I thought so. BTW, I for forgot, congrats getting them up and running.
What if a measurement could be found that shows that the 'high distortion valve amplifier' produces less distortion and many SS (not necessarily all of them) amplifiers. That would certainly turn the tables, would it not? But it would be a very different method of measuring distortion.
And no, the DPA-300B is anything but a high distortion valve amplifier. Without any feedback around the output transformer (something we found to be wrong... but that is another story) it produces much less distortion than what you would expect.
I'm sure the amp you mention is better than the one measured here but it's the same tubes at least. 🙂
https://www.audiosciencereview.com/...tube-amplifier-review-and-measurements.44020/
https://www.audiosciencereview.com/...tube-amplifier-review-and-measurements.44020/
The Vacuum State DPA-300B is not like other 300B amplifiers. If is fully differential circuit 'long tail' and that includes the 300B x 2 output stage. It only works in Class A, and even when it clips, it is still class A.
Also, when people say the word "distortion" there is an unspoken word, because when they say the word they really mean voltage distortion. Since a speaker only reacts to current, then what about current distortion?
But I think that all would like to resolve the question.
Why is it that amplifiers that don't measure all that great can sound rather good and there are amplifiers that measure very well don't sound so good? That does not mean that you can't get an amplifiers that measures well and also sounds good. That too is possible. But are we measuring the wrong thing? In the above measurements, they are all voltage measurements. The current of the amplifier is largely considered in terms of headroom: Do we have enough current available to drive the speakers? The idea that the amplifier can distort there on the current side, that is largely dismissed. But that only means that we can look at the voltage of the amplifier, and there has proved to be no answers there.
Are we "audiophools" when we hear important differences in amplifiers?
Earl Geddes, with whom I have exchanged PMs, is one of the doubters and yet he strangely leaves the door ajar when he said:
"Maybe the problem is with the way we measure distortion in both amplifiers and speakers and if using a different
technique, we may solve this conundrum once and for all - and establish that most audiophiles who say that they can
hear differences are not audiophools."
[Underlining added - this was not from a PM, but something he said in a YouTube video]
Yeah, a conundrum for sure. It has been going on for decades.
Last edited:
Modified Waveguide arrived today and now the SB26CDC tweeter is a good fit.
The Waveguide before on the left, and the modified one on the right. You can clearly see how much narrower it looks. Now it gets right inside the throat and when we measure it on Sunday, weather etc permitting, it should measure much better and not have that null/dip at 9KHz.
The Waveguide before on the left, and the modified one on the right. You can clearly see how much narrower it looks. Now it gets right inside the throat and when we measure it on Sunday, weather etc permitting, it should measure much better and not have that null/dip at 9KHz.
Earl Geddes said this in regard to what you just said in this post? I wouldn't have expected that. Could you please provide context for the quote to help me understand?
@gedlee
It was a YouTube video (as I said) and when I came across it, I made an effort to type it into my PC. Maybe I should try to find that video again and I wish I had jotted it down and how many minutes in - but he did say it.
A good example here of an amplifier that sounds good but does not have very good measurements.Or perhaps sounds good because it does not have good measurements?
https://www.google.com/url?sa=t&rct...usg=AOvVaw0pOvDt3SonDsg27_N0vn7G&opi=89978449
https://www.google.com/url?sa=t&rct...usg=AOvVaw0pOvDt3SonDsg27_N0vn7G&opi=89978449
Some are saying that if you hear differences in two properly functioning amplifiers, it is a delusion.
If you don't agree and you actually hear definite differences, then you have no answer to those who are saying you are delusional. How can you prove that you are not delusional? Getting confirmation from somebody else, that they are hearing differences as well, then you are up against the argument that you are both delusional. So it does not matter how many agree with you, you are still delusional and they will argue that being delusional is very common. But of course, they are not delusional, but you and I are.
What would it take?
Find different measurements where the results correlate with the sound quality heard.
That is the only solution to this conundrum (a puzzle or finding the answer to a difficult question.)
"Maybe the problem is with the way we measure distortion in both amplifiers and speakers and if using a different
technique, we may solve this conundrum once and for all - and establish that most audiophiles who say that they can
hear differences are not audiophools." Earl Geddes
We are not "phools" of any kind.
But how do you prove it?
If you don't agree and you actually hear definite differences, then you have no answer to those who are saying you are delusional. How can you prove that you are not delusional? Getting confirmation from somebody else, that they are hearing differences as well, then you are up against the argument that you are both delusional. So it does not matter how many agree with you, you are still delusional and they will argue that being delusional is very common. But of course, they are not delusional, but you and I are.
What would it take?
Find different measurements where the results correlate with the sound quality heard.
That is the only solution to this conundrum (a puzzle or finding the answer to a difficult question.)
"Maybe the problem is with the way we measure distortion in both amplifiers and speakers and if using a different
technique, we may solve this conundrum once and for all - and establish that most audiophiles who say that they can
hear differences are not audiophools." Earl Geddes
We are not "phools" of any kind.
But how do you prove it?
Why does it matter?
Statistics and proof do not count when the sample size := 1.
When you have a sample size of 1 sceptic you can not prove anything to anyone.
A short story:
My wife just bought a new car a month ago with a upgraded in-cabin sound system by Harman Kardon. On a 300 mile road trip yesterday my wife says to me, "these speakers sound nice". That is proof enough for me and Harman Kardon.
These days in-cabin sound systems are designed by crash test dummies with GRAS microphones for ears.
Earl tells us that trained human listeners are unreliable, at least at Ford. Golden ear humans need not apply.
Thanks DT
I hate the golden ear phrase. We are not delusional.
No golden ears are required!!!
Here is a morality tale:
A pair Dyna Mk3 monoblocks were dropped off to me, they just needed minor work. They had the Vacuum State PP-1 circuit built in to them, so they were not exactly standard. The KT88 output tubes were wired in Triode by tying the Screens to the Anode via 220R resistors.
Without telling my friend, I replaced the 4 x 220R resistor with 30V 5W Zeners that were bypassed by 220uF capacitors, higher grade Panasonics electrolytics.
He picked up the amplifier and took near three hours to drive home, where he promptly put them into his system. He then called me back:
"Did you do anything to the amplifier? Did you increase the power?
"No Walter, it is still around 35 Watt and they are still triode."
"Then what did you do, it sounds so much more powerful?"
"But is it better?"
"Better. Even that Beatles CD now has bass. I never thought I would hear Beatles and bass." [I had given him a copy of the "Love" album]
"And better all right?"
"For sure!"
No golden ears were required. The difference was noted immediately and the 'subject' had been kept in the dark. He reacted appropriately and without any prompting. This is about as blind as you can get. There are more than just the 'approved' way of testing things. I knew his reaction before he was even aware that he was supposed to react to anything.
Another example, I am a tube guy. But I recently bought a March Audio P262 amplifier, which uses Purifi Eigentakt Class D modules. It did not take me long to realise that this was the best solid-state amplifier I have heard. It is hugely better than the Hypex Class D amplifiers that I have owned previously by a huge margin, and I've sold sold them. One of them were the NCore 400 monoblocks. It could not reproduce the complex overtones of my son's real Spanish made concert level guitar (did I mention that my oldest son is a classically trained guitarist?). The inability of the Ncores to get that right was instantly obvious. No golden ears required!
I have just bought another P262 for a friend who uses my tube phono/line preamp. He also has Elsinore Mk5 and the P262 will replace an aging tube power amp that has become to unreliable. He is getting it on my recommendation - and the P262 gets those complex overtones right where the Ncore did not. I would not have recommended it otherwise.
We just don't need to have golden ears and we are not delusional - even if you call us that, we are not. The above case with Walter is a slam dunk - he heard the difference easily, even if he were kept totally in the dark, and the difference was not small, it took him greatly by surprise.
So what is really needed to settle this?
A NEW KIND OF MEASUREMENT!
Sorry for shouting. But even Earl left the door open for that possibility.
I have a suggestion:
1. We test speakers when connected to amplifiers. Fine.
2. Why not test amplifier with speakers connected.
3. And then why not measure distortion on both side of the amplifier, test both the voltage and current?
If we connect a resistor to an amplifier as a load, what happens if we measure both voltage distortion and current distortion.
They are the same!
Connect the amplifier and now make the load a real load an actual speaker. Now what happens if we measure both voltage distortion and current distortion?
They are not the same!
Are we doing something wrong? Trust me, there is a lot more distortion on the current side than the voltage side.
If they are not the same, then which are we listening to?
The voltage?
Or are we listening to the much higher current distortion?
In dynamic drivers, the answer to that question is already know.
Yet there is a complete reluctance to even considering current distortion. Why? It is actually quite logical - if the current gets distorted, that is what we end up listening to!
No golden ears are required!!!
Here is a morality tale:
A pair Dyna Mk3 monoblocks were dropped off to me, they just needed minor work. They had the Vacuum State PP-1 circuit built in to them, so they were not exactly standard. The KT88 output tubes were wired in Triode by tying the Screens to the Anode via 220R resistors.
Without telling my friend, I replaced the 4 x 220R resistor with 30V 5W Zeners that were bypassed by 220uF capacitors, higher grade Panasonics electrolytics.
He picked up the amplifier and took near three hours to drive home, where he promptly put them into his system. He then called me back:
"Did you do anything to the amplifier? Did you increase the power?
"No Walter, it is still around 35 Watt and they are still triode."
"Then what did you do, it sounds so much more powerful?"
"But is it better?"
"Better. Even that Beatles CD now has bass. I never thought I would hear Beatles and bass." [I had given him a copy of the "Love" album]
"And better all right?"
"For sure!"
No golden ears were required. The difference was noted immediately and the 'subject' had been kept in the dark. He reacted appropriately and without any prompting. This is about as blind as you can get. There are more than just the 'approved' way of testing things. I knew his reaction before he was even aware that he was supposed to react to anything.
Another example, I am a tube guy. But I recently bought a March Audio P262 amplifier, which uses Purifi Eigentakt Class D modules. It did not take me long to realise that this was the best solid-state amplifier I have heard. It is hugely better than the Hypex Class D amplifiers that I have owned previously by a huge margin, and I've sold sold them. One of them were the NCore 400 monoblocks. It could not reproduce the complex overtones of my son's real Spanish made concert level guitar (did I mention that my oldest son is a classically trained guitarist?). The inability of the Ncores to get that right was instantly obvious. No golden ears required!
I have just bought another P262 for a friend who uses my tube phono/line preamp. He also has Elsinore Mk5 and the P262 will replace an aging tube power amp that has become to unreliable. He is getting it on my recommendation - and the P262 gets those complex overtones right where the Ncore did not. I would not have recommended it otherwise.
We just don't need to have golden ears and we are not delusional - even if you call us that, we are not. The above case with Walter is a slam dunk - he heard the difference easily, even if he were kept totally in the dark, and the difference was not small, it took him greatly by surprise.
So what is really needed to settle this?
A NEW KIND OF MEASUREMENT!
Sorry for shouting. But even Earl left the door open for that possibility.
I have a suggestion:
1. We test speakers when connected to amplifiers. Fine.
2. Why not test amplifier with speakers connected.
3. And then why not measure distortion on both side of the amplifier, test both the voltage and current?
If we connect a resistor to an amplifier as a load, what happens if we measure both voltage distortion and current distortion.
They are the same!
Connect the amplifier and now make the load a real load an actual speaker. Now what happens if we measure both voltage distortion and current distortion?
They are not the same!
Are we doing something wrong? Trust me, there is a lot more distortion on the current side than the voltage side.
If they are not the same, then which are we listening to?
The voltage?
Or are we listening to the much higher current distortion?
In dynamic drivers, the answer to that question is already know.
Yet there is a complete reluctance to even considering current distortion. Why? It is actually quite logical - if the current gets distorted, that is what we end up listening to!
The tea from China is like the speaker measurements.
It takes oolong to get here.
Could you explain the 84dB bit? That is where you lost me.
FWIW, I try to find meaning in measurements, even if others have given up. If something is heard and cannot be explained by the measurements we are using now, it means that we haven't found the right measurement yet. It has nothing to do with Golden Ears.
Even Earle Geddes has left that door ajar. But some are anxious that it be shut?
But please, no silly tall tales about non-existing Golden Ears.
Tin Ears maybe? 😉
Sorry, could help that last one.
FWIW, I try to find meaning in measurements, even if others have given up. If something is heard and cannot be explained by the measurements we are using now, it means that we haven't found the right measurement yet. It has nothing to do with Golden Ears.
Even Earle Geddes has left that door ajar. But some are anxious that it be shut?
But please, no silly tall tales about non-existing Golden Ears.
Tin Ears maybe? 😉
Sorry, could help that last one.
Could you explain the 84dB bit? That is where you lost me.
FWIW, I try to find meaning in measurements, even if others have given up. If something is heard and cannot be explained by the measurements we are using now, it means that we haven't found the right measurement yet. It has nothing to do with Golden Ears.
Even Earle Geddes has left that door ajar. But some are anxious that it be shut?
But please, no silly tall tales about non-existing Golden Ears.
Tin Ears maybe? 😉
Sorry, could help that last one.
Back when I started with a 0.55 cubic foot enclosure with, if I remember correctly, a single 6-1/2 inch peerless mid-bass driver installed, the same as used in the original Elsinore speaker. At low to moderate SPL output that speaker sounded pretty good to my normal human ears. However when the SPL increase to above ~84dB's the sound output was increasingly irritating to my normal ears. You know, strained and not fun to listen to. That is proof enough for me, no measurements required.
I do have the fun microphones and analyzers on my bench. I did go back and measure that NAD 216 THX amplifier and peerless speaker pair. With increasing SPL output there was increasing levels of THD, two-tone IMD side-bands and current distortion measured across a 0.1R current sensing resistor.
Sorry, @gedlee there is, in this case, a strong positive correlation between THD and perceived distortion. Correlation does not necessarily mean a causal relationship.
Is THD the worst offender? Likely not.
The most likely worst offender is Two-Tone IMD side bands caused by Force Factor (BL) Modulation or Amplitude Modulation.
For the long winded explanation see the AES paper by @lrisbo .
Thanks DT
Have any thoughts about how to isolate Sd distortion in measurements?
Last edited:
I tried one of those Purifi Class D and did not like it at all.Lots of clarity and detail and even tonally quite decent but it it was musically and rhythmically disjointed.Imaging was not that good either -especially centre fill image and almost phasey sounding.Just not very engaging.I can see why people might be impressed at first but reckon they will lose interest in listening to their system after a while-and probably wonder why.
^^^
Yes I’ve had the same experience with connecting my Purifi monoblocks to some Quad electrostats. The low crosstalk panned the image far left and right. A traditional amp with 50dB crosstalk and a bit more distortion sounded more “normal” and brought back the “richness and depth”
That’s why Purifi are clever enough to sell their OEM modules with only 13dB of gain.
Allows manufacturers to differentiate from each other and add their pre-gain or pre-amp in whatever variety of silicon or tube they want.
Not everyone likes to drink plain water.
And until a large organisation body sets a standard for studio monitors for near field and midfield and room acoustics/treatment/correction, we may be going around the circles of confusion in music reproduction- so whatever brings you fun and joy and tears in writing and playing and listening to music, no one can really argue with individual listener preference
Yes I’ve had the same experience with connecting my Purifi monoblocks to some Quad electrostats. The low crosstalk panned the image far left and right. A traditional amp with 50dB crosstalk and a bit more distortion sounded more “normal” and brought back the “richness and depth”
That’s why Purifi are clever enough to sell their OEM modules with only 13dB of gain.
Allows manufacturers to differentiate from each other and add their pre-gain or pre-amp in whatever variety of silicon or tube they want.
Not everyone likes to drink plain water.
And until a large organisation body sets a standard for studio monitors for near field and midfield and room acoustics/treatment/correction, we may be going around the circles of confusion in music reproduction- so whatever brings you fun and joy and tears in writing and playing and listening to music, no one can really argue with individual listener preference
It happens here every day.
The individual with a strong "Personal preference", transitions to the guy with golden ears and the argument is on.
"Personal preference" is not really so innocent after all.
To go way back to the question in 5620, Tube amp distortion is pretty. Speaker distortion is generally not so pretty. I like listening to amplifiers that have different charactoristics to their sound. I can appreciate it when a speaker doesn't get in the way of that. Some amplifiers grab speakers by their cajones and fling them around like toys. Some amplifiers need to be babied. Each has its merits. 🙂