I actually like that site and I like Paul’s videos. The antagonism is really getting old. And on top of your Snark, you had to call them children? That’s pretty revealing.
All that matters is that you enjoy listening to your system Other people's opinions don't matter, nor do their measurements. Some people need reassurances that the fortune they've spent hasn't been wasted.
"If it measures good and sounds bad, -- it is bad. If it sounds good and measures bad, -- you've measured the wrong thing." - Daniel R. von Recklinghausen, HH Scott
As an engineer,indeed this gotta be your philosophy. You need to strive towards making all audible differences also measurable. And many things are measurable with much higher precision than any ear can judge. Great engineers do many times more measurements and more advanced measurements than what you find in marketing materials. For example, amps can sound very different when dealing with spikes in the signal getting at or close to clipping level. Tubes are often preferred for their more smooth and soft clipping behavior doing even harmonics distortion rather than how transistor amps often clip the signal like a scissor. The differences can easily be seen on oscilloscope measurements but you can’t translate them into marketing numbers. If an amp engineer decides to improve the sonic behavior at clipping level using a transistor circuit, you might enjoy it to sound better but you won’t find it impacting the sales measurement numbers.
@@l.s.1709 What is wrong is that you are playing very loud (concert level) where the peaks hit the clipping level. With high quality music preserving the original dynamics, those peaks can be quite louder than the average. Another benefit is that soft clipping helps protect your tweeters from burning when your amp is clipping. Soft clipping should not be audible at normal listening levels of course. The point here is that you can have two amps that have the same standard specs on a spec sheet and the one with soft clipping will sound better at max volume. Another factor for better audio fidelity at louder volume is how well the power supply can maintain the voltage e.g. using more capacitance to support heavy sub bass content. This is also not visible from typical amp data.
@@ThinkingBetter that's not a response to my question. No one at home is playing music at concert level loudness unless they're crazy and even then their hearing will be f-ed up and fidelity is not an issue anymore. Furthermore, all audiophiles will have invested in amplifiers that have beyond ample headroom (in relation to the sensitivity of their speakers) so clipping is a complete non-issue.
Great answer from Paul... When you know what to look for, a spectrum analyzer can give you a good hint about what to expect from a component, but very few have that knowledge. Good audio designers probably have their opinion and know what they want to see, but there are probably not two designers that totally agree on how the result from a spectrum analyzer should look like. There is no "best" and ultimate goal to meet. One lesson learned is that how the harmonics of distortion are configured is much more important than the total level of distortion. If you can make a design with a distortion pattern of the harmonics that fairly closely mimics the detection levels of you ear you can accept several percent of distortion and still think it sounds fantastic. But if you have 12dB too high 7th harmonic distortion compared to the rest of the distortion harmonics it totally ruins it for you. It screams brightness and fatigue. Cross over distortion from the output stage can create very odd harmonic distortion patterns and if badly designed a lot of higher harmonics. Class A biasing will help to reduce higher harmonics, but it also can be too much of a good thing where there are too little of higher harmonics and mainly 2nd and 3rd harmonics left. So in my point of view, finding the right amount of bias is essential to get the best sound from an audio product. A high bias class AB design can absolutely sound better than a full class A design just because it is better at following the detection levels we have for different harmonics of distortion. Personally I detect 2nd harmonic distortion at 0,6% and 3rd at 0,25% and I have found that it is essential to keep that ~9dB of difference at listening level if I should find total pleasure in a product. Please read what Nelson Pass wrote about distortion many years ago and understand that distortion can create distortion and even low numbers of measured distortion with sine waves says very little about distortion during playback. It can be 10-100 fold in amplitude depending on intensity and complexity. So that 0,01% can suddenly be 1% and then it's important that it's the right configuration of harmonics in the distortion. For me it's almost impossible to listen to a system where 3rd harmonic is higher than 2nd. It sound too clean and bright. Too much 2nd over 3rd will sound too soft and with too little edge on transients. But when I hit a 2nd that is 6-12dB higher than 3rd and the rest of the higher harmonics are at a low level I usually find it to sound perfectly neutral and "live". So, do I have the knowledge I wrote about in my first line? Can't say, probably not, but it's my truth based on my experience of designing DIY audio products on a hobbyist basis for +40 years.
I recently turned 70. Back in the day when I was a teenager every one I knew that was interested in a stereo only looked for one spec, THD. None of us knew what it meant but we had to have the lowest number😄
I’ve quoted these brilliant words from Herb Rechert many times, which goes something like, “We can hear most of what we’re able to measure, but we can’t necessarily measure everything we hear”.
Although, best audiophiles ears didn't hear what was going on whith MQA, or with digital delay and masters already used in vinyls around the eighties, or the «loud» high frequency noise of some Hi-Res tracks. They didn't even hear high, strident speakers, extreme reverb times, loud standing waves resonances, or the lack of baffle step compensation or crossover problems in highly regarded speakers. Easy to measure, hear and feel, except for some audiophiles.
Decades ago Consumer Reports magazine rated HiFi speakers. Their reported measurements included frequency response in an anechoic chamber. It was a useful guide to the “potential” of a speaker. If it had large roll off at both ends of the audio spectrum, or severe peaks and valleys, you knew it would likely sound awful in any room.
Ah, the infamous "suck meter". It does exist; they're called "ears". Thanks Paul for the clarification that hundreds of parts compared to hundreds of parts all out of the same parts bins can sound different. I am used to manufacturing tolerances that anything manufactured will have and why "exactness" between two devices would come at enormous cost with little benefit.
Of course! Take the exemple if two cars, same motor, same fuel, same cm³ or same inch³, same manufacturesr, same model, same version. If one of rhe cars has 100kW of power, it is not true all samples also have 100kW. Is is far more probable one has 102.7kW, other has 99.92kW, other has 103.1kW. Top speed can also be slightly different...
That differences are too small. The ear cannot detect them. If they were enough different, Paul would match each pair of power amplifier monoblocks for the same customer. And the integrated amplifiers would have L and R power modules inside each box carefuly mached, choosing close enough pairs in each run of production. I think nobody does that thing because it is not meaningfull at all...
If two of the same amps will not necessarily sound alike, what is the point of auditioning say an amp that is on a dealer’s shelf and then buying the same type of amp pulled from the stock room?
It beats having never listened to the amp, and purchasing it blindly. Our host went a bit too far, in my opinion, stating that two of the same amps will sound different. Technically, he is correct. But no one is going to actually hear any differences. Expounding on his example: Your two speakers will not sound exactly the same. But can anyone actually hear any difference between two speakers from the same manufacturer that are the same model? If no two speakers sound exactly the same, then why audition speakers before you purchase them? So listening to an amp, or any components, will tell you 99.99% of exactly how that component will sound. If you covertly swapped both of Paul's monoblock amps, with each other (move the left one to the right, and moved the right one to the left), Paul would never notice any change in the sound of his stereo. But technically, there would be a change; one that our ears would not notice.
Nothing you hear in a showroom will sound exactly the same in your home. The room has much more influence on the outcome than the technical components.
@@NoEgg4u You have component tolerances that will yield some differences and it could be that the left and right channels sound different. This could easily be true with using tubes. But such differences are often easier to find through measurements than listening. One amp channel might have an SNR of 110dB in the left channel and 106dB on the right channel but nobody can hear it.
In my younger days I was absolutely obsessed with specs. To the point I wouldn't even bother giving a second look or listen to anything that didn't fit the "number" I thought was important. I wasted alot of money and more importantly alot of time on gear that looked great on paper. One of the few regrets in my life. I'm 44 now and am completely satisfied with my three systems, none of which contain any components I would have considered 20-25 years ago, simply due to certain specs I would have considered mediocre
No offence, but when you did an Audiometry last time? I just wonder how good your ears are? They are teally CAN hear whole 20Hz to 20kHz spectre with an absolute linearity and sensitivity through whole range? Not much people all around the globe can boast of such a Golden Ears ^_^
@@Ashen2501 Does it matter if you hear up to 10kHz or 20kHz? Not at all, because all still have adopted their way of listening adjusted to their limitations. Just as an experiment, put an earplug in your left ear and go to an audio shop and listen on new products. You still can tell if you think a product sounds good or not. It's easier with two ears, but you still can make a judgement with only one. Human hearing have a fantastic ability to adopt to the circumstances.
@@flex-cx9bi Yeah, sure.... It doesn't matter if one your eyes have a perfect sight while another (or both) have myopia or hyperopia - if you're experienced enough you will see everything perfectly clear on any distance just like a person with all perfect sight. Right? Correct answer is NO! As an experiment me and my colleagues took an exercise, where we listened to a different music, using different headphones and speakers. Those, who weren't able to hear above 14 kHz failed in a blind test to determine if it were a CD, MP3 or high-res music. Barrier of 17 kHz weeded out difference between a CD or high-res (96/192 kHz and 24 bits). Result is - if you have a hear loss, there's no need to spend kilotones of money on overexpensive hardware. You'll get same (or better!) quality for a much affordable prices, while leftovers could be spend on something truly necessary for your life, which will increase it's quality ^_^.
@@Ashen2501 That is complete BS. Differences between CD, MP3 or other high-res music like FLAC of Tidal or Qobuz lies in the way they compress harmonics and amplitude. They have different algorithms of removing audio signal information to create smaller file sizes that don't need so much resourses to stream from their platforms. You do not know what you are talking about at all because you do not understand how compression is done. The compression is performed from DC to what ever upper limit you select like a 22kHz brick filter of CD or 96kHz. It's not a limitation of high frequencies. The better system you have the easier it is to hear the sonic differences between a CD or a compressed MP3 file. It has nothing to do with the audio information at treble only. Your statement is truly false and if you didn't hear any difference it's more likely that you system wasn't good enough to let you hear them.
@@flex-cx9bi Your statement is based on marketing - buy a multimillion system and you WILL hear a difference between a high-res lossless codec and lossy codec. My statement is based on science - if you physically CAN'T hear high frequencies and both of your ears have no absolute symmetry in linearity and sensitivity on a whole range even a billion worth system won't let you hear a difference between MP3 and even a CD during a true blind test. It's called a science, peach ^_^
Lots of measurements are done also at PS audio and are needed. Frequency response, decay, distortion , impedance, and dispersion on the Aspen speakers are important measurements. On the amps you go great lengths to ensure the power rails are clean from noise and stable when a power amp has peaks. And even THD, TIM and frequency response for amps are usually not determining for a good sounding amp, you still measure them. It’s easy to discredit measurements but they are extremely important also together with listening tests that can reveal things measurements don’t capture. PS audio makes awesome products because products are measured and listened to for best possible perceivable performance. Btw, I’ve also many times seen people say something sounds great and then when you do a measurement you realize it’s not great under certain conditions. Perhaps the music you used for the listening tests didn’t have particular content in it that revealed a measurable problem and some other music would have revealed it. This is particularly true for speakers.
@@nirodha35 Check what is a speaker waterfall measurement. You can measure and graphically represent how the sounds at different frequencies persist in the time domain (x=frequency, y=level, z=time).
@@nirodha35 Decay in loudspeaker terms is most likely energy storage because of self resonances in the cabinet. With an accelerometer connected to the cabinet and running a sinus sweep you can pick up where these resonant behaviour occurs. Usually presented in a water fall measurement. Look at any measurement done by Stereophile. In most speakers you see ridges that have longer decay because of resonances. That is what you measure and try to avoid since it's basically timesmearing of signal and something that will create colouration.
I think it is like trying to measure a dish a chef made. You can analyze all you want with instruments, but the stats can't tell you how it delicious it is, even assuming that EVERYONE has the same taste profile.
If two amplifiers of the same design sounded substantially different from each other then stereo imaging would be be a total mess. After all, EVERY stereo amplifier has two amps of the same design. It can happen but reputable manufacturers usually have better quality control than that. To suggest that you can always hear the differences between ostensibly identical amplifiers is a pure fabrication. Paul steadfastly refuses to use double blind A/B listening tests to support his proclamations.
There's one product recently that has led me to add a footnote -- the Theoretica Bacch system, which is arguably 100% measurements based, with its basic DAC vs DACs well beyond my budget. It may be my next upgrade, but needing to set up a Mac mini with a webcam is a bit cumbersome and messy looking in a living space, so I am still on the fence. If it were more packaged, I would jump.
I’m confused, are you saying that each of the same model of a component made by PS will sound different? Or any other company by that matter. If so, how could you use as a guide any product review? I’m a customer and own a Sprout. Quite pleased.
Literally, after years and years into that hobby, I can clearly state. Different amps (even of the same topology) sound different. Speakers for sure as they interact with your room. A pure measurement parcous will not tell you if it is a good sounding component to your ear. I have a strong background in the test & measurement industry, having handled linearity figures of +/- 0.1 dBm and better and accuracy measurements in the ppm realm. It is essential for certain types of equipment. In the audioworld, a listener would seriously not detect a 0.5dB non linearity for instance, in a speaker design. Listening to music as a whole and not frequency sweeps, multitone distortions with an analyzer, total flatness, impulse response etc, is so different. A well measuring component needs to fulfill, of course, certain standards. A speaker with too agressive peaks towards the end of spectrum might sound ear piercing and vice versa. And amp not handling a speaker well (or totally load mismatch) will falter and sound stressed. If well chosen (the tech specs are only the beginning of this journey) basic satisfaction is guaranteed. But to really enjoy it deeply, one must first listen to their choice. Best in their home if possible. Regarding 2 same amps measuring different..... Absolutely normal even in the T&M world. Even after a full calibration run 2 of the same high accuracy DMMs will have a difference of perhaps 0.001%. And for audio. You will never hear a difference between a THD of 0.05 or 0.005%. Full stop. Our ears are not measurement sensors. One day you hear better than the other day. More important is you enjoy the music thru your system and do not care about the last decimal points after the comma.
It's like with any of our senses, when we're seeing something, it's not just our sight, the brain kicks in and connects what it's seeing with context of the past and expectations of the future, all that then connects with our emotional impact, all that equates to sight. Hearing is no different, hearing is getting the sound and then interpreting it based on what it reminds us of in the past, or what's happening at that specific moment, and then the emotions that equates to. Measurements are very 2 dimensional, senses are far beyond anything we currently understand at a scientific level.
So just from your opening statements if I buy two monoblocks from you they are not going to sound exactly the same. Which is exactly what I would want to monoblocks to do: sound exactly the same.
I would hope two of the same product/model from the same manufacturer would sounds identical, or at least have no perceivable difference in sound. Otherwise, there would be a detriment to using monoblocs.
If a specific driver is used and the sonic reproduction is recorded from 1 and it could extend above 100 percent from ringing or over reproduction would give a more precise estimate.
1. Listening test, hear things beyond measurement. 2. Measurement, measure audio way beyond any human ability to sense. Never the two will meet. The confusion is that something that measures well, may not be the coloration that someone prefers. Individual static measurements came about decades ago, as it was all we could do then. A SysID test would bridge the gap, but no one is doing it in audio. Ie. simultaneous acquisition of THD, multi tone IMD, compression, PSU effects ect under all conditions and frequency. As for two amps measuring the same, but sound different, well you didn't aquire enough / the right measurement.
Measurements are not objective (as the setup is subjective). If you use the same measurement setup for different products, the measurements are internally consistent. As to Amir's "objective" measurements: you tell me how a sine wave translates to music quality.
I'm watching the episode whilst timing. Specifications mean nothing V sonic capability. How the equipment sounds depends on the carbon based lifeform listening.
Until I find a better philosophy I say after a unit has achieved an SNR of 110db, a THD+N of 0.0001%, (making for) a SINAD of (roughly) 100db, a channelnseparation of 90db those specs don't matter anymore; and I look for things like valves, discrete topology, analogue circuits, galvanic or optical separation, seperated power supply, linear or even battery power supply, other forms of separation such as dual mono design, clock speeds, filtering... or anythingbthat can be clearly heard as superior over the YT audio compression.
I was going to mention the scope on the power plant as a prime example of the BS he is talking about, but he did it for me. You literally cannot hear what the thing does so they put the scope on to show it’s “doing something” but he admits it’s meaningless. I guess kudos for back handed honesty.
Hala freaking luuuuyaa Measurements on an engineering level is designed to create a slew of products as close to each other as possible, in the lab you can’t randomly cook a meal and put different ingredients in it and expect the same meal. Happy days, what about for the end user, it’s a rough guide. It will give you a rough indicator of what is going to happen, but it does not take into consideration the frequency response of your own personal hearing, your preferences, or what your plan to synergise with it in the rest of the system. Online you see bloodbath conversations expecting a product to be better than another product just because especially these days there has been slight manipulation of higher measurements to stand out, one of the quickest arguments I see from group such as ASR is a topping DAC is better than a chord Dave This narrative gets exacerbated unfortunately by echo chamber groups, specially those who have very little experience with equipment components, and rely on nothing but literature Akin to the guy at school you knew who could recite the Kama Sutra in 10 different ways and had never ever had a girlfriend Excellent video, and yes, use it as a guide but let your hearing be the judge because you will only look at these measurements once but hopefully you will be listening to your system for years
Should be measuring the pleasantries. There are ranges, and shapes of signals that do show up on a scope. You can use a very sensitive scope and compare to a cheapie, see how much harshness comes through. Would be nice for the industry to get back to pleasant audio instead of bright and harsh stuff, because “it measures good.” My comment is in general of the industry, nothing specific to a brand. Cable companies are the worst with this.
Messurements are often not that honest anyway. Atleast for watts. Many messure peak at 1 channel driven at 1kHz and 1-10% THD instead of actually continuous power to all channels (20-20.000Hz and below 0,5% THD. Alot of watts is nothing without amps behind it. Then It's just volts. Same as an engine.. Doesn't help with alot of horsepower if there is no torqe.
All my hard-nosed measurement-audiophile friends keep giving me a hard time for having spent a king's ransom to have all my rig's premium audiophile cabling-which is already outrageously expensive as it is-shipped around the world and back again, making a full lap around the globe and thus becoming Lenz law-neutral in respect to the earth's magnetic field, giving me thus a slightly wider sound stage and more relaxed mid-highs... They keep saying they can't measure any difference in the output signal, but I have to remind them that my picky, sensitive audiophile ears cannot be reduced to the sterile regularity of mathematical equations; you have to put your entire soul into the listening to appreciate the difference, I say. Then they claim I'm full of s- and start challenging me to an A/B showdown. (" -.-)
The funny thing is that even if shiny cables and cool looking dacs won't change the sound in reality, you could still hear a difference. And if those cables helps your brain to fool you into thinking that now your system sounds incredible, then you might have gotten your money's worth. There is also the possibility to just rely in the measurments and accept that your sound memory and hearing is not very trustworthy and save loads of cash, but that might be a hard thing to accept for some folk.
I get what Paul is saying and agree, but perhaps beyond the scope of the question of published data or precisely quantify, we can measure and know the difference of some things. Drivers/speakers could be measured for harmonics and people typically prefer either odd or even (but I guess it still falls under "educated guessing" 😂). I've also seen and heard amplifiers that was measured and refined, but again I'm not sure if it's possible to quantify the measurements even though it's visible and audible (like Paul said).
Totally agree with you Paul. There are way too many factors that we don’t even realize yet that contribute to the given components even if they have identical parts and circuit design. Maybe one day will be able to figure these differences out.
Two amps with identical components will sound and perform the same, the ones that don't, are companies not measuring basic specs of their devices. PS Audio is one of those companies.
@@PetraKann And releasing the (same) model with changed components later on in device life cycle. Example would be Samsung releasing the same phone, but depending the region, you get different cpu.
It's really good to hear you say that "we don't have the measurements." There are too many people that delude themselves and others about the sanctity of measurements. I think measuring what makes good sound is no less complex than quantifying consciousness
@@l.s.1709 Just like Paul said. If there were measurements for "sonic qualities" it would be a holy grail. Perception of sound is subjective to the individual. Any textbook on sense and perception will lay it out. There is a good book about the word quality called "Zen and the Art of Motorcycle Maintenance" you should check it out
@@PlaybackMansion you're just repeating Paul's point, but badly. The sonic qualities are in the creation: the writer, artist, producers, engineer, mixer are creating the sonic quality and the aesthetic subjective part: this is indeed unmeasurable and subjective. However, the hifi system should only reproduce what was heard at the mixing console and what the artist and the team decided to bring to the world. You're romanticising audio playback instead of music creation.
Measurements of what? Distortion and other problems are practically non existant in todays equipment yet it all sounds different. Measurements for measurements sake?
Oh boy what is happening here? Marketing blablabla.....this channel is kind of snake oil factory. Funny thing is if they do not know what makes the difference, how can they produce products that are best for their customers? I think PS Audio products are good, no question, but take the word of this gentlemen with a grain of salt.
