True in formal circumstances where the question is both reasonable and clearly stated. Are we in those circumstances?
In this case many may not have noticed that many things, both potentially true and false, were presumed.
Just trying to make the best out of a chaotic circumstance and maybe find a little truth, boss! ;-)
Arny, you might be correct. So if there is technical "difference" in file preparation, isn't it surprising if many cannot hear the difference?
After hearing the original, I thought I could do ABX well between that and ANY file, AND with any position in the recording range (it can be shown with the "training mode"), simply because the original file has much stronger "impact", as if the sound level is much higher.
No doubt about it. The real test is in picking preference (but of course, you can say that the circumstances do not allow for preference, or no sound is good enough to prefer).
There are many things in the sound that is not affected by sound level or the difference in time it is recorded. The sound of C, is one of them. It has the sound characteristics of expensive speaker drivers or expensive electronics. Normally this is expensive opamp too (Funnily, NE5532 has a bit of sound characteristics of expensive stuffs, but boring). I don't know if LM4562 is expensive or not, but the specifications are really "expensive".
Last edited:
Just to clarify something about the initial aims of the test. If you look back at the first post, the idea was actually to make this test different (in the hope of getting more to take part) by having the offer of sending one or more files relating to the specific devices to anyone that asked. So if you wanted to compare the 5532 with the 4560 then I would have sent just those files on an individual basis. The idea was to see if your preconceptions of a particular device, and how it might sound would be identifiable by listening. No one has asked for any files.
So we are left with result of 'there are differences' and 'there are not differences' (audible ones of course) which has morphed the question into a 'which do you prefer' type of exercise.
So we are left with result of 'there are differences' and 'there are not differences' (audible ones of course) which has morphed the question into a 'which do you prefer' type of exercise.
Lol 🙂 Well is it true or not ! Do you think there is a difference in level between the six files ?
The only way I could even guess at the level being the same would be to use something like 'wavestats' in Audacity to look at average rms and peak levels.
All I can say with certainty is that only the device under test was changed.
Sorry about that. I wouldn't know one device from another as I have no technical knowledge at all.
Never the less, I did find it interesting that there were differences and that it would be possible to tell them apart.
I'll let you technical lot get on🙂
Yes, that is correct. I stand by my judgement that the files are properly level matched. But one of them has some other fault that should I reveal it, many will start hearing it quite readily. It is also readily measurable.
That's no problem 🙂 In many ways just taking it all at face value and letting your ears decide is the best approach.
Thanks for the reference file. It did not change my opinion of the opamps. None were bad, but all seemed to "erase" a bit of the music and the ambiance. A was worst in that regard. I had a hard time telling the others apart.
I can see how in an analog mixing domain, you'd want to be careful. If you had to bump tracks back and forth, mix and route thru many of these devices, you'd start to lose stuff. The sound could quickly become dull if you aren't careful.
I have not heard the DAC-ADC loop, so don't know how much of the "erasing" is coming from there, of course. Hard to say.
If there is a "tell" in one of the files, I'd love to know what it is. It's like so many other things, you don't see it right in front of your face until someone points it out. 🙂
I can see how in an analog mixing domain, you'd want to be careful. If you had to bump tracks back and forth, mix and route thru many of these devices, you'd start to lose stuff. The sound could quickly become dull if you aren't careful.
I have not heard the DAC-ADC loop, so don't know how much of the "erasing" is coming from there, of course. Hard to say.
If there is a "tell" in one of the files, I'd love to know what it is. It's like so many other things, you don't see it right in front of your face until someone points it out. 🙂
I'm going to go and retrieve the key to all the files in a few minutes.
I wonder about this to, particularly when the likes of D Self (who has designed consoles) suggests that everything we hear has already passed though dozens if not a hundred of such devices.
Remember we have ended up with the curious situation of two 'reference' files. The first is the direct rip from CD, the second posted yesterday due to repeated requests was of just the Micromega player direct into the AD convertor.
Me to 🙂 I was worried earlier, that some trace of the original file names might have been detectable as meta data or something but I put one of the files up on a PC forum and asked if anyone could retrieve any hidden file names. They could not... phew
I'm going to post the key in a few minutes 🙂 so its OK to say if you have found anything amiss.
The big reveal. What the files were.
Here we go, the key to the files and devices.
A = TL072
B = LM833P
C = TLE2072
D = NE5532
E = LM4562
F = JRC4560
I'm going to have a read back at your impressions shortly and see if any pattern emerges.
And as always, thanks to all that took part.
Here we go, the key to the files and devices.
A = TL072
B = LM833P
C = TLE2072
D = NE5532
E = LM4562
F = JRC4560
I'm going to have a read back at your impressions shortly and see if any pattern emerges.
And as always, thanks to all that took part.
The humble TL072. I've always championed this device having had consistently good subjective results from it over the years. The TLE2072 should really be more of the same 'but better' and yet it doesn't quite seem to work out that way.
The LM833P (it must be that version or its equivalent in SOIC) was the one I was really interested in seeing how it performed because it has the quasi comp output stage and is a radical departure from the original LM833 National Semi device.
The 833, the 5532 and the 4562 should really (you would expect) fall into a group of being all fairly similar.
- Status
- Not open for further replies.