The third sample is distorted so all bets are off.
-
Welcome to ASR. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!
Serious Question: How can Op Amp rolling improve the sound of an already well measuring device? Are that many confused? (Master Thread)
- Thread starter antcollinet
- Start date
It's like someone who doesn't know what they're doing randomly changing parts in a well-designed circuit, hoping that by chance, these other parts will fit in with the surrounding circuitry and actually improve sound quality.
So is the NE5532 at 1/100 the cost and a much smaller package.
OP
antcollinet
Grand Contributor
- Thread Starter
- #264
Oscillation, perhaps?
Geert
Major Contributor
- Joined
- Mar 20, 2020
- Messages
- 2,198
- Likes
- 4,415
I was still wondering why we couldn't get a better null via DeltaWave, and why the null had such large phase differences in the high mids.
So I checked the phase with a multiband phase correlation meter and DeltaWave had it right; in the second track the phase of the high-mids, compared to the rest of the frequency spectrum, is reversed. Single screenshot, capturing the phase correlation of both stereo tracks at the same point in time:
That's crazy
There's more going on here then a harmless opamp swap.
So I checked the phase with a multiband phase correlation meter and DeltaWave had it right; in the second track the phase of the high-mids, compared to the rest of the frequency spectrum, is reversed. Single screenshot, capturing the phase correlation of both stereo tracks at the same point in time:
That's crazy
There's more going on here then a harmless opamp swap.
Last edited:
audio_tony
Addicted to Fun and Learning
Am I correct in saying that the tests were carried out by replacing the OPAMP in the buffer / input stage of a class D amp?
If so, any distortions etc. will be completely masked by those of the amp surely?
I suggest that the proper way to test for differences (audible or otherwise) between OPAMPS would be to use a reference circuit with an average gain (around x3 to x5) and then insert this reference in the source path and listen / measure.
To compensate for the additional gain, a simple resistive divider could be inserted on the output of the reference circuit, to enable level matching.
If so, any distortions etc. will be completely masked by those of the amp surely?
I suggest that the proper way to test for differences (audible or otherwise) between OPAMPS would be to use a reference circuit with an average gain (around x3 to x5) and then insert this reference in the source path and listen / measure.
To compensate for the additional gain, a simple resistive divider could be inserted on the output of the reference circuit, to enable level matching.
JeremyFife
Major Contributor
What would be the point of that test - it won't show the measurable/audible impact of an opamp in actual use.
The original test does exactly that.
It's clear that different opamps measure differently. The point of this thread is to determine if that makes any practical difference for hifi ... it doesn't
audio_tony
Addicted to Fun and Learning
The OPAMP would be in actual use though, as part of the reference circuit - if trying to demonstrate audible differences, it doesn't matter what circuit it's inserted into.
So long as the external circuit parameters remain the same, any differeces in sound would be clearly audible (if they existed of course).
Obviously I know that's the point of the thread. I also know that it doesn't make any difference, regardless of circuit.
However... my original point was that if.... if it was possible to hear audible differences between OPAMPS, the last place to listen for any audible differences would be in the input stage of a class D amp, where any (potentially* subtle) differences would be masked by the power amp itself.
If it was remotely possible to hear subtle differences between OPAMPS then surely the best test method would be a reference circuit as I suggested, followed by a very good quality headphone amp and headphones.
* there won't be any audible difference
What reference circuit? Some opamps need external compensation. Some opamps are better at driving capacitive loads so need a smaller external isolation resistor. Some opamps can drive larger loads etc., etc. Thats why manufacturers make dozens of different opamps. When designing you pick the opamp thats best suited to your application, the circuit its in, including the source and load. Than how do you make a reference circuit thats good for all opamps ?
I think so, yes.
Yes, of course.
This kind of test may be useful, but that's not what this thread is about. It's about opamp rolling, replacing an opamp in a given circuit by another one.
johny_2000
Addicted to Fun and Learning
IMO, a good application for testing op amps would be a DAC with all three output op amps swappable. Thus, their effect will be direct (not masked) by any other subsequent amplifying circuits. A quick search didn't reveal any new high-performance DACs with these options from the factory. There are a few that offer a socketed buffer op amp.
Only the Asus Xonar Essence STX and Creative Sound Blaster ZxR sound cards came with all three (four) op-amps (IV L/R and Buf) in the sockets.
Only the Asus Xonar Essence STX and Creative Sound Blaster ZxR sound cards came with all three (four) op-amps (IV L/R and Buf) in the sockets.
Last edited:
johny_2000
Addicted to Fun and Learning
Something like this NEOHIPO DC10 ($60) has a proper implementation of swappable op amps in a recent (Q4/24) USB DAC. This DAC is not a SOTA since it uses the ES9028Q2M chip with minimal effort on output specs (103dB SINAD), but the general concept seems like a good template for rolling op amps.
johny_2000
Addicted to Fun and Learning
Also, this Fosi Audio ZD3 ($180, 117 dB SINAD) equipped with the latest DAC chip (ES9039Q2M) has the ability to replace balanced (2 pcs., Left/Right) and unbalanced (1 pc., both channels) operational amplifiers.
Last edited:
... which doesn't make sense as the LM49720 is one of the very best opamps available. Chances are high that rolling in other types makes things worse, not better.
solderdude
Grand Contributor
I/V converting is not the same as a buffer circuit though.
The RCA output converting is more like a buffer circuit but relies on CM input differences a bit more.
Which op-amp performs technically better in which position is one question.
What the option to roll is about is sales .. that wonderful new 'op-amp rolling for different sound' hobby is what the DIP sockets are all about.
This is feeding the 'DAC chips do not matter as much as the analog output stage' myth.
The RCA output converting is more like a buffer circuit but relies on CM input differences a bit more.
Which op-amp performs technically better in which position is one question.
What the option to roll is about is sales .. that wonderful new 'op-amp rolling for different sound' hobby is what the DIP sockets are all about.
This is feeding the 'DAC chips do not matter as much as the analog output stage' myth.
And not all buffers are equal. DAC inputs are capacitive and some OAs are better with this kind of load (stability and distortion). A common method to deal with cap loads is an isolation resistor on the output of the buffer. How large a resistor depends on the OA as much as the load. So the wrong OA swap will cause distortion and maybe instability. That this is not usually shown in data sheets dosnt help.
Thighsofsteel
Member
The latter. He is a charlatan.
babadono
Active Member
- Joined
- Jul 8, 2021
- Messages
- 260
- Likes
- 225
+1,000,000 yes yes yes...what is it that people do not understand about this.
babadono
Active Member
- Joined
- Jul 8, 2021
- Messages
- 260
- Likes
- 225
And in the unit pictured they used high quality machined sockets, at least it appears so.
Old_School_Brad
Major Contributor
- Joined
- Sep 10, 2024
- Messages
- 1,829
- Likes
- 2,474
It's an ego-driven, narcissistic tendency that prevents admitting to being wrong or having been misled by a clever salesperson. Acknowledging that you’ve been outsmarted and essentially lost money can be a tough pill to swallow.
And in other cases it's much simpler..
Similar threads
- Poll
