly, Esperado, this link may help understand where our discussion is going WRT noise: http://www.analog.com/media/en/training-seminars/tutorials/MT-047.pdf
I cannot ensure no one of the numerous OPAs in the big mixing desk we used at this time was not oscillating at 60MHz... or 1GHz.
But I can ensure-you we never used OPAs in a situation they were supposed to be unstable.
I don't feel strange that some people can prefer X against Y. and others the opposite. Tastes and colors !
That was not our purpose. We were comparing 1X gain VS 50X, trying to find the OPA with the less difference between the 2 situations.
And little details like "ease of listening", separation between instruments, presence, micro-dynamic...
You will notice that our preferences during this contest can be justified on pure technical basis: we prefered OPAs witch had their bandwidth not affected by the gain factor (CFAs).
Peeking ? I remember this session was long and boring, if you see what i mean, and in witch direction we could had been influenced ;-)
Just listened to some Segovia through 5532's, just checking not boring, so who's opinion rules? Just for your information I don't care.
Esperado, my son just married a girl from Cape Verde, we are now countrymen in a way so peace.
Last edited:
Scott, i had read with a lot of interest so many of your application notes since so many years, that i feel you like a member of my family ;-)
My best wishes of hapiness to your son. Is he working in electronic area ?
Last edited:
We used Scott/Walt's 624/744 phono stages in Urie mixers in a night club installation near an international airport where radar sweeps would be audible through the standard Urie phono stage. We converted the Technics turntable internal wiring to fully balanced and shielded for each channel and installed the new phono preamps. With 20k watts and fully horn loaded system - bass horns were 8 feet wide x 42 high and 7 feet deep at full gain the system noise was broadband hiss with no 60 Hz or harmonics and no radar sweeps. We also added Sulzer regulators and metal film resistors to the Urie.
Esperado, you and I, and I'm sure some others here have been in this op amp evaluating business for a long time. It was part of our job. We had to make the decisions of how to make the best recording and reproducing equipment for decades. Neither you or I, or most others had any particular bias toward any IC op amp or hybrid, in fact, we would have been very happy with the 'perfect' workhorse IC that did almost everything well enough, just like I found the uA741 to be while designing servos at Ampex Research back in 1969. Tape servo's for video recorders are relatively slow and are happy being designed with IC's that are fairly rugged, self unity gain compensated, and with a slew rate of less than 1V/us.
It was only when others tried to make analog recorder record-reproduce electronics or studio boards with uA741's that we found big problems, and yet many consoles, etc were produced in the early '70's with uA741's. Did the IC manufacturers warn against this? NO! In fact, it took us years of patient research to PROVE that the uA741 was inadequate for audio. The audio IM analyzers that were the primary design equipment, in the USA at least, showed almost nothing wrong. Even Crown, who made the most popular IMA did not realize the limits of the IM test at the time. It was not intuitive then, by just looking at the test waveform.
By the mid 70's, and the introduction of the Sound Technology 1700 series test equipment, we could more easily find what was wrong with low slew rate IC's. Yet, even we, trying hard, could not justify a higher slew rate than 5V/us for an IC (preamp for example), so the NE5532 should have been enough. Then WHY did both Esperado and I, separately, find it wanting? It is because there is more that we yet not know about op amp behavior that can only be found with actual listening tests. Every time I try to put forward newer ideas like PIM, subtle xover distortion, thermal feedback, etc, I get shot down as if these problems do not exist. But they do! Even today.
It was only when others tried to make analog recorder record-reproduce electronics or studio boards with uA741's that we found big problems, and yet many consoles, etc were produced in the early '70's with uA741's. Did the IC manufacturers warn against this? NO! In fact, it took us years of patient research to PROVE that the uA741 was inadequate for audio. The audio IM analyzers that were the primary design equipment, in the USA at least, showed almost nothing wrong. Even Crown, who made the most popular IMA did not realize the limits of the IM test at the time. It was not intuitive then, by just looking at the test waveform.
By the mid 70's, and the introduction of the Sound Technology 1700 series test equipment, we could more easily find what was wrong with low slew rate IC's. Yet, even we, trying hard, could not justify a higher slew rate than 5V/us for an IC (preamp for example), so the NE5532 should have been enough. Then WHY did both Esperado and I, separately, find it wanting? It is because there is more that we yet not know about op amp behavior that can only be found with actual listening tests. Every time I try to put forward newer ideas like PIM, subtle xover distortion, thermal feedback, etc, I get shot down as if these problems do not exist. But they do! Even today.
Don't want to continue here, I opened a new thread. I will jump in there, time permitting, if there's anything worth debating (given the general tone, I am rather skeptical at this point).
Back when the 741 came out we have to get up at 4 o'clock in the morning, 3 hours before we went to bed, have a handful of cold gravel for breakfast etc. Yawn.
Not got any stories relevant to the 21st century? I'm sure you must have won at least one award with a design with some op amps in?
Not got any stories relevant to the 21st century? I'm sure you must have won at least one award with a design with some op amps in?
Benchmark AHB2 Power Amplifier - Benchmark Media Systems, Inc.
This should divide opinions. Either the best measuring commercial amplifier on the planet or cold, sterile and souless. Or for me too expensive
This should divide opinions. Either the best measuring commercial amplifier on the planet or cold, sterile and souless. Or for me too expensive
- Status
- Not open for further replies.