Announcement

Collapse
No announcement yet.

How much SPL do you really need?

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • #16
    Re: How much SPL do you really need?

    Really?

    Originally posted by Pallas View Post
    My opinion (with which you may also disagree, though you did not address the issue one way or another) is that efficient speakers with smooth polars and extended bandwidth will always sound better than inefficient speakers with smooth polars and extended BW. Even at the same average levels, and even if those levels are well within the theoretical capability of either speaker.
    ************************************************** *******
    Sorry, I can't buy the above statement, either, as it seems to not be scientifically valid; a less sensitive/efficient speaker system simply needs more power to reach a specific SPL level, and if the drivers are capable of that and the driving amplifier has adequate power and headroom, then the sound should be essentially identical as compared to a more sensitive/efficient system driven with an appropriate amplifier. I must admit, however, I've never been able to do such a comparison. Also, I did read and understand what you were saying even with your using efficiency and sensitivity interchangeably.

    As to the statement below, I have no doubt you're correct about the loudness levels in the 5th-7th row, but I never sat that close all the years I attended live symphony concerts.
    Paul
    ************************************************** ******
    Actually, inasmuch as we disagree it's over a few dB. I think, and rather clearly wrote above, that a system should ideally be able to handle peaks of ~105dB broadband, and more in the midbass and below. The latter is perhaps a matter of opinion. The former is more firmly grounded: 105dB peaks is cinema reference*, and studies have shown measured peaks in 5th-7th row center floor seats in major orchestral halls (Musikverein, etc.) to be in the 104-106dB range, with average levels of up to about 85dB depending on the piece.

    Comment


    • #17
      Re: How much SPL do you really need?

      Originally posted by Deward Hastings View Post
      For "reproduction" and listening at realistic (and even a bit above realistic) levels yes . . . for "show and tell" maybe not.
      That is so true. My wife always asks me why I crank it up so loud when someone comes over to listen to my speakers. She says "you never ever listen to it that loud." She's right.
      Craig

      The lowest possible F3 box alignment is not always the best alignment.

      Designing and building speaker projects are like playing with adult Lego Blocks for me.

      Comment


      • #18
        Re: How much SPL do you really need?

        Originally posted by spasticteapot View Post
        I'm curious - how much SPL, regardless of frequency, do you think is required at the listener for a hi-fi experience? 90dB at your ears seems to be a general minimum for continuous listening, but what about high-power transients?
        There's an awful lot of dB this and dB that being thrown around with no mention of weighting.

        You can't talk about how loud things need to be "regardless of frequencies". Our ears are more or less sensitive to different frequency ranges, and how we measure SPL takes that into account. At least in the US, the only two weighting systems in common use are A-weighting and C-weighting.

        A-weighting (like bkeane1259 is using in that screen shot) is either a good choice or a useless one depending on what you are trying to measure. It's a reasonably good approximation of what frequencies our ears are most sensitive to, but those frequencies are found above 1k. It follows a simplistic Fletcher-Munson equal loudness curve.

        C-weighting makes a lot more sense if you are concerned with the overall SPL, including bass frequencies, and is probably a better way to talk about music (vs industrial noise). It's possible to have very low A-weighted SPLs while royally ticking off the neighbors.

        And even after all that, unless you are measuring noise, there's no good answer. Music is too varied, too dynamic, to make sweeping generalizations. I will say 85-90dB A-weighted studio-recorded popular music is "very loud", 95-100dB C-weighted is also "very loud". Anything about 95dBA and above and my fingers go in my ears. That starts to hurt. A kick drum at 95dBC not so much. It just depends. Low end peaks I would say 105dBC where you are sitting is a reasonable upper limit.

        But it is almost impossible to give any kind of guideline answer. It really just depends. Electronic dance music is not an orchestra is not a jazz trio is not a noise band.

        So, yeah.

        Comment


        • #19
          Re: How much SPL do you really need?

          Originally posted by Paul K. View Post
          Really?

          Sorry, I can't buy the above statement, either, as it seems to not be scientifically valid; a less sensitive/efficient speaker system simply needs more power to reach a specific SPL level, and if the drivers are capable of that and the driving amplifier has adequate power and headroom, then the sound should be essentially identical as compared to a more sensitive/efficient system driven with an appropriate amplifier. I must admit, however, I've never been able to do such a comparison.
          You're making an assumption that I think is not valid in the real world. Namely, that small-signal behavior linearly translates to large-signal behavior. With more sensitive speakers, the vast majority of what one hears is (very) small signal behavior. When you start talking about 83-85 dB/W/m speakers, it starts not to be. And peaks require the drivers to shed quite a bit of heat.

          I don't honestly know what it is - I had assumed the difference noted was dominated by thermal/power compression, but a paper by John K suggests otherwise - but there is something that seems to make a more efficient speaker simply better than a less efficient speaker driven by a massive amp. Having experimented both ways, there just isn't a compelling reason to go low-efficiency.

          I was struck most recently by the difference when I heard the Statements, which have 2 8" Dayton woofers, 2 4" (I think) TangBand midranges, and a Fountek (I think) ribbon tweeter, with a sensivitity in the upper 80s, and then heard the Orions (a variant with different woofers I've never seen before, not the Peerless XXLS, but the upper drivers seemed to be the expected Seas Excel units) at AXPONA. Both of them had rounded-off sounding peaks. And while I didn't get my choice of program material on the Orion, the Statements also "sounded loud" (~300W separate amp) whereas my reference speakers (which have less volume displacement but are 96dB/W/m sensitive, powered by a ~140W amp in an AVR) simply do not. They just get loud.

          But most people don't get a chance to experiment with more efficient speakers. Efficient drive units capable of high fidelity performance get expensive fast. Also, too many people are still fixated on getting "full range" performance out of their mains, despite the copious research into smoothing room modes with multiple subwoofers.
          --
          "Based on my library and laboratory research, I have concluded, as have others, that the best measures of speaker quality are frequency response and dispersion pattern. I have not found any credible research showing that most of the differences we hear among loudspeakers cannot be explained by examining these two variables." -Alvin Foster, 22 BAS Speaker 2 (May, 1999)

          Comment


          • #20
            Re: How much SPL do you really need?

            Originally posted by Sydney View Post
            Presuming the 80 - 85db is avg. level and 103 - 104 peaks..
            That is a window of 18 - 24db.
            In doing normalizing and spectrum analysis that is in-line with the level differences I'm accustom to observing.
            Dependent on the source material of course and how far the dynamics are used.
            ( What was the track used? )
            I thought we were in the 80-85dB range more during loud passages. I'd put our range between 70-90dB, but there was a wide variety of program material.

            I'm pretty sure we hit above 100dB on Stravinsky's Firebird Suite more than once, but I also recall, at one point, the Dave Matthew Band - "Too Much" track being pumped up and hitting close to or above 100dB, and that was (no pun intended), TOO MUCH. Obviously depends on the program material with regard to ear sensitivity.

            I could not imagine for a second, listening to any of the new, awfully compressed crap at 90+dB for any length of time. However, isn't this exactly what our poor kids are doing?
            Bryan K.

            Midwest Audio Club

            Speedster | Sub Attachť | The Wildeman | Sean's NLA Towers | CO‹GAR, COUGAR II and CO‹GAR JR | Triton | Lithium | J-Boom | Trym MLTL | Docere MLTL

            Comment


            • #21
              Re: How much SPL do you really need?

              Originally posted by Pallas View Post
              You're making an assumption that I think is not valid in the real world. Namely, that small-signal behavior linearly translates to large-signal behavior. With more sensitive speakers, the vast majority of what one hears is (very) small signal behavior. When you start talking about 83-85 dB/W/m speakers, it starts not to be. And peaks require the drivers to shed quite a bit of heat.
              What kind of speakers are you listening to?

              Also, I've seen some high-SPL tests of the ScanSpeak 6600 tweeter and others. They measure very well even at high output.

              Comment


              • #22
                Re: How much SPL do you really need?

                Originally posted by Pallas View Post
                and then heard the Orions (a variant with different woofers I've never seen before, not the Peerless XXLS, but the upper drivers seemed to be the expected Seas Excel units) at AXPONA. Both of them had rounded-off sounding peaks.
                ORION never used Peerless XXLS woofers, through version 3.3 they used the XLS. The drivers you heard were Seas L26ROY, which in ORION 3.4 and the "commercial" ORION 4 will be replaced with a custom driver from Seas based on, but slightly different from, the L26. The drivers in ORION have "sensitivities" of 88-89 dB/Watt, so the speakers easily provide realistic concert levels at less than a Watt average power input, and 105dB "peaks" at less than 50 Watt input . . . all well below the driver's thermal and mechanical ratings.

                You're welcome to like or not like ORION, as you wish, and the same with the demo music used at AXPONA . . . but to allege "compression" or "rounded-off peaks" is simply false. ORION give you the dynamic range that is on the CD, intact and undistorted.
                "It suggests that there is something that is happening in the real system that is not quite captured in the models."

                Comment


                • #23
                  Re: How much SPL do you really need?

                  Originally posted by fastbike1 View Post
                  If you have 90dB at your ears for periods longer than about 20 - 30 minutes you're damaging your hearing.
                  Simply not true. There are standards that you can google.

                  Do you have a link that says 90dB for 30 minutes will damage hearing?? All my links and even work standards say 8 hours.

                  With your opinion anyone that has been to a Rock Concert should have damaged hearing :rolleyes: It does not work that way.

                  Besides in a full treated room with higher quality designs 90dB is not that loud at all. It must be all the crap rooms and limited driver choices that makes 90dB sound too loud for people

                  Comment


                  • #24
                    Re: How much SPL do you really need?

                    Originally posted by AMC View Post
                    +1, sort of makes you reconsider current dogma on sensitivity.
                    Dogma

                    Several facts.

                    1. A common distance in custom HT rooms is 15 feet.

                    2. Peaks in movies can be 25 to 30 dB.

                    3. If you are listening at 15 feet and you want clean peaks when listening at 75 dB you will need speakers that can do 115dB. Not only the speakers but the amps have to handle the peaks cleany.

                    Conclusion, lets just say HIGH END custom HT designs do not use low sensitivity designs ( <= 92dB) with wimpy 6.5" woofers. I know this isnt exactly what the majority here who support so many small/low sensitivity drivers meant more for 2 channel small room music but its just the simple fact that there is a scientific reason designs from Seaton, JTR, Danley are just superior for HT usage.


                    Its up to each individual to do their own calculations to find out how much their drivers are distorting/compressing (bad SQ) and how much their amps are clipping (Bad SQ). They can then choose their compromise.

                    For my custom HT, low sensitivity design simply sound like crap including B&W 802Ds!!!

                    Comment


                    • #25
                      Re: How much SPL do you really need?

                      Originally posted by spasticteapot View Post
                      What kind of speakers are you listening to?

                      Also, I've seen some high-SPL tests of the ScanSpeak 6600 tweeter and others. They measure very well even at high output.
                      I have never seen quality 110dB tests from any Dome tweeter.

                      If this is an HT application you have to move beyond the limits of a dome design. They simply have limited directivity control and limited dynamics.

                      Comment


                      • #26
                        Re: How much SPL do you really need?

                        True. OSHA 1910 allows 90 dB for a period of 8 hours. However, prolonged exposure to sound greater than 85 dB will cause hearing damage according to the National Institute on Deafness. Here is another source (audio related) citing 90-95 dB causes permanent hearing damage.

                        With regard to your comment about rock concerts, I would say that yes, anyone who has attended a rock concert likely has some hearing damage. From that second link, you will see a concert is around 115 dB. OSHA standards would say you could be exposed to this type of loudness for 15 minutes a day. Now, I'll admit OSHA probably has at least a 3x safety factor before any research would show that their recommended levels cause permanent damage, however a concert is usually 3-4 hours long.

                        If 90 dB is "not that loud" to you, you probably already have some hearing damage or your background level is quite high.

                        Originally posted by penngray View Post
                        Simply not true. There are standards that you can google.

                        Do you have a link that says 90dB for 30 minutes will damage hearing?? All my links and even work standards say 8 hours.

                        With your opinion anyone that has been to a Rock Concert should have damaged hearing :rolleyes: It does not work that way.

                        Besides in a full treated room with higher quality designs 90dB is not that loud at all. It must be all the crap rooms and limited driver choices that makes 90dB sound too loud for people

                        Comment


                        • #27
                          Re: How much SPL do you really need?

                          Originally posted by Deward Hastings View Post
                          You're welcome to like or not like ORION, as you wish, and the same with the demo music used at AXPONA . . . but to allege "compression" or "rounded-off peaks" is simply false. ORION give you the dynamic range that is on the CD, intact and undistorted.
                          Interesting on the woofers. I didn't know Seas made an XLS competitor. Where's the crossover now? Perhaps the new woofers are forcing the crossover to the relatively small (Seas Excel 8") and unloaded midranges down too far? I could often see the poor little guys moving.

                          Actually, I liked them quite a bit, considering "Show Conditions." I can't say that about many other rooms. (In truth, the only thing I left AXPONA wanting was that McIntosh mini-system to put on display in my office!) I expected I would, of course. We're talking Siegfried effing Linkwitz here!

                          The one thing I didn't like about them is that they did sound dynamically limited. Better than most of the speakers at that show, to be sure, but still a little rounded-off. Sorry, that's just what I heard.

                          Have you used any genuinely efficient speakers, i.e. speakers that need 6dBW+ less power to reach the same levels, that are likewise designed with "hi-fi" priorities?

                          Originally posted by penngray View Post
                          For my custom HT, low sensitivity design simply sound like crap including B&W 802Ds!!!
                          That's not really a fair comparison, unless you're trying to ape B&W's house sound. Those elevated upper mids get annoying to me after a short listen.
                          Last edited by Pallas; 05-11-2011, 11:14 AM. Reason: replace subjective language ("over emphasized") with objective language ("elevated")
                          --
                          "Based on my library and laboratory research, I have concluded, as have others, that the best measures of speaker quality are frequency response and dispersion pattern. I have not found any credible research showing that most of the differences we hear among loudspeakers cannot be explained by examining these two variables." -Alvin Foster, 22 BAS Speaker 2 (May, 1999)

                          Comment


                          • #28
                            Re: How much SPL do you really need?

                            Originally posted by penngray View Post
                            Besides in a full treated room with higher quality designs 90dB is not that loud at all. It must be all the crap rooms and limited driver choices that makes 90dB sound too loud for people
                            Originally posted by tculverhouse View Post
                            If 90 dB is "not that loud" to you, you probably already have some hearing damage or your background level is quite high.
                            Penngray makes a good point that is, I think, distinct from a discussion of pure SPL.

                            On a set of really good speakers with adequate power, music doesn't "sound loud," or at any rate doesn't sound louder at high actual SPL than it does at low actual SPL. There's just more of it. (I differ from him in that I'm not sure the room has much to do with it. A crap rooms sounds like a crap room at any SPL.) One learns to keep volume levels in check by speaking aloud and seeing how much effort is required to be heard over the music. Lesser speakers, by contrast, will audibly "sound loud." I don't know exactly what distortion components are involved, but it's trivial to hear if you have speakers with enough contrast in swept volume and sensitivity. Or really good headphones adequately powered. They likewise never "sound loud."

                            The flip side of that is, yes, quibble about the numbers and thresholds but the undisputed fact is that prolonged exposure to high SPL will eventually degrade a person's hearing. If you don't trust yourself to keep levels in check, a less capable speaker is probably a good idea, because it will give you audible clues when levels are getting out of hand.
                            --
                            "Based on my library and laboratory research, I have concluded, as have others, that the best measures of speaker quality are frequency response and dispersion pattern. I have not found any credible research showing that most of the differences we hear among loudspeakers cannot be explained by examining these two variables." -Alvin Foster, 22 BAS Speaker 2 (May, 1999)

                            Comment


                            • #29
                              Re: How much SPL do you really need?

                              Originally posted by penngray View Post
                              I have never seen quality 110dB tests from any Dome tweeter.

                              If this is an HT application you have to move beyond the limits of a dome design. They simply have limited directivity control and limited dynamics.
                              Oh please.

                              the range that a dome tweeter covers does not require it to hit 110dB. Music, AND movie content have a downward power curve as you go up in frequency. 110dB pink noise produced by a full range speaker has less than a third of the power concentrated in the high frequencies.
                              R = h/(2*pi*m*c) and don't you forget it! || Periodic Table as redrawn by Marshall Freerks and Ignatius Schumacher || King Crimson Radio
                              Byzantium Project & Build Thread || MiniByzy Build Thread || 3 x Peerless 850439 HDS 3-way || 8" 2-way - RS28A/B&C8BG51

                              95% of Climate Models Agree: The Observations Must be Wrong
                              "Gravitational systems are the ashes of prior electrical systems.". - Hannes Alfven, Nobel Laureate, Plasma physicist.

                              Comment


                              • #30
                                Re: How much SPL do you really need?

                                Originally posted by penngray View Post
                                Dogma

                                Several facts.

                                1. A common distance in custom HT rooms is 15 feet.

                                2. Peaks in movies can be 25 to 30 dB.

                                3. If you are listening at 15 feet and you want clean peaks when listening at 75 dB you will need speakers that can do 115dB. Not only the speakers but the amps have to handle the peaks cleany.

                                Conclusion, lets just say HIGH END custom HT designs do not use low sensitivity designs ( <= 92dB) with wimpy 6.5" woofers. I know this isnt exactly what the majority here who support so many small/low sensitivity drivers meant more for 2 channel small room music but its just the simple fact that there is a scientific reason designs from Seaton, JTR, Danley are just superior for HT usage.


                                Its up to each individual to do their own calculations to find out how much their drivers are distorting/compressing (bad SQ) and how much their amps are clipping (Bad SQ). They can then choose their compromise.

                                For my custom HT, low sensitivity design simply sound like crap including B&W 802Ds!!!
                                115db is LOUD! I personally don't know anyone who wants their movies to go that loud. Even so, those peak sounds aren't going to be coming out of 1 speaker, but the sub and probably the three main fronts for loud explosions(maybe even the rears).

                                I'm not sure that you really need one speaker that can do 115db.

                                I do agree that sensitivity is important, but the speakers you listed are not common and not really NEEDED for good home theater sound. For the people who demand sound pressure levels that will damage their ears and have extreme budgets, then they can shoot for the stars to reach 115db and beyond.

                                I agree that some of the low 80db sensitivity speakers are not going to make very good HT speakers hooked up to a off the shelf AVR. Get near 90db, and most people will be happy with the results.

                                Also, not everyone sits 15' back. Even Onkyo has a HTIB that is good for THX levels at 6' back.

                                Comment

                                Working...
                                X