Another listening on distortion
Orange has the highest distortion. It is I'll defined, masked, rounded, just like how tube amps are favored even tho they are not accurate. Damn, they must know something when they say 741 will win
Peach has the least distortion. It should be the original
Hahaha this time I'm sure with what I hear. I focus myself on distortion. And you know what? One stands out for the least distortion, one stands out for the worst distortion. There could be some imperfection with digitizing but I'm sure with my ears.
Orange has the highest distortion. It is I'll defined, masked, rounded, just like how tube amps are favored even tho they are not accurate. Damn, they must know something when they say 741 will win
Peach has the least distortion. It should be the original
Last edited:
Yes, my favorite is banana. I have stated many times that blind test should not reveal the straight wire because electronics can add something preferable to the sound. It doesn't mean that I want distortion. Original music is very dynamic and musical. If this sonic and musicality can be achieved with little compromise, why not? I have long found what is necessary for musical enjoyment, even invented new measurement parameters never been used by amp/speaker designers.
Banana has limitation but that is the most musical to my ears.
Slew rate analysis
Many think that opamp with slew rate of 1v/us is sufficient for this low voltage output application but I disagree. I found that SR is one parameter has the highest correlation with performance. For musical enjoyment it is necessary that the sound can be "fast" or sonically "dynamic".
I noticed aggressiveness in pomegranate. It has that "special quality" but surely on the bottom of my preference.
Then I checked for the SR. Best 3 are opa134 (20v/us, the only one I have heard), LM4562 (20v/us) and TLE2072 (45v/us!!!). So TLE is outstanding in this blind test!
Now with such a circuit used by Mooly, do you think such a wideband opamp will have no issue? May be it is the smear ascociated with pomegranate.
I have never heard lm4562, but I wish it is the banana. Opa134 is well known for its musicality and they both have 20v/us but I guess the banana is the lm4562. Just a wild guess.
Many think that opamp with slew rate of 1v/us is sufficient for this low voltage output application but I disagree. I found that SR is one parameter has the highest correlation with performance. For musical enjoyment it is necessary that the sound can be "fast" or sonically "dynamic".
I noticed aggressiveness in pomegranate. It has that "special quality" but surely on the bottom of my preference.
Then I checked for the SR. Best 3 are opa134 (20v/us, the only one I have heard), LM4562 (20v/us) and TLE2072 (45v/us!!!). So TLE is outstanding in this blind test!
Now with such a circuit used by Mooly, do you think such a wideband opamp will have no issue? May be it is the smear ascociated with pomegranate.
I have never heard lm4562, but I wish it is the banana. Opa134 is well known for its musicality and they both have 20v/us but I guess the banana is the lm4562. Just a wild guess.
Yeah, I heard that too, won't give away the identity, but I'll most probably vote it as worst sounding. There are a few others that clip (perhaps mildly) on the first vocal selection.
I may have to download the .wavs instead of the .mp3s to identify the best sounding though - about 3-4 of them are fairly close.
The files are not level matched to the required amount (it should be equal within 0.1dB). This makes the test results useless if not fixed.
Testing for preference can't be done with an ABX test, as this kind of testing only tests for audible difference between 2 files.
You need ABC/HR for testing preference.
ABC/Hidden Reference Free Download
This is a much more elaborate testing protocol and takes up quite a bit of time.
I wonder who is willing to take the time and effort to do a proper ABC/HR test on these files, I certainly am not.
Testing for preference can't be done with an ABX test, as this kind of testing only tests for audible difference between 2 files.
You need ABC/HR for testing preference.
ABC/Hidden Reference Free Download
This is a much more elaborate testing protocol and takes up quite a bit of time.
I wonder who is willing to take the time and effort to do a proper ABC/HR test on these files, I certainly am not.
The other side to this is,
If you attend a HIFI convention and "pick" a sound you like they won't be level matched but you can still pick a sound you like.
The same is true if you go to buy components CD players amps tuners etc, you cant say I want them level matched to 0.1dB.
And the same is also true with live bands (music). I want to listen to all the bands at a concert adjusted to etc.
However you can still hear the difference. Even if the output of a CD is deliberately increased by the manufacturer to give a false sense of bass drive etc.
I don't think the issue is with a scope I can see clipping etc , its what sound do you think is preferred by yourself. And how does this compare to the Opamp type. (when its revealed)
Regards
M. Gregg
Last edited:
The test is really intended to be a subjective "yes, I like that" or "no, that doesn't sound too good" etc and as such it seems to be proving difficult. I'm certainly no expert on using software to look for differences but, even if you find a difference you are still left the choice of "which do I prefer"
And that's what I would really like you to vote on.Thats how I look at these kind of tests. Hopefully you listen and something captures your attention, and given the choice that is the "component" you would carry off home.
There is a very good reason the level requirements are so strict: People perceive slightly louder signals as better, even if there is no other difference between the files. Its just how we humans hear things.
If you do not fix the level differences the test results are invalid!
For me the different levels and types of audible distortion in the WAV samples are quite apparent - if one wants to call that a 'sound' that's fine, I prefer to call it by what I perceive it to be, distortion ...
Edit: loudness doesn't come into it, if anything the slightly louder version will emphasise the irritating qualities of audible distortion, or even provoke any element of the playback system which is adding distortion in its own right to perform slightly worse.
Edit: loudness doesn't come into it, if anything the slightly louder version will emphasise the irritating qualities of audible distortion, or even provoke any element of the playback system which is adding distortion in its own right to perform slightly worse.
Last edited:
I understand what you are saying but they are not test results, they are a simple what do I like and vote.
What you are saying if I'm correct is if they were matched to 0.1dB what you like will be different. This may or may not be the case however you could as easily say that the volume of your playback equipment is identical and it is linear across all frequencies. The same as everyone else's.
Regards
M. Gregg
The topic title is: ULTIMATE OPAMP SHOOTOUT... Where you get to decide.The test is really intended to be a subjective "yes, I like that" or "no, that doesn't sound too good" etc and as such it seems to be proving difficult. I'm certainly no expert on using software to look for differences but, even if you find a difference you are still left the choice of "which do I prefer"
If its just for fun, then you should have picked an other topic title. People can be fooled into thinking the test results are worth something.
Subjective testing, witch is measuring with your ears instead of with machines, has to be done in very strict way's in order to avoid false positives or false negatives.
Let me try to make it clear with an objective testing method we all use: Say you want to measure the voltage output of an amp with different input voltages. Then what you need to do is use the same reference for all measurements, usually we take the signal earth for that. If you randomly plug in the ground probe in the circuit, then the measurement results have no meaning. This is exactly what is happening with this listening test if you don't fix the level differences.
When doing testing, in any form of science, you should only have 1 variable. If you have more than 1 variable, you don't know what is the cause of the result.
I agree,
However are you saying you can't walk into a shop with a HIFI playing and make a decision if you like it or not without a scope in your hand?
Have a look at the thread "sound quality vs measurements" it has been discussed in detail.
I think that is the point of this thread (I could be wrong).
Listen and make a choice, now if that choice is different in a different listening room is subjective and we don't have that option. However I'm pretty sure what I like on these tests will tell me something. (It might be that I can't tell the difference between a high end Opamp and a 741) If that is the case then so be it.
If then I could go on to have a scope etc and see the difference and then say that's the best one. then put it in my equipment and prefer the inferior opamp. then what would I listen to? (The one I like to listen to)
because whats the point of not being happy with the sound?Regards
M. Gregg
Last edited:
It depends on what you want from your hifi system.
Just make sure you do the level matching, if you want to make an honest decision.I think that is the point of this thread (I could be wrong).
Listen and make a choice, now if that choice is different in a different listening room is subjective and we don't have that option. However I'm pretty sure what I like on these tests will tell me something. (It might be that I can't tell the difference between a high end Opamp and a 741) If that is the case then so be it.
If then I could go on to have a scope etc and see the difference and then say that's the best one. then put it in my equipment and prefer the inferior opamp. then what would I listen to? (The one I like to listen to)
Regards
M. Gregg