I've seen that movie.
-
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
Okay, so be it, but it's damn embarrassing.
It is what it is... Cash grab.Okay, so be it, but it's damn embarrassing.
Meanwhile, lots of people (including me) are using power- or integrated amps based on Hypex or Purifi modules, the younger versions of whose are rightfully considered state of the art - with NE5532 input buffers that aren't socketed.
A comparable "rolling" scene or craze like on the cheaper chip-based amplifiers just doesn't exist, simply because SMD soldering isn't a widespread skill. And yet, close to nobody, as in approaching zero percent, is "missing" something. The amps amplify, and excellently so in all regards that matter, and that's it. Big badaboom and great enjoyment listening to your favourite music. Psychologically and objectively, the amplification problem is solved.
Objective data aside for a moment, I really think it's nothing but pure psychology. You buy a cheap class D chip amp for 100 or a mono pair for 200 moneys, and then feel like "missing out", because the whole "rolling" thing has been heavily advertised and "is a thing" and easy to do and over 9000 uToob videos tell you it's totally worth it to invest in sparkos whatevers that cost more per pair than your whole amp does. You do it because the low initial price tricks you into thinking "it's good but could be better". Then in the end you spent a total amount that is already substantial - hooray sunk cost fallacy. Et voilà, suddenly you arrived at a satisfying sound.
While if you spent significantly more in the first place, the usual 500-1000 for Ncore amps for example, you're much less likely to fall for that, because you KNOW it's excellent, without any load dependency flaws for example. The price you paid thoroughly agrees, if you know what I mean. And as said, replacing opamps is difficult and requires handy skills 99% of users don't have, so it's "not a thing" in the first place.
A comparable "rolling" scene or craze like on the cheaper chip-based amplifiers just doesn't exist, simply because SMD soldering isn't a widespread skill. And yet, close to nobody, as in approaching zero percent, is "missing" something. The amps amplify, and excellently so in all regards that matter, and that's it. Big badaboom and great enjoyment listening to your favourite music. Psychologically and objectively, the amplification problem is solved.
Objective data aside for a moment, I really think it's nothing but pure psychology. You buy a cheap class D chip amp for 100 or a mono pair for 200 moneys, and then feel like "missing out", because the whole "rolling" thing has been heavily advertised and "is a thing" and easy to do and over 9000 uToob videos tell you it's totally worth it to invest in sparkos whatevers that cost more per pair than your whole amp does. You do it because the low initial price tricks you into thinking "it's good but could be better". Then in the end you spent a total amount that is already substantial - hooray sunk cost fallacy. Et voilà, suddenly you arrived at a satisfying sound.
While if you spent significantly more in the first place, the usual 500-1000 for Ncore amps for example, you're much less likely to fall for that, because you KNOW it's excellent, without any load dependency flaws for example. The price you paid thoroughly agrees, if you know what I mean. And as said, replacing opamps is difficult and requires handy skills 99% of users don't have, so it's "not a thing" in the first place.
Last edited:
If op amp A and B fit in device X technically, if both A and B work as they should, then only the imagination factor remains.
You linked that distortion test in #131. Good. It can give an indication of how much may be needed. IF we hear distortion, it is most likely only when the amplifier is driven into clipping and with bad/mediocre headphones and/or speakers.
What we can really hear, however, are differences in FR. I mean swings of around 10 dB to 15 dB in the bass area (depending on the impact of the room).
You hear that, no imagination there. Even smaller deviations than that. To get an idea of that, you can try this:
I know you know what I just said and you've probably already taken the test I'm linking to now. It's more of a tip for others reading the thread.
Blind Testing a 1 dB Level Difference
Discover the smallest difference in level (dB) you can reliably hear.
www.audiocheck.net
I put this in as a bonus, for those who are curious to try:
Blind testing a 24 dB Dynamic Range
Checking your listening room's noisefloor, online.
www.audiocheck.net
Replacing bjt input OAs with fet input OAs can cause problems in the circuit. These rollers forget the OA is part of a circuit that should be optimized for the specific OA. The circuit is responsible for stability, a major factor and there are others including rail noise/ripple/rf, load (some OAs handle reactive loads better), offset voltage, etc.If we start from the premise that there really was an audible and not imagined difference, then it cannot be due to differences in distortion, as pointed out by others.
Does that OPA2604 fit into the Aiyima 07? I can guess (I'm no electrical engineer) that it should be possible to read from the technical data under which conditions the OPA2604 operates and then compare it with the NE5532.
OPA2604 Datasheet, PDF
OPA2604 Datasheet. Part #: OPA2604. Datasheet: 435Kb/26P. Manufacturer: Texas Instruments. Description: Dual FET-Input, Low-Distortion Operational Amplifier. 22 Results. Datasheet: 214Kb/12P. Manufacturer: Burr-Brown (TI).www.alldatasheet.com
NE5532 Datasheet, PDF
NE5532 Datasheet. Part #: NE5532. Datasheet: 125Kb/12P. Manufacturer: NXP Semiconductors. Description: Internally-compensated dual low noise operational amplifier. 169 Results. Datasheet: 69Kb/10P. Manufacturer: Texas Instruments.www.alldatasheet.com
Or there was something wrong with your OPA2604 if you heard clear differences. That on the condition that we can rule out (difficult in practice) the imagination factor.
And all of your music recorded before 2000 has already been thru a dozen of these. It must be really, really muddy and fat.Holy shit, this is what Aiyima themselves have to say about what it sounds like with different op amps in their A04 & A07!
NE5532: The resolution is average, the high frequency is relatively dry, and the low frequency is relatively muddy and fat.
It begs the question, if you have to design the circuit to be safe with a vide variety of OA how optimal is it ? Would it just not be even better if the designer optimised around a given set of components ?
I suspect they make the product worse so the end user can faff around with it and feel satisfied that way ?
And of course , some overenthusiastic fellow riding the peak off the dunning Kruger curve would not stick to the approved list of suitable OP amps :/
And make a mess, because he read something wonderful about the musical properties of some OP amp ?
Arguably well into the 2000s. Every single channel (as in instrument) has run through half a dozen NE5532 or similarly cheap equivalents. Plus several more for groups and summing stages. Plus various periphery like mic preamps, equalisers, compressors, and whatever gear the producer chose to use. Fully digital production (as in sound generation and recording and processing) only started in the mid to late 2000s. And only in electronic music styles you can probably expect no ne5532 in the signal path today - without insurance. It's literally everywhere and has been for decades.
And nobody on the producing side ever gave a damn. Unlike the usual audiophool suspects, for whom that tried and tested chip is somehow not good enough for mere listening.
audio_tony
Major Contributor
The reality is that many opamps (particularly those intended for audio usage) will work quite safely in a comptently designed circuit.
There are a few exceptions where instability could be an issue, and it is this instability that is often perceived as a difference in sound, because the circuit is oscillating (but not all the time, the oscillations can be triggered by differing amplitudes or frequencies - so observing the output with no signal will show no obvious issues).
It's not difficult to design a circuit that is both 'safe' and optimal, the two can go hand in hand. The problem is usually caused when excessive cost cutting measures are implemented, causing the elimination of power supply decoupling capacitors for example.
Last edited:
Interesting, but I would not dare anyway before I aquire the skills to design the whole circuit and pcb to begin with ? I can not run simulations either ? I can and do some engineering but not this . I know enough to not do it .
I wonder what makes random people think they can second guess a competent designer without being one of them ?
A wonderful technology laden hobby for the technical illiterate
with points for participation if you go to other forums than ASR ….
Via the Wikipedia link about NE5532 I found something that Douglas Self said. I looked up and in hisArguably well into the 2000s. Every single channel (as in instrument) has run through half a dozen NE5532 or similarly cheap equivalents. Plus several more for groups and summing stages. Plus various periphery like mic preamps, equalisers, compressors, and whatever gear the producer chose to use. Fully digital production (as in sound generation and recording and processing) only started in the mid to late 2000s. And only in electronic music styles you can probably expect no ne5532 in the signal path today - without insurance. It's literally everywhere and has been for decades.
And nobody on the producing side ever gave a damn. Unlike the usual audiophool suspects, for whom that tried and tested chip is somehow not good enough for mere listening.
Self, Douglas (2010). Small Signal Audio Design
.. and on page 117 he writes:
With horrible inevitability, the very popularity and excellent technical performance of the
5532 has led to it being criticized by subjectivists who have contrived to convince themselves
that they can tell op-amps apart by listening to music played through them. This always makes
me laugh, because there is probably no music on the planet that has not passed through
a hundred or more 5532s on its way to the consumer.
For those who don't know Douglas Self. He knows what he's talking about:
Douglas Self is a British electronics engineer and author with a particular interest in audio. He received a first class honours degree in engineering from Cambridge University, and then studied psychoacoustics at Sussex University.[1] He is the author of six books on audio electronics, published by Focal Press.[2] He has also contributed many articles to Wireless World magazine, some of which were compiled into a book along with articles by Peter Baxandall.
[3][4][5]
Self's books have been well received. His Audio Power Amplifier Design Handbook was recommended by Walt Jung and described as "famous" by audio website hifisonix.[9][10] The second edition of his Small Signal Audio Design received a very positive review in Sound on Sound magazine.[11]
Douglas Self - Wikipedia
That alone is worth applause. It would seem knowing your own limits mysteriously mutated into a rare skill these days.
Some OAs will work fine in a circuit without decoupling so they figure why include them. Then a roller puts one in thats not and things go sideways.
I think they want to count being an audiophile for a long time and owned lots of expensive equipment as experience
Optimal for all OAs? Try using a unity gain unstable OA in a buffer circuit. Try using a bjt input OA in a guitar input circuit that was designed for a fet input.
audio_tony
Major Contributor
Which is why I placed emphasis on a comptently designed circuit - as no competent designer would omit even the most rudimentary decoupling.
Yes, I should have added a caveat for opamps that are not unity gain stable. I have built many, many test circuits with a moderate gain of around x3 to x5 and subsituted many different types of opamps with no issue at all, simply because I provided adequate decoupling.
Obviously with some very wideband opamps, instability can be an issue - but then most wideband opamps are not typically intended for audio use.
Last edited:
Audionaut
Major Contributor
Yes, but let's stick with ASR and say that what has different values and measurements can usually be heard.
These are OP amps, as they are often changed with the TPA3255 and which sometimes have very different values.
I have not yet swapped an OPA, but I can well imagine that different OP amps in the appropriate topology can also be audibly perceived differently.
Excuse my bewilderment, but WHAT THE FUCK is this? My old timey electronics degree (probably close to educated layman level nowadays) says this is utter bullshit and defeats the whole purpose of any proper opamp circuit.
- Joined
- Dec 12, 2019
- Messages
- 16,458
- Likes
- 37,357
hmmm... are you sure about that?
Rando internet images above, all.
Original copies of all of the above ('cept the Coltrane) are upstairs here, though.
Similar threads
- Locked
- Poll
