[Head Injury]

Quote:

As already stated, he didn't measure or listen to particular opamps.

NwAvGuy didn't? He has an entire article about them and why he used the ones he used.

 

[frenchbat]

Ahemm ...
 
You might want to go and read the article about the Op Amp Measurements before saying he didn't measure particular op-amps. Here's the list :
 

1. NJM2068
2. NE5532
3. OPA2134
4. OPA2227
5. OPA2277
6. LM4562
7. TL072

 
Quote:

As already stated, he didn't measure or listen to particular opamps.

 
EDIT : Head Injury was faster than me, but I put on the list ...

 
 
 

 

[mikeaj]

Quote:

Yes, it is. The opamps in this amp were chosen because they had the best performance in the circuit of all the ones tested. If you roll opamps, the changes you hear are due to degraded performance or placebo, until you prove otherwise with measurements. You should really assume this about any amp, because it's nice to think the engineers knew what they were doing.

IMHO this is kind of a stretch, or you need to be more careful about the wording.  If you consider battery operation power consumption and cost, the chosen gain stage NJM2068 was easily the best of those tested.  If you consider just power consumption and performance, NJM2068 was the best.  Going by raw performance, OPA2227 did better, and some alternatives had lower distortion but higher noise.  But this is based just on just a few tests, which are not particularly conclusive.  It may be premature to declare a winner based on just that data, if you're looking at just performance and ignoring the power consumption.
 
The point is that aside from the outdated TL027, everything did really well, to the point that differences are quite boring and marginal--other than potentially noise levels, which could be relevant if you're using very sensitive IEMs in some very quiet listening environment.  There's not much motivation to look harder at alternatives or investigate the performance more carefully when a relatively cursory overview already finds something sufficiently good that's cheap.
 
There are definitely parts with higher performance, even for audio purposes and even for a gain stage.  However, with that circuit and PCB and other surrounding parts, you might not be able to reach the potential of those expensive op amps.  (This is aside from whether or not those improvements in the gain stage would be audible, which seems highly unlikely for reasons previously discussed.)

 

[upstateguy]

Quote:

 
Yes, when someone hears a difference with OPA627, AD797, AD8620 etc. it's thanks to "degraded performance", clearly... that is what I'm hearing, 'OK'.
 

The audio DiffMaker is perfect for testing differences between opamps.  If you can hear a difference the diffMaker program will isolate it for you.
 
Ti Kan did a manual null for me between the OPA637 and the AD8610 in an M^3.  The difference was at -45dB and probably due to doing it manually.
 
So the real challenge is to post some diffMaker recordings of these differences so we can all evaluate it.

 

[Naim.F.C]

If he measured it to be a near perfect neutral response with a flat curve and inaudible noise, I very much doubt it'd make a difference even if superior opamps were used. You can't really fix something that has nothing there to be fixed. Unless of course it could bring less noise than already inaudible noise or something unnecessary like that. 
 
Quote:

As already stated, he didn't measure or listen to particular opamps.
 
If an engineer builds a merry go round and paints the horses, it doesn't mean he used the best paint, especially if he never even saw or tested that paint.
 

 
EDIT: Also, why has this been moved to Sound Science? Lol. Surely it belongs in the Headphone Amps section? Scientific or technical discussions about amps are only natural in amp threads.
 

 

[mikeaj]

Quote:

EDIT: Also, why has this been moved to Sound Science? Lol. Surely it belongs in the Headphone Amps section? Scientific or technical discussions about amps are only natural in amp threads.

This is not that uncommon.  Once the number of "contentious" posts surpasses a certain threshold, the powers that be relegate the discussion to the ghetto of head-fi:  Sound Science.  There it may continue to fester without disturbing the ordinary netizens.  Draygonn totally called it. 
 
One interpretation is that such arguments about blind testing, measurements and listening impressions, etc. are old and stale, and nobody ever comes to an agreement, so it's kind of annoying and pointless to see happen again.  It's cleaning the garbage to move the discussion elsewhere.  The other interpretation is that such discussions give rise to thinking that would not be favorable to head-fi's many sponsors, some of which rely on some amount of misinformation or false hype to generate sales.  e.g. If you take the view that most well-designed solid-state amps aiming for accuracy (and accomplishing that) should sound the same, then you're less likely to purchase new solid-state amps, even ones that are very good.  Also, for various reasons, the powers that be aren't exactly fans of V.
 
Those responsible for the move maybe think they have reasoning more along the lines of the first interpretation, while V would be convinced it's clearly the second.  I would think that both are factors, but of course I can't know for sure.

 

[kiteki]

Quote:

NwAvGuy didn't? He has an entire article about them and why he used the ones he used.

 

 

[Head Injury]

Don't try to derail with pony images, now. The latest episode was about being assertive, and I'm putting my hoof down.

 

[uelover]

Unsubscribed.

 

[kiteki]

Quote:

Don't try to derail with pony images, now. The latest episode was about being assertive, and I'm putting my hoof down.

You can put your hoofs down. =]
 
In post #1345, I am only at stage 1.
 
 
Quote:

 
There are definitely parts with higher performance, even for audio purposes and even for a gain stage.  However, with that circuit and PCB and other surrounding parts, you might not be able to reach the potential of those expensive op amps.

Sure, but before we put it in the stable of most transparent hi-end amplifier ever necessary, there should at least be some experimentation, at least in my view.  There are electrical engineers which think opamps sound different for reasons beyond their dScope.

 
Quote:

The audio DiffMaker is perfect for testing differences between opamps.  If you can hear a difference the diffMaker program will isolate it for you.
 
Ti Kan did a manual null for me between the OPA637 and the AD8610 in an M^3.  The difference was at -45dB and probably due to doing it manually.
 
So the real challenge is to post some diffMaker recordings of these differences so we can all evaluate it.

That's interesting, but then we are listening to the sound of recordings with opamps, that is why I don't see the examples of "100 opamps in a studio recording" or the Meyer & Moran ADC/DAC CD duplicates as perfectly valid.
 

 

[mikeaj]

Quote:

Sure, but before we put it in the stable of most transparent hi-end amplifier ever necessary, there should at least be some experimentation, at least in my view.  There are electrical engineers which think opamps sound different for reasons beyond their dScope.

 
Considering all that I don't know, and the level of education and thinking of plenty of electrical engineering graduates, I must say that some electrical engineers thinking anything is not much of a surprise to me.  It doesn't particularly signify much, all things considered.  And a lot of times for audio you hear from audio engineers, who may have a much different background.
 
But I'm always down for more experimentation.  Who's up?

 
 

That's interesting, but then we are listening to the sound of recordings with opamps, that is why I don't see the examples of "100 opamps in a studio recording" or the Meyer & Moran ADC/DAC CD duplicates as perfectly valid.

 
Maybe I'm just being really dense, but I don't quite follow what you're saying here.  When you say "sound of recordings with opamps" are you talking about op amps in our playback chain or in the studio recording/mastering chain?  What does this have to do with a null test?
 
At least for the O2 and these gain stage op amps, for a single tone test, we'd be looking at a difference in the 0.001% (-100 dB) range or so, or a little lower.  For real music, which has multiple tones, or an IMD test, you would probably expect a little more differentiation, but not like an order of magnitude, which would still be vanishingly small.

 

[kiteki]

     Quote:

Originally Posted by mikeaj /img/forum/go_quote.gif
 
Maybe I'm just being really dense, but I don't quite follow what you're saying here. When you say "sound of recordings with opamps" are you talking about op amps in our playback chain or in the studio recording/mastering chain? What does this have to do with a null test?

Yeah I mean the two above are different.

When upstateguy said "recordings of these differences so we can all evaluate it."  It looked like he meant listening to the latter, using the former.
 
 
Quote:

 
But I'm always down for more experimentation. Who's up?
 

That's more like it, I think I'll start with NE5532 versus OPA2134 (since they have been tested), if I can't hear any difference at all, then move on to something like NE5532 versus OPA627, still no difference?  Then I'd find that quite interesting and most likely quit, with some satisfaction of the result.
 
 

 

[mikeaj]

Why would you suspect that op amps in the playback system would be different than those in the studio?  These are often some of the exact same models, used in similar circuits for similar purposes.  I can't think of what the distinction would be.

 

[Willakan]

There are engineers that believe lots of things. None of these things have ever been produced under controlled conditions. I wonder why.
 
Also, how did this end up in Sound Science? Kinda ironic, surely, moving a thread that is turning to the subject of opamp rolling into Sound Science...
Also, where's the V200 vs O2 review?

 

[kiteki]

Quote:

Why would you suspect that op amps in the playback system would be different than those in the studio?  These are often some of the exact same models, used in similar circuits for similar purposes.  I can't think of what the distinction would be.

If opamps have a certain sound-quality or performance inherent to them, or even a type of signature, then that will be inherent to all music played through them, from all studios and decades.
 
How am I supposed to listen to the performance of X opamp in X studio using X opamp?