1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.

    Dismiss Notice

Testing audiophile claims and myths

Discussion in 'Sound Science' started by prog rock man, May 3, 2010.
829 830 831 832 833 834 835 836 837 838
840 841 842 843 844 845 846 847 848 849
  1. Don Hills
    I recall the predominant theory it was to do with the different acoustic space dimensions. The smaller the room, the lower the actual SPL for the same perceived SPL.
    I'm still looking, the relevant archive is on an old machine which I will need to resurrect.
    castleofargh likes this.
  2. analogsurviver
    Correct. It has to do with the size of acoustic space environment. Which, at home, is usually MANY times smaller than the original recording venue.

    The ONLY way to play back music while preserving all the dynamic range AND spatial cues is binaural reproduced over EARSPEAKERS - not headphones.

    Binaural suffers on the recording side of things; it works best outdoors - absolutely no adjacent boundaries and consequent echoes from the walls, etc - followed by a large venue - and can be absolutely atrocious sounding in small highly reverberant rooms without much or any acoustic treatment.

    Catch is that one has to know that - and use that knowledge accordingly. Binaural is NOT the answer or solution to all the real world recording scenarios,
  3. gregorio
    Loudness is a complex thing, which involves a number of different and concurrent variables unrelated to the actual SPL. Most here are aware of the Equal Loudness Contours but this is only one of the variables, another of the variables is as you mention, room size (acoustic space dimensions). This has been well known for many decades and is most obviously demonstrated/exhibited in the film sound world. Sound for film is edited and designed in relatively small rooms, typically bedroom or small living room sized rooms but it is intended for playback in relatively large rooms (cinemas) and therefore mixed in "mix stages" that are also cinema sized. Cinema sound has a specified/defined SPL calibration at the listening position: 85dBSPL = -20dBFS. However, that ONLY applies to cinemas and the similarly sized mix stages, it does NOT apply to the sound edit suites or sound design rooms where 85dBSPL = -20dBFS at the listening position sounds much louder, and therefore the translation of the edited content from edit suite to mix stage wouldn't work as intended with the same SPL calibration level. This difference in perceived loudness due to room size is not only known about but well quantified and has been defined for decades. Dolby explicitly specified this for film sound at least 30 years ago (to my knowledge) and probably more than 40 years ago: 85dBSPL = -20dBFS for cinemas and mix stages but 78dBSPL = -20dBFS for smaller rooms (edit suites and sound design rooms). These two significantly different SPL calibration levels result in roughly the same perceived loudness, although it should be mentioned that the 78dBSPL calibration level is variable depending on the exact size of the room. In practice it can be as low as 72dBSPL in very small edit suites or as much as 80dB or so in larger sound design rooms.

    There's another very important variable as well. If I pose the question: "What will sound louder, a peak at 95dBSPL or a peak at 85dBSPL?" and eliminate the other variables (IE. Both peaks are the exact same recording of the same sound, listened to in exactly the same room, with exactly the same system and even the exact same listener), the obvious answer would be the 95dBSPL peak. However, that's not necessarily the case, the 85dB peak could sound louder because there's another variable, it also depends on what precedes the peak. If we precede the 95dB peak with several seconds of 90dB content and precede the 85dB peak with several seconds of 40dB content, even if the peak and content are exactly the same, the 85dB peak will sound louder. This is because the ear/brain perceives loudness as a relative quantity, the 95dB peak is only 5dB higher than the preceding content whereas the 85dB peak is 45dB higher. In other words, the contrast (or dynamic range) is itself a determinant of loudness. This fact has been well known about (and employed) by composers for over 200 years. It's what made Beethoven's 5th Symphony so shocking in it's day and is a fundamental tool employed in film sound (pretty much ubiquitously in horror films) for many decades. This variable applies to consumer music listening, a live symphony concert for example typically has a fairly high noise floor, despite a great deal of time and money spent on concert hall acoustics and sound isolation. We've got 90 odd musicians and many hundred (up to a few thousand) members of the audience, all of whom are moving and breathing (plus of course some HVAC to enable them to breath) and therefore a noise floor typically around 45dBSPL (in some cases it can be lower and in others higher). If we assume that in an ideal seat an audience member might receive a peak level of 100dB, that's a dynamic range of around 55dB. Let's say using HPs in their own home that the noise floor is 35dB, 55dB higher than that would be 90dBSPL and would be perceived as the same level as the live event because the dynamic range is the same, even though the peak level is 10dB lower. This is in fact why IEMs were invented in the first place, to allow performing (pop/rock) musicians to "monitor" at far lower levels, thereby not having to use stage monitor speakers at very high levels, which was potentially damaging to the musicians' hearing and often impacted the the main (audience) mix.

    Last edited: Feb 9, 2019
    Steve999, muza_1, bfreedma and 3 others like this.
  4. gregorio
    1. But why on earth would we want to preserve all the dynamic range and spatial cues??? The actual dynamic range at say a symphony concert can be very small, because an audience member will be a considerable distance from the orchestra but right in the middle of the audience and therefore the music will be quieter and the audience noise louder. In practice, the brain will concentrate on the music and reduce the audience noise, giving us the experience of a dynamic range which is significantly greater than what actually enters our ears. The same is true of room reverberation, the brain will concentrate on the music and effectively change the balance between direct and reflected sound (increase the direct sound/reduce the reflected sound). A competent recording engineer/music producer will try to create an idealised recording of the sound EXPERIENCED at a live performance, NOT the sound that actually existed at a particular seat in the audience. That is why competent recording engineers never use just a couple of mics at the seating position but numerous mics, most of which are very much closer to the orchestra than the audience (which increases the ratio of orchestra sound to audience noise and increases the direct sound relative to the reverberant sound).

    2. Hang on, binaural best preserves "spatial cues" (which are largely defined by "consequent echoes") but "suffers on the recording side of things" and works best when there are no "consequent echoes". Extremely impressive self contradiction there!
    2a. What large venues? Stadia and arenas for example are often a complete nightmare, with massive slap back echoes which present a very serious problem (both to the live sound engineers and a would be binaural recording). Classical symphony halls on the other hand, particularly the more modern ones, are very carefully designed to increase reverberation/reflections ("consequent echoes"), particularly in the lower freq band. Even old concert halls, both large and small, extensively use wooden panels and flooring specifically for this purpose (of making the venue more reverberant) and the best contemporary classical concert halls often incorporate specifically designed echo chambers/areas. Or, what about other common large music venues like churches and cathedrals? Massively thick stone walls, floors and ceilings and consequently, some of the most highly reverberant spaces ever made!

    3. One has to know what, how to contradict one's self?
    3b. Which patently you are not! Because:
    3c. Have you any idea what an (un-amplified) orchestra sounds like outdoors? So that rules out "outdoors" classical music recording in binaural (and most other types of music). Almost all large concert venues and many smaller ones must also be ruled out because they're highly reverberant and can/would "be absolutely atrocious". And the main reason we have recording studios in the first place is so that we can alter the "spatial cues" during recording and/or mixing, so that rules them out too. So what's left? When IS binaural "the answer or solution" and how do you "use that knowledge accordingly"? And, how does your statement that "binaural is NOT the answer or solution to all recording scenarios" reconcile with your first statement that binaural is "the ONLY way to playback music while preserving all the dynamic range and spatial cues"?

    It's just utter nonsense analogsurviver, your own assertions don't even agree with your own assertions, let alone the actual facts!!!

    taffy2207, bfreedma and sonitus mirus like this.
  5. GearMe
    Stole this from the Schiit thread...electron microscope image of stylus on record, for the occasion of the Grammy Awards.

    NOW I get it!

    You can even see the better soundstage, warmer/fuller/richer mids, etc. :wink:

  6. GearMe
  7. mbphotox
    Well, given everything that goes along with Vinyl, I do find that it "sounds better".
    May not be technically better, but there's much more involved with listening.

    You put the LP on the player, carefully brush it, select the right speed and gently set the needle down, then raise the volume on the amp.
    Listening to Vinyl is so much more intimate and therefore "feels" better.

    I don't need an expensive player, cables, or amp, but they look great and that helps the "sound experience" immensely!

    What I've been wondering, but cannot find the source anymore:
    Some tube amps deliver significantly less power at higher impedance levels and vice versa.
    If you use a 5W tube amp on a Focal Grande Utopia (impedance ranges from 2 to 16 ohms), you're bound to hear differences to a more powerful amp, because the SPL at different frequencies will be audibly different. (we're talking 3dB+ dips and rises in the curve)

    Would this already be considered "clipping", because no sane person uses a 5W amp for such a hungry speaker?
  8. castleofargh Contributor
    one can enjoy vinyl playback and even prefer it to digital playback. nobody here claimed that it was wrong to have personal taste and to act on them. just like the love of ritual you mention is clearly a reality for many people(probably everybody TBH). being a listener is a passive hobby, just like we enjoy tweaking our playback chain and picking special stuff to contribute in some ways, we can enjoy being more involved thanks to vinyls. there is certainly nothing impossible in all that.
    what most of us oppose actively is claiming that the subjectively better felt by some people, means that vinyls have objectively higher fidelity. we reject that idea because many measurements clearly disprove it. but otherwise, several people in here who would never agree that vinyl has better fidelity than digital playback, own vinyls and enjoy them a great deal. there is no contradiction in that.

    about speaker amps, you're right that some designs, or simply the impedance output can affect the frequency response by up to several dB sometimes. but power notions are mostly unrelated. ideally, we should always purchase the amp to go with our transducers and not the other way around. so it is assumed that our amp can supply X watt into a specific load and that X is the amount of watt you need to get as loud as you'll ever want to go on that system. beyond that, power isn't causing much on its own. if it does, it's only a sign that this specific amplifier isn't a good match for those specific transducers and your listening habits, and probably shouldn't have been purchased for that job in the first place. the only reason I wouldn't entirely blame the user in such a situation, is because the specs provided by manufacturers can sometimes be all over the place. but there is a simple fix for that too. if specs aren't clear about what is given and how it was measured, I suggest to go purchase gears that do provide proper specs and nomenclature. this too works in a loop, if we stopped purchasing gears with BS published specs, manufacturers would stop publishing BS specs to fool consumers. power problem pretty much solved ^_^.
    mbphotox likes this.
  9. KeithEmo
    From your description you seem to be talking more about "the experience" than how vinyl actually sounds.

    Actually, in most cases, the situation with the output power on tube amps is reversed from what you described.

    Most solid state amps are what we call "load invariant" - which means that they deliver the same output voltage regardless of the load (within reason).
    With amplifiers that follow this "rule", you will get more power into a 4 Ohm load than into an 8 Ohm load, with the same volume setting.
    Because of this, most solid state amps are rated to deliver more power into low impedance loads.

    In contrast, most tube amplifiers use output transformers to match the load impedance seen by their output devices to the impedance of the load.
    (Power tubes work best into very high impedances, on the order of a few thousand Ohms; the output transformer matches this to the 4 - 8 Ohm load presented by most modern speakers.)
    However, most output transformers have separate output taps for specific speaker impedances, and most have at least a 4 Ohm tap and an 8 Ohm tap....
    And, for MOST amplifiers that have multiple taps on their output transformer, their power output rating will be identical into each of those impedances.
    (So, for example, a typical solid state amp may deliver 50 watts into 8 Ohms and 100 watts into 4 Ohms, while a typical tube amplifier will deliver the same 50 watts into either.)

    Note that this applies to virtually all amplifiers designed to run speakers - but many headphone amplifiers exclude the expensive transformers.
    (Headphones require relatively little power so getting a lot of power, or good efficieny, really doesn't matter with a typical headphone amp.)

    "Clipping" refers to a very specific type of distortion commonly caused by overloading an amplifier.
    The waveforms are actually "clipped off" and become flat on the top and bottom - because the amplifier cannot keep up with what it's being asked to do.
    Clipping results in a rather distinctive - and unpleasant - sort of distorted sound.
    You are certainly more likely to cause an amplifier to clip if the amplifier is under-powered and the speakers are inefficient - but you can avoid doing so by turning down the volume.
    (In general, most amplifiers can actually "run" most speakers, as long as you don't ask them to deliver more power than they are capable of putting out...)

    mbphotox likes this.
  10. analogsurviver
    Although correct, the above is a form of generalization.

    You have left one, but IMPORTANT form of amplifier - HSD OTL. HSD = Hollow State Devices = tubes = valves = Rohren,
    OTL = output transfoermer less

    These amps are the best for driving the best transducers - electrostatics. Either headphones or speakers.

    There are also special solid state amps for transformer coupled electrostatic speakers - which I nicknamed "welding apparatus". They MUST be stable well into sub 1 ohm impedance region - and heavily reactive one at that. The VERY last thing you want in such an amp is any form of current limitting - because, any of the schemes yet tried will bury the amp in sonical terms when driving a typical ES load.
    These amps are capable of almost perfect doubling of output power/current into halvcing the impedance of the load - and would not self destruct when presented with a dead short.

    That means a BIG amp - with big linear power supply and a VERY powerful output stage device(s) - on a hefty heat sink arrangement.

    In real life, there is very little chance an amplifier will ALWAYS be running within its output capabilities - either voltage or current - unless it is a particularly high efficiency transducer at the end or the amp is not extremely big and therefore expensive. Tube amps clip less severely/more gracefully than do solid state amps in general - transfer function of a typical tube does not change drastically, whereas transistor is extremely linear - BUT only up to the limit, above which is a sharp cutoff that has as the consequence the most unpleasent sounding ( all measurable, of course ) harmonic spectrum clipping characteristics.

    Many solid state amp designers have tried to ameliorate this FACT ( where tubes are, de facto, better than SS ) by trying to simulate at least the front end of a solid state amp to simulate tubes as much as possible.So, maligning tube designs as inferior to SS by default much practiced in this thread is simply - wrong.

    Above is not something "all amplifiers sound thew same" brigade is likely to love to hear - and although amp and load characteristics that do not require such extreme and therefore expensive measures are not rEquired definitely do exist, it is equally true that there are - at least as numerous - cases where an amp should pack under its cover every trick known to mankind in order to "survive" the WAY out of "normal" load conditions experienced in real life.
  11. KeithEmo
    As I said... most...

    OTL tube amps are in a pretty small minority these days...
    As are solid state amps with output transformers...
    Although both do still exist.

    I would, however, disagree about most amplifiers running within their output capabilities.
    There are many solid state amps that are powerful enough that they usually remain within their intended operating range (unless you have very inefficient speakers or play your music very loudly).
    The best way to avoid worrying about the overload characteristics of an amplifier is simply to use an amplifier that is powerful enough to remain within its linear operating range the vast majority of the time.
    (I agree that most tube amps overload more gracefully than most solid state amps - but I would simply prefer not to hear an amplifier overloading at all.)

    I also agree that driving electrostatic speakers is a special case.....
    However, I would point out that designing an amplifier to run them using solid state circuitry wouldn't be especially difficult today.
    Solid state devices with the required voltage and current capabilities to do so are in fact readily available nowadays.
    However, they tend to be expensive, and few people seem to be designing electrostatic speaker amps these days.
    (If you're designing the amplifier from scratch it obviously makes sense to avoid the transformer and drive the panels directly.)

  12. bigshot
    Not having to go through the ritual of playing a record is one of the best things about my music server. I don’t have to worry about accidentally damaging my records, and I can jump from one song to another without having to dig through stacks. It brings me closer to the music.

    The best tube amps sound just as accurate as typical solid state amps.
    Last edited: Feb 21, 2019
  13. KeithEmo
    I absolutely don't miss "the ritual" of playing vinyl.

  14. castleofargh Contributor
    I somehow do. it's strange because I would never consider getting myself a turntable again, but I had rituals that included putting on a vinyl and several other stuff, that would put me in a glorious mood when I'd perform them. like breakfast on a weekend would never feel as good without putting on Stevie Wonder or Marvin Gaye on the old turntable, then proceed to make freshly pressed orange juice, boiled egg and soldiers and go enjoy the last few song while looking at the mountains(outside if the weather allowed it). 110% ritual, and somehow pressing play on my playlist just doesn't feel the same(at that moment @castleofargh knew he had become an old fart. but somehow it felt fine).
    autosleeper, GearMe and taffy2207 like this.
  15. bigshot
    I have a Thorens transcription table, and it doesn't have auto return/stop. Whenever I would try to do something like make breakfast, my hands would be covered with pancake batter or maple syrup whenever the record reached the end of the side and it would sit there clicking in the inner groove until I got finished cooking.
829 830 831 832 833 834 835 836 837 838
840 841 842 843 844 845 846 847 848 849

Share This Page