Sparkos SS3602 Opamp Rolling In Fosi P4

[Mnyb]

Sincerely, you have been gaslighting. The nitpicking and FUD-casting on other people's measurements and data. Jumping from topic to topic to keep from having to account for the things you are claiming. Misinterpreting. Not answering direct questions. Appeals to authority, and elitism. You seem to think you have unique access to stuff and experiences, which is odd given the forum you are posting at. For example, I have a box full of your unicorn OPA627. Let's check them out.

But first, briefly return to an example earlier in the thread, the OPA2604, and establish a baseline for distortion performance before I shoehorn some OPA627 into the preamp I have been using here.

I loaded pairs of NE5532, OP275, TL072, OPA2227, OPA2228, LM4562, OPA2604, OP270, OPA2132, OPA2134, and LF412C into the preamp. I swept the output to clipping. Here are THD+N and THD for the group. I zoom in on THD to get a better view.
View attachment 471896

The NE5532 and a handful of OpAmps all stand out with best distortion performance in this application. The OPA2134 is among the best for distortion, which is great since I also have the OPA134, and I can make single- to dual-OpAmp adapters.

View attachment 471897

I made two adapters since my preamp has two gain stages. The adapters prevented me from closing the lid on the preamp. But I see lots of OpAmp rollers who can no longer close lids on their gear. Cut holes. Etc. So I'm fine.

To establish a baseline between the OPA2134 performance and the OPA134 mounted on my adapters, I did four runs. I did a run with all OPA2134 in the preamp, a pair of runs with just one OPA134 adapter configured with an OPA2134, and a run with both OpAmps using the OPA134 adapters. All of these are line on line matched for distortion, which I measured since you would have cast doubt on the validity of the result since my adapters are crude, not made from exotic materials.

View attachment 471903
I get the same results with the OPA134 as the OPA2134. Let's compare the OPA627!

Next, I measure the OPA627 across output voltage. The OPA627 has slightly lower distortion at 2-4V.
View attachment 471906
This is great performance, the differences are tiny, not audible.

Here are OPA627 compared to OPA134 at 2V output.
View attachment 471905
There is simply no audible difference between these. The slightly different 2nd to 3rd HD between the OpAmps is just not notable at this low level.

Try not to overinterpret the gain again, I am attenuating to 6dB gain with a volume control - it's a preamp after all. The internal gain is much higher. I measure stuff where it will most commonly be used. Not in some corner case configuration like you keep suggesting.

And I measured the performance across output for you again, so that's covered. And I made sure I wasn't doing something silly or unfair to either OpAmp by overloading an input, despite some of your odd coaching you gave me earlier. Which is why I was laughing at you when you accused others of gaslighting.

Hey, since you were speculating, what other physical property determines how sound is reproduced?

wow good work sir

Btw i assume there is a bit of randomness to the noise etc due to physics ?

So over-nitpicking the graphs is a bit pointless as measurement of the same thing several times can look a tiny bit different I assume ?

 

[MAB]

wow good work sir

Btw i assume there is a bit of randomness to the noise etc due to physics ?

So over-nitpicking the graphs is a bit pointless as measurement of the same thing several times can look a tiny bit different I assume ?

Each OpAmp truly does have slightly different performance. The OPA627 vs. OPA134 difference is absolutely tiny.

The OPA627 has ~2dB better THD than the OPA134 at 2V output:

It's at the limit of my analyzer, way beyond my hearing. There is a technical difference, just not an audible one. Also, if you did some circuit optimization, and for sure got rid of the socket did a better job of ground and decoupling caps, you could likely get better performance out of both of these OpAmps. The removable socket itself, and crappy circuit layout that carries a small penalty at this level of performance. Although not that bad since the OPA134's on the riser didn't test much different than the OPA2134, so my hack isn't too bad. For sure high speed devices will perform less stably though. If the goal was best performance, most of these sockets would be eliminated.

Even among the dual-OpAmp models, the TL072 and OPA2604 aren't going to be audibly worse, but for sure they consistently measure differently in this circuit. The OPA2228 is high bandwidth, and I think needs some HF attenuation to control the noise, it isn't a likely candidate for rolling, I just had to try.

The classic NE5532, the OPA2132 and 2134, the LM4562, and the OPA2227 are indistinguishable. The results might change slightly if I change the internal gain of the preamp, which I can do with DIP switches, but these are all going to be pretty close in this non-inverting design. This group is nearly indistinguishable from run to run variation.

For sure there are real differences here. But the application is way more important than swapping nearly equivalent parts, or accidentally degrading things.

 

[JSmith]

If you have poor speakers or power amplifier, you will never hear the difference.

Rubbish.

I disagree.

You are welcome to do so, but if you want to be taken seriously then you'll need to provide proper testing evidence of your claims.

I believe you are trolling this thread.

gaslighting

I believe you are doing both these things.

JSmith

 

[MaxwellsEq]

Each OpAmp truly does have slightly different performance. The OPA627 vs. OPA134 difference is absolutely tiny.

The OPA627 has ~2dB better THD than the OPA134 at 2V output:
View attachment 475226
It's at the limit of my analyzer, way beyond my hearing. There is a technical difference, just not an audible one. Also, if you did some circuit optimization, and for sure got rid of the socket did a better job of ground and decoupling caps, you could likely get better performance out of both of these OpAmps. The removable socket itself, and crappy circuit layout that carries a small penalty at this level of performance. Although not that bad since the OPA134's on the riser didn't test much different than the OPA2134, so my hack isn't too bad. For sure high speed devices will perform less stably though. If the goal was best performance, most of these sockets would be eliminated.

Even among the dual-OpAmp models, the TL072 and OPA2604 aren't going to be audibly worse, but for sure they consistently measure differently in this circuit. The OPA2228 is high bandwidth, and I think needs some HF attenuation to control the noise, it isn't a likely candidate for rolling, I just had to try.
View attachment 475229
The classic NE5532, the OPA2132 and 2134, the LM4562, and the OPA2227 are indistinguishable. The results might change slightly if I change the internal gain of the preamp, which I can do with DIP switches, but these are all going to be pretty close in this non-inverting design. This group is nearly indistinguishable from run to run variation.

For sure there are real differences here. But the application is way more important than swapping nearly equivalent parts, or accidentally degrading things.

The designers of the NE5532 did an astonishingly good job.

 

[Mnyb]

The designers of the NE5532 did an astonishingly good job.

They basically "solved" the audio op amp problem for a very vide range of applications ?

Suppose there a few ones where a more suitable component could be found , but again due to the actual circiut it should work with , not due to "sound" .

This seems to rile up some audiophiles to no end ? But makes life's easy and cheap and reliable for engineers and designers , "lets put an NE5532 there it would be good enough "

 

[gino1961]

Very interesting thread expecially now that i am a happy P4 owner
however some headphone amps show even better performance
and for similar price
and they can be used as line preamps as well

i am currently evaluating the option seriously
what do you think ?

 

[JeremyFife]

Very interesting thread expecially now that i am a happy P4 owner
however some headphone amps show even better performance
and for similar price
and they can be used as line preamps as well

View attachment 475815

i am currently evaluating the option seriously
what do you think ?

I think you will always find something that measures better (or differently) ... take a step back and consider if this has any impact on audibility (is the difference simply more inaudible!). Chasing perfection at the expense of more than good enough is a waste of your time. Enjoy the music

 

[gino1961]

Hi thank you very much for the very kind and valuable advice
I grew up reading audio mags especially from Uk always hunting for best buys
Then it came the period of the giant killers
I have compared the P4 to an old Electrocompaniet ec4.5 and a Mission Cyrus preamp (fwiu also a good unit in its days ?)
The differences if existing are quite small with the Electro a little more rounded and therefore softer on ears
I mean when the cost of the unit is small i would buy all compare them and try to resell what i like less

 

[Joe Smith]

Amir did sometimes comment on his listening experience, and i think he still does with some categories, such as headphones. But of course reporting one person's subjective experience, no matter how trained their ears are, is ultimately that...one person's experience. About as useful as listing to gear on posted You Tube videos. The best thing the reviews here can do is point us towards equipment that measures well and has no discernible design or UI flaws. After that...get the item and see if you like it, how it sounds TO YOU compared to other equipment you have on hand.

This is why all the op-amp swapping makes no sense...you can't measure the difference beyond a scintilla of a gnat's wing, and what any one person thinks they hear regarding such a minute change has absolutely no bearing on what I'M going to hear...

 

[staticV3]

Audio electronics and loudspeakers are tools with well defined performance goals and emotionally driven instrumentation like ears should be kept far away from the research and development process.

Whatever Nelson Pass, Franco Serblin, or Frank Zappa thought they heard in a piece of equipment has zero value to me, or to the equipment's functionality as a tool for audio reproduction.

This is common sense in every other industry, only in HiFi do people stubbornly hold onto this romantic delusion.

We don't develop cinema projectors by eye, combustion engines by smell, or water filtration systems by taste.

That would be absurd, just like developing a speaker or Amplifier by ear.

 

[M.C. Logan]

There's no listening test, for electronics, *because* they are subjective and inherently flawed.

Plenty of interesting reading around soundstage on this site: tldr: its not affected by the electronics, its an artefact encoded in the source and modified by speakers in a room and further modified by our perception.

Unless the electronics are *not* transparent perhaps? At which point it might be interesting to determine what particular non-transparent features/artifacts are affecting the perception of 'soundstage', no?

 

[JeremyFife]

Unless the electronics are *not* transparent perhaps? At which point it might be interesting to determine what particular non-transparent features/artifacts are affecting the perception of 'soundstage', no?

Why would you want kit that was so fundamentally flawed (broken) that it audibly changed the signal? There's no need for that today.
Tubes, and odd personal preference excepted.

I joined ASR after my new amp was reviewed here and found to be objectively poor (probably inaudibly, but really not good). This was a wake-up call for me. I changed the amp, binned my old upgrade plan, stayed on ASR and I've saved a fortune while massively improving my sound.

Drop the 'perhaps' thinking, realise what the measurements actually say, learn from the experts here (not me, the real experts) ... and enjoy music

 

[M.C. Logan]

Speakers and headphones are not solved.

No transducers are, nor will they probably ever be.

 

[oleg87]

Music is an artform. The art disappears once the music is committed to some signal with the dimensions of amplitude and time (until it is recreated once again by the transducer wiggling our ear drums). From there, it enters the discipline of engineering.

There is not some profound distinction between an Audio Precision box and a professional recording interface. If music can be recorded with electronics, music can be measured with electronics.

(But it most definitely *cannot* be measured with human hearing, which cannot make the faintest pretense to repeatability, objectivity, etc without *rigorous* controls, hence the proliferation of absolutely absurd, impossible, and comical beliefs among audiophiles who put a completely unmerited degree of faith into human perception. Indeed I find it extremely improbable that a serious engineer would spend a career in audio and not catch on to this reality, but clearly some have found it more profitable to embrace the silly side of this hobby and posit themselves as golden-eared gurus.)

 

[VientoB]

Music is an artform. The art disappears once the music is committed to some signal with the dimensions of amplitude and time (until it is recreated once again by the transducer wiggling our ear drums). From there, it enters the discipline of engineering.

It's an artform manufactured using signals in the form of sounds which are translated into electrical signals in the case of recorded music.

 

[RickS]

Latest troll rampage through this thread has been eradicated.

Many of you put up a good fight so was reluctant to delete the responses, but his last post clearly showed he was just here to spew subjective blather rather than contribute productively and is best just to not be reminded.