there exists no linear relationship between an objective quantity and the subjective quantity; the human frequency response is all over the map

@@johnholmes912 having any objective measure or correlation with the subjective experience would be a huge step forward. There must be people that understand what creates some of these experiences, but there doesn't seem to be any scientific understanding of these phenomena in the audiophile community

Bruno Putzeys states something similar in his article about "No such thing as too much feedback". The cause is indeed that too little NFB is used. linearaudio.net/sites/linearaudio.net/files/volume1bp.pdf

@@hidjedewitje Thanks, that's gold. The only time you can not have negative feedback is if the gain in your amp is very low - like maybe x 2-3.... as a guess. Once you introduce high open loop gains you must have error control. Error control = precision; you are performaing a mathematical function. No feedback means your amp is basically out of control. People who claim no feedback is best, must ignore the thousands of op amps that were involved in laying down the original recording - all of them used negative feedback to produce distortion free recordings. Its almost magical / religious thinking to have a negative view of negative feedback.

@@mikevincent6332 The open-loop gain needs to be high for high disturbance rejection. Low close loop gain is perfectly fine. I don't even see how one can make an amplifier with a OL gain of 2-3x. A typical transistor already has Hfe of up to ~400, but typically about 150 (though power transistors have less and the capacitances limit HF performance). It's also the case that a little amount of feedback does more harm than good. When you lower sensitivity in the lower frequencies, you increase sensitivity for high frequencies. That's just the bode sensitivity integral. If you use little amounts of feedback, you'll end up with more sensitivity to disturbances in the midrange where you are very sensitive for them. In that case feedback sounds bad. You just have to apply as much feedback as you can to push the sensitivity out of the audible bandwidth. Some of the best amplifiers have out there have about 60-70dB OL gain at 20kHz.

@@hidjedewitje I think the low voltage gain circuits must still have forms of negative feedback such as emitter degeneration resistors etc, which will of course knock back the gain of the circuit considerably and extend the linearity, but I would have to study a circuit to find examples (some of the Pass amps perhaps) this type of feedback is les opbvious than global feedback. I have designed an amp myself which seems to have great performance. it uses (gosh) 4 feedback loops and I think it is called a common-emitter, complimentary pair darlington output stage (collectors are tied together at the output) 275W RMS into 6R. I am not formally trained but simulated my ideas in spice before building. I use an OPA op amp to drive the whole thing. The o/p stage has a voltage gain of 6, its a very unusual design

You only hear what you want to hear and do not understand the main point of the video. Measurements and spec tell very little about how we perceive sound, so everything you preach about zero testing etc is pure BS .. Measurements and specs are good for telling if a device is working as intended, but largely useless for defining High-end. Crossovers SR , THD do "not exist" in a well-designed amplifier. But still, they sound very different for many reasons.

The owner of ASR was sitting in that presentation.

@@jonathannovick3766 Hmm. Then I wonder why the ASR guys and the SBAF guys seem to hate each other so much. The SBAF guys seem to take your approach, in that they use measurements to try to validate what they're hearing, whereas the ASR guys seem to be taking the, "If you can't measure it, it doesn't exist" approach. Does Amir just not agree with you?

@@BrianKern I thought Amir liked the presentation. I will say this though. I do believe that any difference you can hear is measurable. The only question is what the test conditions should be to quantify the difference. This is where many people give up. After all, it is hard to prove that finding the secret sauce will lead to greater financial success therefore companies don't fund the research.

@@jonathannovick3766 I recently joined the SBAF Forum and noticed they had a thread created entirely to critique Amir's approach. I don't understand what all the measurements mean well enough to know the difference and didn't understand what the fuss was about but after doing some research on both sites it seemed those two groups of people dislike each other very much. I find it a bit bizarre. They both like audio equipment and you'd think they just want to listen to equipment that makes their music sound better but they really seem to not enjoy each others presence at all. One says only measurements matter and the other says both matter. And for whatever reason that's enough for people to massively dislike each other and both of them take digs at the "other guys across the pond". I guess this is more of a post commenting on the strange behavior of humans. You'd think they'd let bygones be bygones and just agree to disagree. Oh well, I suppose. And I agree that anything that has a physical property can be measured. I liked your presentation a lot because you're pointing out the possible "confounding variables" that can mess up your results. Very useful to point out to the community and something I haven't seen discussed very much among people that do measurements.

Ummmm. The fellow asking the questions at the end (58 minute mark) is Amir. The owner of ASR. 😉

Perceptual tests are complicated and not always agreed upon. PESQ and POLQA testing is used a lot in telephony applications but I have seen these tests give very low scores with encrypted speech signals that actually sounded very clear. Applicability comes down to the types of problems being measured, the weighting applied to problem and the characteristics of the signal path being tested. PA systems are often tested for their intelligibility using STIPA testing that accounts for the acoustics of the environment. There is a perceptual music test called PEAQ that has been around for years but is not widely used. I am not familiar with its short-comings but the lack of adoption would be an indication that they exist.

+Corrado VRz (CorradoVR6z) Remember that the lossy audio compression used in online video could potentially magnify the distortion such that it becomes perceptible earlier in the video than in person. (Though I agree with you, I heard the distortion several seconds earlier, too.)

+Corrado VRz (CorradoVR6z) Its probably also easier with headphones :)

Clearly you didn't listen to the entire discussion; the exercise was to identify when the distortion became offensive, not audible.

I watched the entire video and distortion is offensive to me, it doesn't matter if it's a little bit or a lot, I don't want to hear any kind of distortion.

Well sorry to say you literally cannot listen to any reproduction of music then, as there is no system that is free of distortion. Again, you've completely missed the point of the exercise.

I think it was this guy audiosciencereview.com/forum/index.php?threads/master-index-for-audio-hardware-reviews.2079/

@@rollanb The guy in pink shirt?

Well at this point they have to make a difference, whether they make a difference or not!

FoxGhrocks perfect paws grooming ost7