Does Op-amp Rolling Work?

[daniboun]

@amirm
I think it would be rather interesting in the end to measure a bypass of the op amp instead. Audiophonics has released a bypass op amp (ghost opa) and I think we might be surprised by the results

AUDIOPHONICS GHOST OPA DIP8 Bypass

The purpose of the Audiophonics Ghost OPA is to bypass the interface card on the 450ET range of power amplifiers

www.audiophonics.fr

 

[Bleib]

The price of the OP-amps puts in the Topping PA5 II price territory. Better get that instead if there is no need for bass and treble EQ

 

[Hiten]

Thanks for review. It may appear another opamp rolling review has nothing new to show. But it does help new members who has not searched or found any precise info or are totally ignorant about it. Besides a reminder about subject never hurts, because such myths/experiments/mods comes on audio forums at regular intervals.

How about the plexiglass is eye-candy, aimed at a specific "type of customer" and potentially a feature that may have sold amps just for this reason alone.

varying color led lights serves no purpose. I am no expert but if implemented wrongly it may introduce noise and distortion depending on design i guess.

here is a discussion about it (just to be clear the amplifier and problem may be completely different )
LED Noise

Regds.

 

[Mnyb]

The point is, Douk are selling snake oil on an otherwise fine product. I've never seen an op amp swap improve performance in a measurable and significant way. That would require the original designer to have made a terrible decision on the original opamp.

1+ another paradox the terrible design you can improve I would rather wonder what else is wrong if does not have the correct op from the factory ?

 

[RickNRoll]

There are many applications for them that require different designs. An "instrumentation" op-amp for example produces least amount of DC offset. A "video" one has very high bandwidth. Some come with one or man op-amps in one package. Some cost more than the others. And yes, there are performance differences between them.

The basic point of the op-amp design from the start was to make it appear to be completely irrelevant to the components. Each op-amp in an ideal world would be exactly the same as the next.

Of course, it's not a perfect world, but the designs are good enough that for purposes like this any competently designed and built op-amp will sound just the same as the next.

 

[solderdude]

When opamps can make a difference:

In high gain / high bandwidth circuits (also in audio applications like microphone/MM/MC inputs)
In very low noise circuits (also in audio applications like microphone/MM/MC inputs)
In circuits that require an (extremely) high input resistance (instrumentation/sensors, high Q gyrators)
In circuits designed to work rail-to-rail input/output conditions (low power supply audio for instance)
In circuits where there can be a huge difference in - and + input voltages (comparators, processing of signals etc)
In circuits designed for very low (a few V) or >36V circuits (could be some audio gear)
When it has to drive loads that are below 100ohm (audio applications for instance like headphone outs)
In circuits where a very high CMRR is needed
In circuits where a very high PSRR is needed
When a circuit is used as a (+/-)1x buffer (audio applications)
When a circuit does not require SOTA performance or when it does require SOTA performance in certain applications

Because of these (and more) reasons so many op-amps exist, as well as the fact that money can be made selling op-amps.

 

[beefkabob]

1+ another paradox the terrible design you can improve I would rather wonder what else is wrong if does not have the correct op from the factory ?

With something as theoretically straightforward as audio components, I'd like brands to just sell me the product done right. Don't offer me opamp combos. Don't offer different power supplies. Just sell me the total package. It's not like a car where people are good with different options. Even with cars, it's not like they offer the 16-bit ECU and the 32-bit ECU for a higher price.

 

[KL....]

Basically the performance is dominated/limited by the class-D output circuit.

Yes, that is a very reasonable conclusion, isn't it?, that is the downstream dominates and is not as transparent as the feeder. Perhaps, swapping opamps in this case can be called a synergy (perhaps), couldn't it?.... but nice to see it is available, isn't it :=)

edit: with Douk allowing, in this instance, opamp rolling (as it is called), they really should provide the necessary opamp specs (and the range of specs) that will permit safe opamp rolling/plugNplay, shouldn't they?

 

[beefkabob]

When opamps can make a difference:

In high gain / high bandwidth circuits (also in audio applications like microphone/MM/MC inputs)
In very low noise circuits (also in audio applications like microphone/MM/MC inputs)
In circuits that require a very, very high input resistance (instrumentation/sensors)
In circuits designed to work rail-to-rail input/output conditions (low power supply audio for instance)
In circuits where there can be a huge difference in - and + input voltages (comparators etc)
In circuits designed for very low (a few V) or >36V circuits (could be some audio gear)
When it has to drive loads that are below 100ohm ((audio applications for instance)
When a circuit is used as a (+/-)1x buffer (audio applications)
When a circuit does not require SOTA performance or when it does require SOTA performance in certain applications

Because of these (and more) reasons so many op-amps exist, as well as the fact that money can be made selling op-amps.

Yes, but who is going to benefit from swapping op amps unless their product came with improperly specced opamps for the use case/design?

 

[amirm]

I think until we Test it we never know.

Thankfully I have tested them: https://www.audiosciencereview.com/...-opamp-review-sonic-imagery-vs-sparkos.10325/

They actually made things a hair worse compared to the stock opamp.

 

[amirm]

In high gain / high bandwidth circuits (also in audio applications like microphone/MM/MC inputs)
In very low noise circuits (also in audio applications like microphone/MM/MC inputs)
In circuits that require an extremely high input resistance (instrumentation/sensors)
In circuits designed to work rail-to-rail input/output conditions (low power supply audio for instance)
In circuits where there can be a huge difference in - and + input voltages (comparators, processing of signals etc)
In circuits designed for very low (a few V) or >36V circuits (could be some audio gear)
When it has to drive loads that are below 100ohm (audio applications for instance like headphone outs)
In circuits where a very high CMRR is needed
In circuits where a very high PSRR is needed
When a circuit is used as a (+/-)1x buffer (audio applications)
When a circuit does not require SOTA performance or when it does require SOTA performance in certain applications

Good list but as you know, none of that explains the larger soundstage, better bass slam, less fatiguing, etc. that folks describe when they swap opamps.

 

[solderdude]

Yes, but who is going to benefit from swapping op amps unless their product came with improperly specced opamps for the use case/design?

Manufacturers and distributors of expensive op-amps benefit from this in general.
Manufacturers building audio gear that has specific provisions for 'op-amp rolling' benefit as people want the 'experience'.
Also the buyers/swappers benefit as they feel 'their ... device sounds MUCH better' while technically nothing has changed.
In some rare cases there might be a genuine improvement in technical performance or there might be worsening.
In some cases there might even occur oscillations which may or may not be audible.

Good list but as you know, none of that explains the larger soundstage, better bass slam, less fatiguing, etc. that folks describe when they swap opamps.

Placebo does... but that is not in the list of reasons to select the right op-amp for the job.

The weird thing is that no matter how often measurements show no discernible differences the op-amp swappers will continue to swap and continue to 'report' their findings and encourage others to 'swap' a bunch of op-amps because they want to hear 'better sound quality'.
There is a large audiophile community that is built on placebo and no matter how well we try we can't change their minds... in fact it makes the gap between the worlds even bigger.

What reviews like this ARE good for is for people that like to see things measured and realize audio electronics is just electronics and is bound to physics.
Unfortunately this group of people is very small compared to the audiophile and audiophool group.
Fortunately the actual test data is made available here which is a good thing .... as well as the fact that it can be discussed when measurements are made.

Perhaps, swapping opamps in this case can be called a synergy (perhaps), couldn't it?.... but nice to see it is available, isn't it :=)

There is no 'synergy' here but audiophiles love that word. The circuit they are in does not have an influence of the overall performance of the device it is in except when a not well suited op-amp is used in which case the performance could be worse. Fortunately for swappers this could still mean an improvement.
It is nice for 'op-amp swappers' that the option is easily accessible.
It is nice for op-amp-manufacturers that they can sell a few more opamps (gear manufacturers order by the thousands).
It works wonders for owners who can finally hear things they want even if there is no actual improvement happening.

 

[Sokel]

I think we're missing the whole point of the thread here.
To me it's not about the op-amps but about the way to "see" what they do and the potential damage they can cause.

Amir was lucky here testing them on a device which can probably tolerate some deviations.
What I have seen though for example, is some really nice LME49720 (original, Mouser sourced) having 30dB worst noise and an abnormal FR against even an ancient RC4558P salvaged from a SONY all in one! (couple of caps near fixed it, but still)

It's all about the ability to see, to know what every change does. It does not need an AP, a decent ADC and a free REW or Multitone Analyzer can show you in seconds.

Thanks Amir!

 

[morillon]

we can't stop pointing out dac machines etc to the high-performance 1k sinad, which at the same time, points out the 1611/12 etc other "recent' racing "aops having even become an important "readable" marketing element....etc and be surprised by the presence of dip8 and the play of aops (not new) ....

after the case of the little fosi phono remains interesting showing precisely the two marketing elements in action (161* and dip) , even if clumsy on the engineering side and also possible clumsiness on the side of the measurement protocols which can be "missed"...
a textbook case...
- components marketing
- DIY game marketing "customizable and do better"
- the necessary doubt by principle considering the limits of our measurement, simplifying, processes...



manufacturers themselves feeding the subject...it's not likely to dry it up

;-)

 

[amirm]

manufacturers themselves feeding the subject...it's not likely to dry it up

I am seeing that youtubers and subjectivists promoting them more than companies. It has become a lot more fashionable in the last couple of yeas than before.

 

[jools]

Good list but as you know, none of that explains the larger soundstage, better bass slam, less fatiguing, etc. that folks describe when they swap opamps.

I haven't tried but I'm sure I'd find fiddling about swapping opamps for nothing to be highly fatiguing, very quickly.

 

[ObjectiveSubjectivist]

Every time I hear the term "rolling", being tube or op-amp rolling, I roll my eyes

 

[Bob101]

It's highly unlikely that any designer has chosen the wrong opamp, so you can usually only make things worse, and where better is possible it's likely inaudible anyway, though not always, but certainly in 99.9999% of cases.

After informing with Audiophonics they basically said: op-amps are op-amps, and virtually all are interchangeable. Just for fun I tried rolling the stock op-amps in my HPA S-400ET with a working pair of BurrBrown OPA2134PA. After correctly installing them, the amp did NOT like these at all and started having sonical issues. I swiftly replaced them back with the stock ones, and all was well again. There might be an electronical reason (one of the numbers on the product sheet I don't understand that explains this) but my adventures in op-amp rolling ended there. The BB's are back in my Aiyima A04, where they do an excellent job of making no difference at all compared to the stock op-amps that it came with, which I deemed, trough reading many and wrong forums, not 'musical' enough. They will remain the only op-amps I'll roll.

 

[Bob101]

I haven't tried but I'm sure I'd find fiddling about swapping opamps for nothing to be highly fatiguing, very quickly.

 

[Bob101]

I never understood what op-amp or tube rolling should deliver. Even if you want to change the frequency reponse of your device, just use EQ.

Tube-rolling is another matter, because there is measuable variation in characteristics for any individual tube of the same type by the same manufacturer. Because 'tubes'. Manufacturers of tubes often sell measured and selected matched pairs of them for that reason. I'm not into it at all but tube-rolling is defensable, as opposed to op-amp rolling. But feel free to roll all the op-amps in the world, by all means.