|
Post by donjoe on Aug 9, 2016 20:50:38 GMT
A while back I came across this article by Ken Rockwell arguing that bass frequencies all the way down should be reproduced in stereo just like everything else: kenrockwell.com/audio/stereo-subwoofers.htmI haven't found any comments on this from anyone else, especially since he doesn't have a comments section on there, so... does anyone here think his arguments hold water? Are we missing out on arbitrary bass notes by summing them into one channel, because of the canceling effect of phase differences between the left and right originals?
|
|
|
Post by arnyk on Aug 10, 2016 14:11:23 GMT
There has long been a controversy over stereo bass below 40-80 Hz. I don't find your source to be generally credible, whether he has things right in a specific case or not. I find him to often be imaginative, not factual. My experience is that the point where stereo bass is audible depends on many things, but mostly the source program material, the room acoustics, and both the subwoofers and upper range speakers. Low frequencies were generally summed together below 80-120 Hz in the days of vinyl because the medium and the hardware was usually incompetent to handle frequencies this low competently. It frequently added spurious signals due to the mechanical flaws in the medium and equipment that were difficult or impossible to eliminate any other way. Here's a relevant paper: www.davidgriesinger.com/asa05.pdf
|
|
|
Post by donjoe on Aug 10, 2016 18:24:01 GMT
I don't find your source to be generally credible, whether he has things right in a specific case or not. I find him to often be imaginative, not factual. Well, he says as much in his general description of what that blog contains: he says sometimes he will post things just to troll people. It's actually part of why I asked for other (serious) opinions on that piece. Yeah, I read something about that just recently - it was something about how the phase differences between realistically separated left and right bass notes could make the needle jump out of the groove during playback. But I don't think this affects much of what's been produced since the advent of digital media.
|
|
|
Post by arnyk on Aug 11, 2016 8:54:44 GMT
I don't find your source to be generally credible, whether he has things right in a specific case or not. I find him to often be imaginative, not factual. Well, he says as much in his general description of what that blog contains: he says sometimes he will post things just to troll people. It's actually part of why I asked for other (serious) opinions on that piece. Yeah, I read something about that just recently - it was something about how the phase differences between realistically separated left and right bass notes could make the needle jump out of the groove during playback. But I don't think this affects much of what's been produced since the advent of digital media. Actually it does happen since the advent of digital, because a fair amount of irritating and distracting low frequency noise in recordings can often be vastly reduced by summing the very lowest frequencies. Yes, these noises and their bad effects on perceived sound quality were a much bigger problem with vinyl. A lot of vinyl was made with everything below 150 Hz summed to zero. A lot of vinyl was made rolled off strongly below 50 or 80 Hz. Many recordings are made in much smaller rooms that can't really support phase differences in the lowest frequency sound between different parts of a room. So why try to reproduce them? That has nothing to do with how you distribute the recordings - analog or digital.
|
|
|
Post by Ethan Winer on Aug 11, 2016 21:20:31 GMT
All I can add is that one time, many years ago when I was a newbie, I recorded a pipe organ recital at a church. I put stereo microphones far apart to get a "good stereo spread" and sent the master tape to some place that did small runs of LPs. The guy at the LP place called me on the phone and chewed me a new one, complaining about all the phase problems in my recording. I guess he managed to work around that because the church bought a few hundred LPs and sold them at their services.
|
|
|
Post by donjoe on Aug 12, 2016 9:42:23 GMT
a fair amount of irritating and distracting low frequency noise in recordings can often be vastly reduced by summing the very lowest frequencies ... or by not allowing noise into your recording in the first place. And anyway, just because some recordings have low frequency noise that needs to be gotten rid of doesn't mean that all of them do and that all of them have summed-up single-channel bass and thus that stereo subwoofers are generally or absolutely worthless. "Support" phase differences? I don't understand. All you need to get loudness and phase differences at the left and right microphones is for the bass note source to be positioned off-center; then the two direct waves going to the microphones will have different distances to travel and voila: different loudness and different phase for each. And if the room's walls and corners are treated, like any recording studio should be, any reflected components or room modes should have negligible influence. All that should change due to the smaller size of the room are the maximum phase differences possible to have at the very lowest bass frequencies (but not all bass frequencies) and the frequencies of the room modes (which will be higher). For everything other than that there's still a case to be made for listening to such a recording on stereo subwoofers. And then, of course, you have all those other recording rooms that aren't small.
|
|
|
Post by arnyk on Aug 12, 2016 14:46:13 GMT
a fair amount of irritating and distracting low frequency noise in recordings can often be vastly reduced by summing the very lowest frequencies ... or by not allowing noise into your recording in the first place. And anyway, just because some recordings have low frequency noise that needs to be gotten rid of doesn't mean that all of them do and that all of them have summed-up single-channel bass and thus that stereo subwoofers are generally or absolutely worthless. Since when does what I said: "a fair amount of irritating and distracting low frequency noise be vastly reduced by summing the very lowest frequencies" mean the same thing that "stereo subwoofers are generally or absolutely worthless"? If it doesn't, then you are arguing with yourself. Let's look at the facts. Stereo subwoofers are like hen's teeth - almost nobody has them. Speaker management systems with independent outputs for two (or more) independent subwoofers are very rare and primarily found only in the largest and most elaborate systems. Almost every pre-recorded and live source has mono bass. Almost every AVR and surround sound processor provides mono subwoofer outputs. I'm not saying they don't exist because I've heard a few of the few that exist. But they are rare. Does stereo bass raise many veils from those few recordings that actually has stereo bass? IME No! "Support" phase differences? I don't understand. All you need to get loudness and phase differences at the left and right microphones is for the bass note source to be positioned off-center; then the two direct waves going to the microphones will have different distances to travel and voila: different loudness and different phase for each. And if the room's walls and corners are treated, like any recording studio should be, any reflected components or room modes should have negligible influence. All that should change due to the smaller size of the room are the maximum phase differences possible to have at the very lowest bass frequencies (but not all bass frequencies) and the frequencies of the room modes (which will be higher). For everything other than that there's still a case to be made for listening to such a recording on stereo subwoofers. And then, of course, you have all those other recording rooms that aren't small. [/quote] Almost all commercial and private recordings that are made in large rooms have mono sub-bass. Please give label and title for enough exceptions to justify the complexity and costs. Please tell us about your personal system with stereo sub bass. My private system has two subwoofers, but I use them, individually equalized, to obtain smooth bass response.
|
|
|
Post by donjoe on Aug 12, 2016 15:35:19 GMT
Almost every pre-recorded and live source has mono bass. And here we have the basic contradiction between your opinion and Rockwell's, because according to him, 1980[...] Digital cheerfully can record straight down to DC, and the new digital "Compact Discs" (CDs) allowed anyone who could afford a CD player to have perfect sound forever. [...] For the first time in 100 years of electronic sound reproduction, consumers have access to recordings with unlimited deep bass. Since there is no point in mixing or summing the bass to mono, bass has been in stereo in our homes since the introduction of the CD. I can't contribute my own experience to this dispute, as I'm only here doing my homework before ever spending a dime on any kind of "serious" subwoofer (I don't even own a home yet). I guess for my part all I could do is take a random sample of the digital recordings I currently own, low-pass filter them and subtract one channel from the other to see what's left.
|
|
|
Post by arnyk on Aug 12, 2016 18:30:24 GMT
Almost every pre-recorded and live source has mono bass. And here we have the basic contradiction between your opinion and Rockwell's, because according to him, 1980[...] Digital cheerfully can record straight down to DC, and the new digital "Compact Discs" (CDs) allowed anyone who could afford a CD player to have perfect sound forever. [...] There is no dispute at this point and saying so shows how little some understand about audio. It is a well known and true fact that for all practical purposes the CD format and every CD player, computer audio interface, and portable digital player that I've tested (dozens and dozens) can play two well-separated audio channels from a few Hz to 20 or more Hz above DC to the limits of 44.1 KHz digital coding with a few exceptions. For example, some early iPods and other digital players rolled off well above 50 Hz with some common headphone loads. I hedge a tiny bit because no real world player can actually reproduce pure DC. Its low frequency limit is the inverse of the length of the longest track it can handle which is always some finite length, even if hours long. But for most practical purposes, not much harm done if we call that DC. ;-) Just about every real world digital player has some kind of built in bass roll off that is greater than 0.1 Hz, but usually less than 10 Hz. Again, no serious harm done if we call that DC. I can't contribute my own experience to this dispute, as I'm only here doing my homework before ever spending a dime on any kind of "serious" subwoofer (I don't even own a home yet). I guess for my part all I could do is take a random sample of the digital recordings I currently own, low-pass filter them and subtract one channel from the other to see what's left. This is totally and utterly false. AFAIK there are no commercial recordings with anything like unlimited deep bass. I've never found any, and I've looked. I think I know why this is, which is that the bass response of digital recordings is almost always set elsewhere in the production process than the console or the recorder or the player. There is a lot more to recording music and dialogue than recording consoles and digital recorders/players. There is the slight matter of the microphones that are used for all recordings of real world audio. They all roll off somewhere from around 120 Hz down to maybe 5 or 10 Hz. Most roll off at 30=-50 Hz or so, but many, especially the mics that are used to record vocals and speech usually start falling off at far higher frequencies. I can't believe that a person can be credible to anyone as any kind of audio authority who does not know this simple fact.
|
|
|
Post by donjoe on Nov 19, 2016 11:56:47 GMT
I guess for my part all I could do is take a random sample of the digital recordings I currently own, low-pass filter them and subtract one channel from the other to see what's left. So I finally got around to doing that, but with an extra twist: since I chose to believe the claim in this article that blind testing shows most people unable to hear bass directionality once the crossover is set at 60 Hz or lower, I wanted to eliminate those sub-bass frequencies from the analysis as well and only keep the 60-80 region, which is what might be heard differently by human listeners if reproduced incorrectly with a single sub crossed over at 80 Hz, which is still quite a frequently used setting. (Though of course this all depends on how close to the sub you're sitting: the audibly-directional frequencies may go even lower than 60 - at the extreme, I've seen someone in another forum say they need to cross over at 45 to not be able to tell where the bass is coming from.) So I took the 13 songs you can see below, bandpass-filtered them so that 60 Hz remained untouched, with a -48 dB/octave roll-off below that and a -12 dB/octave slope above (typical for the crossovers on the market, though quite mediocre), which goes to -6 dB at 80 Hz. Then inverted one channel and added both to mono to see what left-right difference there was in the chosen "directional subwoofer band". My conjecture is that any song that has audible inter-channel differences in this band can benefit from a stereo subwoofer setup (or, one could say, from never crossing over a single subwoofer higher than 60 Hz, but there's more to discuss along that line of thinking). Now, to my eye at least 7-9 of these tracks (all released in the 2000s and later) have audible stereo bass information, which is more than 50% of them and deviates by quite a large margin from what I've been seeing in some experts' claims that "90-99% of modern music has mono bass". (And yes, the Celt Islam track actually turns out that way, I double-checked.)
|
|
|
Post by arnyk on Nov 20, 2016 14:51:56 GMT
The musical selections shown are obviously from a very tiny esoteric niche musical genre, and thus largely irrelevant to any discussion that has a reasonable degree of generality.
To help the discussion, I was thinking of mainstream music from the US market, such as that which was top-100 in the Billboard charts.
|
|
|
Post by Ethan Winer on Nov 21, 2016 21:24:32 GMT
Aside from the type of music, that a track contains "stereo" bass - different content left and right below 60-80 Hz - has no bearing on whether people can discern that the bass range is different. When people can tell, it's always because a window is buzzing, or a nearby front-firing woofer port is making air noise, etc.
|
|
|
Post by arnyk on Nov 22, 2016 14:13:30 GMT
Yes, an authoritative test would be an ABX listening test involving mainstream sources of music,using a unquestionably relevant criteria (actual listening in a real world system) and involving a wider band of music than just a limited band around 60 Hz.
|
|
|
Post by donjoe on Dec 19, 2016 14:37:18 GMT
that a track contains "stereo" bass - different content left and right below 60-80 Hz - has no bearing on whether people can discern that the bass range is different. This was never about bass range. I was talking about whether people can tell where the bass notes are coming from, if they can localize the bass source. If they can, the presence of stereo bass information means two subwoofers are required for faithful reproduction of the full soundstage. And here we get to that "more things to discuss" part I mentioned in the parenthesis: even if you find the perfect crossover frequency such that your sub will not reproduce any locatable bass frequencies at all (from the recording being played back), there are still other possible "tells", like port chuffing or harmonic distortion or unwanted resonances that may still give away the location of your single sub and thereby disturb the stereo image. All these headaches go away if you simply use two subwoofers positioned not too far away from your main left and right speakers. As for genres, I doubt mainstream top-whatever billboard-whatever music is all that relevant to audiophiles who would care about this debate in the first place; it certainly isn't to me.
|
|
|
Post by arnyk on Dec 19, 2016 19:28:05 GMT
that a track contains "stereo" bass - different content left and right below 60-80 Hz - has no bearing on whether people can discern that the bass range is different. This was never about bass range. I was talking about whether people can tell where the bass notes are coming from, if they can localize the bass source. If they can, the presence of stereo bass information means two subwoofers are required for faithful reproduction of the full soundstage. Bass in the 60-80 Hz range may or may not be reliable perceived as being directional, depends on several things. (1) Since a high proportion of speakers can't reproduce cleanly in this range, the sounds question may shift up an octave, two or more, because that's where the speaker or room-generated harmonics and/or noises may land. At 120 Hz and up, clean bass can have clear directionality, and the problem gets turned on its head - not having sound localize to the speaker that reproduces it. (2) IME with clean speakers in rooms with clean smooth response (no resonances or rattles), directionality disappears when you get much below 60 Hz. Clean smooth response below 80 Hz, particularly at higher listening levels is actually pretty rare, even in real performance venues. There are probably a lot of people with strong opinions about the need for stereo bass who have based their conclusions on bad listening tests in poor rooms with marginal speakers. For example, many people have expensive and highly regarded (by some audiophiles) 2 channel speaker systems that get pretty messy below 80 Hz. Or, they have rooms with panel resonances or nonlinearitites in the walls and floors. They are also suspicious or adverse of subwoofers. In fact their systems are producing many percent of THD at low frequencies when played a little bit loud.
|
|