Last edited:
-
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!
Does Op-amp Rolling Work? [Video]
- Thread starter amirm
- Start date
antcollinet
Grand Contributor
Based on the analysis of the recordings having differences that no op amp swap can create there are two possibilities.Just yet another example of his backwards understanding of science.
- He is either incompetent,
- Or he is a fraud.
I know which one I'm putting my money on.
During the live, he was asked a question re: Amir's review of the XLS Encore, and claimed that you tried to build it in a way that it would measure as poorly as possible.Just yet another example of his backwards understanding of science.
He mentioned that you purposely didn't use No Rez and some other insulation that he recommends for the build, and that you didn't round off the corners to prevent diffraction. But that despite your best efforts to sabotage his product, that it measured great anyway so Amir had to begrudgingly give it a recommendation.
During the live, he was asked a question re: Amir's review of the XLS Encore, and claimed that you tried to build it in a way that it would measure as poorly as possible.
He mentioned that you purposely didn't use No Rez and some other insulation that he recommends for the build, and that you didn't round off the corners to prevent diffraction. But that despite your best efforts to sabotage his product, that it measured great anyway so Amir had to begrudgingly give it a recommendation.
Yep, a bit off topic here but have heard this delusion before so I refer to this in the build thread…
Build of gr-research X-LS Encore
As I wrap up this build, I did complete my validation measurements. Here is the on-axis frequency response, 75db, gated at 1m: Interestingly, I also measured after removing the damping, the response was almost identical. So, the impedance blip ugliness does not show up on a simple REW...
www.audiosciencereview.com
Danny also said he would send his own build for testing but he backed out. This is when we learned he only sends samples if he can control the review conditions. Despite my recommendations (and almost 5 years later), his cabinet drawings still do not call for baffle rounding and the damping material is not part of the website ordering options. But that is not his fault.
P.S. One of the reasons I like email is that it gives me an audit trail if someone has memory recollection challenges. Here is Danny's response from 2020 when I asked about the baffle rounding. Note about halfway through that he admits the Encore design info will be "more clear and precise on the new website". I also included where, despite Amir's recommendation of the Encore speaker, he starts to hedge about listening conditions. He did not post or supply a response for me to post...
The instructions for the kit do not indicate whether the front baffle should get its edges rounded. What is your guidance?
Every now and then I get that question and I of coarse always respond by recommending that the edges be rounded over. Sharp edges will cause some edge diffraction issues as seen in amirm's measurements.
As an experienced builder though I would guess that you already knew this, but since it wasn't specified you did not round them over.
As per your reference to this and regarding damping material I will make it even more clear and precise on the new website.
You mentioned that I would be allowed to respond to what was posted there and I just have a few thoughts to share.
First I have a few questions though.
You mentioned a sound check and Amirm obviously also gave them a listen and was quite pleased with them.
Amirm laid out how they were measured and the whole measurement process in great detail.
But for listening evaluation nothing was mentioned about your system or his. Can you please show pictures of your listening room and gear as well has his listening room and gear that was used in that part of the evaluation?
And then I will put together a response that you are free to post. Or if you need me to I can join and post a response.
Thanks,
Danny
Last edited:
Hi Amir,
I have a question about your testing method. Does feedback cause the op amps to harmonise their output, would the feedback actually suppress the differences, and therefore if both channel Op amps were changed the feedback would be different and the results also different. I ask purely out of a lack of knowledge on my part. I’m sure that your original results are true.
I have a question about your testing method. Does feedback cause the op amps to harmonise their output, would the feedback actually suppress the differences, and therefore if both channel Op amps were changed the feedback would be different and the results also different. I ask purely out of a lack of knowledge on my part. I’m sure that your original results are true.
Opamps are specifically made to use feedback to control the transfer function. They will not function (as a practical matter) without it. So things like frequency response and gain are controlled by the external components, not by the opamp itself. Although things like open loop gain can vary from one device to another, it's invariably high enough to make no difference.
The left and right channels are independent.
antcollinet
Grand Contributor
The left and right channels are independent, no mixing, no feedback no nothing.
Basically. If each channel has a feedback resistance ratio of 100:1 (say), the gain will be set by that ratio (about 40dB in this case). Channel to channel, the only gain variation will be because of resistor tolerances, NOT the opamp.
I'll tell you something maybe a little surprising. There are differences between even the same model of op-amps coming from different brands. I have a circuit here that is stable without feedback bypass capacitors when using Rohm brand 4580 opamps, but is unstable and requires feedback bypassing by 100pF with Diodes Inc 4580 opamps.
DualTriode
Addicted to Fun and Learning
- Joined
- Oct 24, 2019
- Messages
- 961
- Likes
- 627
I have come to believe that both THD+N and SINAD are near worthless as indicators of perceived audio quality.
In this context, OP-Amp rolling, THD+N and SINAD are near worthless indicators of the perceived audio quality between op-amps.
Also scientific, measurements of PEAQ may be discriminating where THD+N and SINAD are not. Evidence of masking of low order harmonics fills the AES Library. High order harmonics are much more annoying even at seemingly low values. This is the stuff of rub and buzz and PEAQ measurements.
amirm you may need to upgrade you APx555 instrument.
Thanks DT
In this context, OP-Amp rolling, THD+N and SINAD are near worthless indicators of the perceived audio quality between op-amps.
Also scientific, measurements of PEAQ may be discriminating where THD+N and SINAD are not. Evidence of masking of low order harmonics fills the AES Library. High order harmonics are much more annoying even at seemingly low values. This is the stuff of rub and buzz and PEAQ measurements.
amirm you may need to upgrade you APx555 instrument.
Thanks DT
antcollinet
Grand Contributor
Let's put to one side the fact that beliefs are like opinions: Everyone has one, and they are nearly always smelly.
You are missing the point. This test doesn't rely on a qualitative assessment of different levels, or components of THD. The point here is that the harmonic distortion DOES NOT CHANGE. So there can similarly be no perception of a change in audio quality. Or at least no perception that is resulting from the sound reaching the listeners eardrums.
It is not surprising: it actually happens that op-amps bearing the same designation but coming from different manufacturers have slightly different specs or produce slightly different measurable end results in some application.
I also know of circuit unstable with ST Micro MC33078/9 but stable with the same reference from the original manufacturer: Motorola (now OnSemi). By the way, the original datasheet of the latter shows vastly different CMRR vs frequency than the same reference from the competition of the current Texas Instruments version of this chip.
I suspect it all boils down to different manufacturing processes or level of quality control or undocumented nuances in the inner implementation of the schematics.
Last edited:
OP
- Thread Starter
- #56
This data was presented:
Low order harmonics are identical. What is not, is easily masked or is below threshold of hearing.
DualTriode
Addicted to Fun and Learning
- Joined
- Oct 24, 2019
- Messages
- 961
- Likes
- 627
Let's put to one side the fact that beliefs are like opinions: Everyone has one, and they are nearly always smelly.
You are missing the point. This test doesn't rely on a qualitative assessment of different levels, or components of THD. The point here is that the harmonic distortion DOES NOT CHANGE. So there can similarly be no perception of a change in audio quality. Or at least no perception that is resulting from the sound reaching the listeners eardrums.
So what point have I missed?
Yours is a fallible argument.
With the thought in mind; a APx555 audio analyzer with a GRAS microphone is not a human ear. A PEAQ measurement made with that same APx555, GRAS microphone and a PEAQ algorithm likely has a better probability of determining differences among a hand full of Qp-Amps.
amirm, your plot above looks to be taken at the amplifier output. What is not shown is the driver output with the the sum and difference modulation side bands that may be that fuzzy sounding annoyance that you can not quite identify what you hearing. The fuzzy annoyance gets louder as you turn up the volume.
If the distortion was not audible at low SPL it very well may be as you turn up the volume.
JSmith
Master Contributor
DualTriode
Addicted to Fun and Learning
- Joined
- Oct 24, 2019
- Messages
- 961
- Likes
- 627
The Apx555 is what is on my bench.
The SINAD application in the APx500 software can use some help. As previously stated, a APx555 audio analyzer with a GRAS microphone is not a human ear. Upgrade as in get the add ons.
Audio Precision also has PEAQ software application(s) available.
PEAQ testing also shows in the APx500 software speaker Rub and Buzz testing applications.
PEAQ is used to evaluate an audio codec vs the original signal, what does that have to do with testing an amplifier?
Similar threads
