The original will not stand out. What will are those with extra sonic and distortion added by the amp stages, unless you focus on thd ( people tend to focus on what is good or terrible)
Indeed I needed to think hard about it. I was confused from previous experience where orange has the highest distortion but is favorable. I'm not familiar with 741 sound but it is hard to believe if it can sound like orange, which is close to tube sound. In my prediction 741 should have sounded hard but vintage chips do sound more open.
But there is clear difference between orange and B here. They have high distortion (orange was worst) but B is very fatiguing. I'm familiar with opa134 and it is not fatiguing because the distortion is second order.
Ahhh ... an old friend, I've only listened to this a few thousand times - off the Denon. It really gets interesting listening to the -60dB version, hearing digital bits spitting and popping - depending on the hardware used ...
It's a nice test. I once ordered 2134's from HK and had already mounted them in an active xover, when I discoverd they were fakes. Didn't bother to change them, because they measured and sounded ok. Question for me still is what opamp is cheap enough to be used as the basis for a 2134 fake and yet sound so nice. Perhaps this test will show me.
I'm thinking if the the 741 is too slow that the "sharpness" in the signal gets rounded off by the slower amp. Negative feedback could cause cause overshoot in the slower chip and give an impression of sharpness too so, heck, I could be wrong on my pick.
A null test would be more definitive as to which chip changes the signal the least, but not necessarily which one sounds the best.
There are all surprisingly similar. I'm going to vote Orange, as it sounds the most open and cleanest to me. Second place goes to Pear, as it seems to distort the least on the choral piece. I've listened on headphones from the computer and on speakers powered by tubes. My impressions were pretty much the same on both systems.
For a better ranking, might have to cut the clips apart and listen that way.
For a better ranking, might have to cut the clips apart and listen that way.
Tube amp users should really pick the ORANGE!
Subjectivists should really pick the PEACH
Musicians, may be the BANANA or pomegranate
I read this some time back and it gives some clues why you don't really want mud slow amps in your system if they can be avoided. When your opamp approaches slew limits it is for all practical purposes running with no feedback, sort of 'dynamic clipping'.
Op amp myths – by Barrie Gilbert | Embedded
G²
I didn't expect it but yes I could. If not because of blind test, I will not be so sure. The second test is actually harder but once I know what to expect from each, it is very easy to tell apart (100% accuracy from 5 trials, and I know it is repeatable). The more time I invest in recognizing each sound, the more differences I can hear.
BTW:
I must tell you that I think my speaker is quite special. I use cheap amp (TDA2030A), cheap source, MP3, but my speaker is the culmination of my knowledge in speaker design. Complex crossover, but resolve better than fullrangers. And much better than headphones I have access to.
My first trial was using $5 headphone and I picked the banana. Later with the speaker I still pick the banana but I know more with the speaker. I don't think headphone can do it
Last edited:
J.S. Bach Prelude from Partita no.3 in E major BWV1006
This one,
Bach Partita No 3 BWV 1006 (I.Prelude) - Rachel Podger - YouTube
AFAIK, they're not in this test, but NE5532, LM833 and LF353 are inexpensive candidates that can be passed off as the OPA2134.
Mooly, I'd like to enter my discrete opamp(s) for some future edition of this shootout - what form-factor is preferred, and how many are required? (I'm guessing 4x DIP8 single opamp).
Last edited:
Yes, something like an LF353 or TLO72/82 would probably pass as an OPA2134. If its bjt and not FET then that is easy to pick up, beyond that and its less obvious without detailed tests.
Well I don't know if there'll be another shootout quite like this one but if you have discrete opamps of your own then its quite easy to make the files yourself to compare against another device.