I am pondering the meaning of the term "actual program loudness."* What kind of time-constant do you have in mind? Or are you referring to the loudness of the specific bass note? Do you mean the anticipated level in consideration of the correction for level being introduced?
*No, I have no trouble understanding the term in other contexts.
An example. A song has a loud part at 0 dB relative to reference level and a soft part -20 dB relative to reference level. Now we listen to that song at -30 dB from that reference level. The louder part needs a different correction curve than the lower part. Let's look at the idealized compensation curves based on ISO226:2003:
In our example the loud part would need the green curve (-30 dB) applied whereas the softer part would need the blue curve (-20 dB - 30 dB = -50 dB) applied.
An externally hosted image should be here but it was not working when we last tested it.
In our example the loud part would need the green curve (-30 dB) applied whereas the softer part would need the blue curve (-20 dB - 30 dB = -50 dB) applied.
Jamikl, thanks for the ganged potentiometer reference. I'll contact them.
Markus76, I agree that doing it "manually" means that it has to be calibrated for each source for best results. It isn't as big a problem as you might think though, depending on the source. For example, most modern music is so heavily compressed ("loudness wars") that one calibration will suffice for most of it. Even before CD, most LPs had reasonably similar levels to each other - set, in this case, by the limitations of the playback cartridge.
DSP is not a complete answer, either. To work properly, it has to know in advance the average loudness of the music you are about to play. Otherwise it's going to be constantly adjusting to variations in the source level. It works for movies because there is a defined reference level for all material, but still requires initial calibration to that level.
I'm beginning to think that it might well be used in conjunction with a PC based media server. It can then use Replaygain or similar to set the source level. In that case, the "loudness control" could be implemented in software on the PC. Even for CD playback, it only takes a few seconds for a modern PC and CDROM drive to read a track and calculate a "loudness" value. It can then start playing the first track while it scans the rest.
Markus76, I agree that doing it "manually" means that it has to be calibrated for each source for best results. It isn't as big a problem as you might think though, depending on the source. For example, most modern music is so heavily compressed ("loudness wars") that one calibration will suffice for most of it. Even before CD, most LPs had reasonably similar levels to each other - set, in this case, by the limitations of the playback cartridge.
DSP is not a complete answer, either. To work properly, it has to know in advance the average loudness of the music you are about to play. Otherwise it's going to be constantly adjusting to variations in the source level. It works for movies because there is a defined reference level for all material, but still requires initial calibration to that level.
I'm beginning to think that it might well be used in conjunction with a PC based media server. It can then use Replaygain or similar to set the source level. In that case, the "loudness control" could be implemented in software on the PC. Even for CD playback, it only takes a few seconds for a modern PC and CDROM drive to read a track and calculate a "loudness" value. It can then start playing the first track while it scans the rest.
I've come to expect smarts when I read a Don Hills post.
Markus - you've postulated a whole theory of how "loudness hearing" works. Hope you realize that? I think Don's implicitly philosophical critique shows that it requires what might be called "teleological" information about the future.
Thinking about the human-ear mechanics of your theory, isn't there something circular about how loud to bang a side drum when the rest of the orchestra is playing softly?
My best guess, for the moment, is simply that the ear is non-linear with respect to the transmission of sounds of different loudness to the brain and it varies by frequency. But that is a bit odd since at lower frequencies, the nerves follow the wave. But if the transmission is non-linear, wouldn't the sound be distorted?
I'm still puzzled after all these years.
Markus - you've postulated a whole theory of how "loudness hearing" works. Hope you realize that? I think Don's implicitly philosophical critique shows that it requires what might be called "teleological" information about the future.
Thinking about the human-ear mechanics of your theory, isn't there something circular about how loud to bang a side drum when the rest of the orchestra is playing softly?
My best guess, for the moment, is simply that the ear is non-linear with respect to the transmission of sounds of different loudness to the brain and it varies by frequency. But that is a bit odd since at lower frequencies, the nerves follow the wave. But if the transmission is non-linear, wouldn't the sound be distorted?
I'm still puzzled after all these years.
No, the louder part does not need a different correction curve.
Imagine you have a track with loud parts and soft parts. Do you turn up the bass during the quiet parts? (You might, but I don't.) The correction curve should be fixed for each track if they are unrelated, or for a complete album if the tracks belong together.
You don't need to turn up the bass during the quiet parts as long as you listen at reference level because the mixing engineer took care of the tonal balance in the mixing process. So the loudness compensation is already "in" the recording. That changes when you listen to that recording at a different level than reference. Louder parts need a different compensation curve than softer parts (see equal-loudness contour) to keep their tonal balance.
That is a clever argument and also explains that a conductor "adjusts" levels for the audience. I'm not convinced the conductor gets sound anything like the audience or that he/she would know the difference or care - they sometimes get excellent halls gutted so THEY can hear better and to hell with the customers.
(In the interest of not being too provocative, for the moment, I'll avoid comments about recording engineers such as about their hearing or taste. It is clear we depend on these talented people and they are a kind of reference, as Markus points out.)
Interestingly, full recording studio loudness (and live music) is one instance where using a loudness control doesn't matter because it is at the top of the nested equal loudness curves anyway. So the issue is moot for that loudness.
Last edited:
An orchestra always plays at reference level. You can't play a forte violin at different sound pressure levels. This can only be done with a recorded violin. So equal loudness volume controls are needed for recordings that get played back at different levels than they were mixed.
The equal loudness curves are in effect throughout the whole range of human hearing, not just within a certain region.
I see what you are getting at, but you still do not quite grasp the concept... If the balance between bass and midrange is correct when listening at "reference level", it will still be correct at lower levels if you have also applied the correct amount of "loudness correction". No additional correction is required.
Assume that at "reference level", the balance between bass and midrange sounds correct. Note that this applies regardless of the dynamics of the music. The music may have soft parts and loud parts but the the musicians, or the mix engineer, will adjust the balance between bass and midrange so that it sounds the way they want it to sound. In other words, you don't have to adjust the bass control during the playing of, for example, Ravel's "Bolero" or Led Zeppelin's "Stairway to Heaven". From what you said above, you understand this.
Now turn your replay volume down by 20 dB. You will also need to boost the bass by about 10 dB, so midrange will be about -20 dB and bass will be about -10 dB from reference. Now play the same music again. The balance between bass and midrange should sound similar to that at reference level, right through the music. You don't have to turn up the bass further during the quiet parts.
Try it by numbers:
Play a 400 Hz ("midrange") tone or 1/3 octave pink noise at "0 dB".
Add a 40 Hz ("bass") tone or 1/3 octave pink noise and adjust it until it is clearly audible (let's say it works out to 0 dB also). That is your "loud music".
Drop the midrange signal to -10 dB. Adjust the bass until it sounds about the same in proportion to the midrange as it did before. The loudness curves predict that the bass level will be about -5 dB. That is your "quiet music".
Notice the relationship in the above: If you lower the midrange by x dB, the bass level needs to drop by about x/2 dB to sound in proportion. If you have Audacity or similar, you can generate these signals and make a WAV file of them for use in the next step:
Next, reduce your overall volume setting by 20 dB. You'll also need to apply about 10 dB of "loudness compensation", according to the loudness curves.
Play your "loud music" and "soft music" again. You should find that the balance between bass and midrange is still correct for both.
Looking at the numbers:
Midrange loud: was 0dB, now -20 dB.
Bass loud: was 0 dB, now -10 dB.
Midrange quiet: was -10 dB, now -30 dB.
Bass quiet: was -5 dB, now -15 dB.
Note that the relative levels between bass and midrange have not changed. For every 10 dB drop in the midrange the bass drops by 5 dB, as predicted.
In short: assuming a properly designed "loudness control" linked to the volume control, the perceived balance between bass and midrange will not alter as you turn the volume up or down. If you first adjust the tonal balance so that it sounds correct at one volume, it will sound correct at all volumes.
So you don't need DSP to automatically adjust the amount of "loudness compensation". The right use of DSP would be to measure and adjust your sources so thay all have the same perceived volume level (Replaygain or equivalent), then have a loudness control linked to your volume control.
You mention the Audyssey - this is designed for HT use where everything is supposed to be calibrated. Soundtracks have their bass / midrange balance set correect relative to 0 dB (105 dB SPL in room). If you set your replay volume to, say, -20 dB, all the Audyssey has to do is boost the bass by 10 dB to match the "loudness curve". I believe the Audyssey does more than this in order to "enhance" the sound for HT use, but this is less relevant for "audio".
Hi Don,
this is only true if the offset of the loudness curves at each frequency does not vary with level. While this is true for the curves found in ISO226:2003 which were derived from pure tones, it might not be true for real world program material. You might want to chime in at "Official" Audyssey thread. - Page 1011 - AVS Forum
Best, Markus
Markus,
I doubt I will ever hear an Audyssey, let alone own one.
But this oft-repeated quote from their web site seems clear enough:
"Audyssey developed proprietary methods that calculate the differences between reference level and playback level in real time. Dynamic EQ is the first technology of its kind to combine information from incoming source levels with actual output sound levels in the room, while taking into account human perception and room acoustics. The result is something never before possible — bass response, octave-to-octave balance and surround impressions that are detailed and accurate at any volume...Audyssey Dynamic EQ selects the correct frequency response and surround volume levels moment-by-moment. The result is something never before possible — bass response, octave-to-octave balance and surround impression that remain as they should be despite changes in volume. This is the first technology to carefully combine information from incoming source levels with actual output sound levels in the room, a pre-requisite for delivering a loudness correction solution."
Chris confirms, in other posts, that the Dynamic EQ function alters the "loudness compensation" on a dynamic basis. For example, if a movie has 10 minutes of quiet sound, it will increase the loudness compensation. If the quiet sound is followed by a period of loud sound, it will back off the compensation.
Fine for movies, maybe, but not something I would care for when listening to music.
You say in the AVS forum:
"Makes perfectly sense but we started the discussion with you saying that different loudness compensation countours need to be applied to the very same track depending on the loudness of certain parts within that recording. I did show that this is not the case when using data from ISO226:2003. What am I missing?"
What you are missing is that he is (a) talking about movie sound, not music, and (b) he wants to promote the Audyssey as providing a function that is not achievable elsewhere.
He says later that their research indicated that the ISO curves "lacked refinement". There may be some truth to his claims, but until / unless he publishes his research or corroborating evidence we do not know. We do know here is a significant body of published research behind the current ISO curves.
I looked back a few months but I did not see him make any claims that the Audyssey's Dynamic EQ function is suitable for music. Have you seen him make such a claim?
In summary, even for movie use, I see dynamic "loudness compensation" as a solution looking for a problem. When I say "dynamic", I mean compensation over and above the "manual" compensation that should be applied when altering the overall volume.
Take an example of a slamming door. Recorded close up, it will be loud and there can be a significant deep bass component. Now say the slam is quiet - that is, recorded from further away. In real life, you don't hear significant bass. The Audyssey will tend to bring up the bass. This will certainly sound more dramatic, unless the director deliberately wanted the slam to sound distant.
However, for music, there is no such justification. Take Ravel's "Bolero". The double bass initially comes in while the levels are still moderate. The loudness of the bass is chosen to be audible but not overpowering. As the piece progresses and the overall levels rise, the bass level also rises but still in proportion to the rest of the players - if the overall level rises by 10 dB, the bass level rises by somewhat less. The point is that "loudness compensation" is built into music by the composers / musicians / mix engineers, and if you make a static adjustment to the volume, you only need to make a static adjustment to the loudness compensation. The rest is already taken care of in the music.
I doubt I will ever hear an Audyssey, let alone own one.
But this oft-repeated quote from their web site seems clear enough:
"Audyssey developed proprietary methods that calculate the differences between reference level and playback level in real time. Dynamic EQ is the first technology of its kind to combine information from incoming source levels with actual output sound levels in the room, while taking into account human perception and room acoustics. The result is something never before possible — bass response, octave-to-octave balance and surround impressions that are detailed and accurate at any volume...Audyssey Dynamic EQ selects the correct frequency response and surround volume levels moment-by-moment. The result is something never before possible — bass response, octave-to-octave balance and surround impression that remain as they should be despite changes in volume. This is the first technology to carefully combine information from incoming source levels with actual output sound levels in the room, a pre-requisite for delivering a loudness correction solution."
Chris confirms, in other posts, that the Dynamic EQ function alters the "loudness compensation" on a dynamic basis. For example, if a movie has 10 minutes of quiet sound, it will increase the loudness compensation. If the quiet sound is followed by a period of loud sound, it will back off the compensation.
Fine for movies, maybe, but not something I would care for when listening to music.
You say in the AVS forum:
"Makes perfectly sense but we started the discussion with you saying that different loudness compensation countours need to be applied to the very same track depending on the loudness of certain parts within that recording. I did show that this is not the case when using data from ISO226:2003. What am I missing?"
What you are missing is that he is (a) talking about movie sound, not music, and (b) he wants to promote the Audyssey as providing a function that is not achievable elsewhere.
He says later that their research indicated that the ISO curves "lacked refinement". There may be some truth to his claims, but until / unless he publishes his research or corroborating evidence we do not know. We do know here is a significant body of published research behind the current ISO curves.
I looked back a few months but I did not see him make any claims that the Audyssey's Dynamic EQ function is suitable for music. Have you seen him make such a claim?
In summary, even for movie use, I see dynamic "loudness compensation" as a solution looking for a problem. When I say "dynamic", I mean compensation over and above the "manual" compensation that should be applied when altering the overall volume.
Take an example of a slamming door. Recorded close up, it will be loud and there can be a significant deep bass component. Now say the slam is quiet - that is, recorded from further away. In real life, you don't hear significant bass. The Audyssey will tend to bring up the bass. This will certainly sound more dramatic, unless the director deliberately wanted the slam to sound distant.
However, for music, there is no such justification. Take Ravel's "Bolero". The double bass initially comes in while the levels are still moderate. The loudness of the bass is chosen to be audible but not overpowering. As the piece progresses and the overall levels rise, the bass level also rises but still in proportion to the rest of the players - if the overall level rises by 10 dB, the bass level rises by somewhat less. The point is that "loudness compensation" is built into music by the composers / musicians / mix engineers, and if you make a static adjustment to the volume, you only need to make a static adjustment to the loudness compensation. The rest is already taken care of in the music.
Yes, the usual wagonload of commercial exaggeration. In reading that kind of marketing bumff, it is helpful to understand something about what the Patent Office looks for in claims. That is why guys hot to get patents on their projects will shout about the most trivial benefits, provided no patent has boasted about those trivial benefits before.
I think there is a crucial question that needs answering. In the abstract it is: what is the time-constant of adaptation to loudness, assuming that concept applies at all. I don't know if there's research on that, just don't know. But what I do know, is that the measurement of equal loudness contours is a far trickier method than any I've ever tried and more beset by assumptions, loose directions to testees, and other of what we in the trade call "threats to validity."
I think there is a crucial question that needs answering. In the abstract it is: what is the time-constant of adaptation to loudness, assuming that concept applies at all. I don't know if there's research on that, just don't know. But what I do know, is that the measurement of equal loudness contours is a far trickier method than any I've ever tried and more beset by assumptions, loose directions to testees, and other of what we in the trade call "threats to validity."
You can do a crude experiment to judge your own adaptation time-constant.
I described a method here:
http://www.diyaudio.com/forums/soli...loudness-control-post1982108.html#post1982108
.. that you can do with minimal tools - a dB meter or suitable digital multimeter, some music, and a source of tones. I would suggest marking the scales directly with dB before you start - no need to pick a 0dB level, just play a tone at a comfortable level and then adjust the controls and make a mark whenever the level changes by, for example, 3 dB.
Using music with "loud bits and quiet bits" for the actual test (after calibration) will add the "time constant of adaptation to loudness". You can also experiment with bringing the bass up slightly during the quiet parts and backing off during the loud parts to see if it sounds more natural.
Before the argument gets too esoteric as they often do here I would just like to reiterate my initial question and that was for some of controlling xmax and
equalisation with and without subs. I definitely do not have tin ears but do know what I am after which I know Don Hills understands and I thank him again for the time he has put into this.
jamikl
equalisation with and without subs. I definitely do not have tin ears but do know what I am after which I know Don Hills understands and I thank him again for the time he has put into this.
jamikl
Hi Don,
We don't know if this is true and I didn't get a direct answer from Chris when I specifically asked for this. I'll check the behavior of Dynamic EQ when my Onkyo arrives but I think this is a functionality of Dynamic Volume which is more or less just a compressor. It can be switched off.
Of course he wants to promote Audyssey but at the same time we can get information that otherwise would never make it into public.
He's also talking about music and I don't see the point why only movie sound would benefit from equal-loudness compensation. Equal-loudness is reality with any signal. The problem with music production is that there's no standardized reference level what makes finding the correct compensation curves a matter of luck. Nonetheless equal-loudness compensation would need to be applied when listening at any other level than reference.
Unfortunately there's no publication I know of about equal-loudness that uses any other signal than pure tones, let alone music. We can only speculate whereas Chris states that their research showed that equal-loudness needs to be applied dynamically, whatever that means.
Dynamic EQ doesn't differentiate between music or movie. If you look in one of the Onkyo manuals then you can find this:
Audyssey Dynamic EQ Reference Level Offset
Best, Markus
We don't know if this is true and I didn't get a direct answer from Chris when I specifically asked for this. I'll check the behavior of Dynamic EQ when my Onkyo arrives but I think this is a functionality of Dynamic Volume which is more or less just a compressor. It can be switched off.
Of course he wants to promote Audyssey but at the same time we can get information that otherwise would never make it into public.
He's also talking about music and I don't see the point why only movie sound would benefit from equal-loudness compensation. Equal-loudness is reality with any signal. The problem with music production is that there's no standardized reference level what makes finding the correct compensation curves a matter of luck. Nonetheless equal-loudness compensation would need to be applied when listening at any other level than reference.
Unfortunately there's no publication I know of about equal-loudness that uses any other signal than pure tones, let alone music. We can only speculate whereas Chris states that their research showed that equal-loudness needs to be applied dynamically, whatever that means.
Dynamic EQ doesn't differentiate between music or movie. If you look in one of the Onkyo manuals then you can find this:
Audyssey Dynamic EQ Reference Level Offset
- 0dB:
It should be used when listening to movies. - 5dB:
Select this setting for content that has a very wide dynamic range, such as classical music. - 10dB:
Select this setting for jazz or other music that has a wider dynamic range. This setting should also be selected for TV content as that is usually mixed at 10 dB below film reference. - 15dB:
Select this setting for pop/rock music or other program material that is mixed at very high listening levels and has a compressed dynamic range.
Best, Markus
Agreed. The current discussion really belongs in its own thread.
It's easy enough to build a loudness compensation circuit for a fixed speaker setup. Building one that smoothly hands over from OB to subs as the level increases is a whole new level of complexity. You first need to determine if it is even possible with your specific combination of OB speakers and subs in your room.
To do this you'll need a multi-amped (active crossover) setup with an adjustable high-pass for the OB woofer drive. The aim is to progressively cut the low bass passed to the OB woofers while replacing it with bass from the sub. You'll need to pick a crossover / high pass frequency that suits both the sub and the OB woofers.
Start without the sub, with no high-pass, and increase the level until you are approaching the limits of the OB woofers. This is your reference level. Now increase the volume further in small (2 or 3 dB) steps) and at each step increase the high-pass cut to keep the OB woofers under control. Then add some sub level to make up for the high-pass cut. After you have it tweaked to your satisfaction at each volume step, note the volume level (+dB from the reference level), high pass cut in dB, and sub drive level change in dB. You should end up with a list of values which can be plotted on a graph to provide the adjustment curves required.
If you're lucky, the curves can be approximated with a relatively simple analog filter circuit. It might have to have three volume controls - one to set the initial reference level, one for OB alone listening, and one for "sub assisted" listening. When the "OB alone" control reaches max, then you start turning up the "sub assisted" volume control.
If it can't be done simply, you'll need a flexible programmable filter unit where the curves can be programmed in. It'll need 3 or 4 independently programmable outputs - one for the sub, one for the OB woofers, one or two for the OB mids and tweeters. It might be possible to program a Behringer crossover to do this. I don't have one so I can't comfirm it.
old school
Comments:
In order to obtain a good audio reproduction at different listening levels, a different tone-controls setting should be necessary to suit the well known behaviour of the human ear. In fact, the human ear sensitivity varies in a non-linear manner through the entire audible frequency band, as shown by Fletcher-Munson curves.
A simple approach to this problem can be done inserting a circuit in the preamplifier stage, capable of varying automatically the frequency response of the entire audio chain in respect to the position of the control knob, in order to keep ideal listening conditions under different listening levels.
Fortunately, the human ear is not too critical, so a rather simple circuit can provide a satisfactory performance through a 40dB range.
The circuit is shown with SW1 in the "Control-flat" position, i.e. without the Automatic Loudness Control. In this position the circuit acts as a linear preamplifier stage, with the voltage gain set by means of Trimmer R7.
Switching SW1 in the opposite position the circuit becomes an Automatic Loudness Control and its frequency response varies in respect to the position of the control knob by the amount shown in the table below.
C1 boosts the low frequencies and C4 boosts the higher ones. Maximum boost at low frequencies is limited by R2; R5 do the same at high frequencies.
Technical data:
Frequency response referred to 1KHz and different control knob positions:
Total harmonic distortion at all frequencies and 1V RMS output: <0.01%
Notes:
• SW1 is shown in "Control flat" position.
• Schematic shows left channel only, therefore for stereo operation all parts must be doubled except IC1, C6 and C8.
• Numbers in parentheses show IC1 right channel pin connections.
• R7 should be set to obtain maximum undistorted output power from the amplifier with a standard music programme source and P1 rotated fully clockwise.
Parts:
P1_________________10K Linear Potentiometer (Dual-gang for stereo)
R1,R6,R8__________100K 1/4W Resistors
R2_________________27K 1/4W Resistor
R3,R5_______________1K 1/4W Resistors
R4__________________1M 1/4W Resistor
R7_________________20K 1/2W Trimmer Cermet
C1________________100nF 63V Polyester Capacitor
C2_________________47nF 63V Polyester Capacitor
C3________________470nF 63V Polyester Capacitor
C4_________________15nF 63V Polyester Capacitor
C5,C9_______________1µF 63V Electrolytic or Polyester Capacitors
C6,C8______________47µF 63V Electrolytic Capacitors
C7________________100pF 63V Ceramic Capacitor
IC1_______________TL072 Dual BIFET Op-Amp
SW1________________DPDT Switch (four poles for stereo)
Comments:
In order to obtain a good audio reproduction at different listening levels, a different tone-controls setting should be necessary to suit the well known behaviour of the human ear. In fact, the human ear sensitivity varies in a non-linear manner through the entire audible frequency band, as shown by Fletcher-Munson curves.
A simple approach to this problem can be done inserting a circuit in the preamplifier stage, capable of varying automatically the frequency response of the entire audio chain in respect to the position of the control knob, in order to keep ideal listening conditions under different listening levels.
Fortunately, the human ear is not too critical, so a rather simple circuit can provide a satisfactory performance through a 40dB range.
The circuit is shown with SW1 in the "Control-flat" position, i.e. without the Automatic Loudness Control. In this position the circuit acts as a linear preamplifier stage, with the voltage gain set by means of Trimmer R7.
Switching SW1 in the opposite position the circuit becomes an Automatic Loudness Control and its frequency response varies in respect to the position of the control knob by the amount shown in the table below.
C1 boosts the low frequencies and C4 boosts the higher ones. Maximum boost at low frequencies is limited by R2; R5 do the same at high frequencies.
Technical data:
Frequency response referred to 1KHz and different control knob positions:
Total harmonic distortion at all frequencies and 1V RMS output: <0.01%
Notes:
• SW1 is shown in "Control flat" position.
• Schematic shows left channel only, therefore for stereo operation all parts must be doubled except IC1, C6 and C8.
• Numbers in parentheses show IC1 right channel pin connections.
• R7 should be set to obtain maximum undistorted output power from the amplifier with a standard music programme source and P1 rotated fully clockwise.
Parts:
P1_________________10K Linear Potentiometer (Dual-gang for stereo)
R1,R6,R8__________100K 1/4W Resistors
R2_________________27K 1/4W Resistor
R3,R5_______________1K 1/4W Resistors
R4__________________1M 1/4W Resistor
R7_________________20K 1/2W Trimmer Cermet
C1________________100nF 63V Polyester Capacitor
C2_________________47nF 63V Polyester Capacitor
C3________________470nF 63V Polyester Capacitor
C4_________________15nF 63V Polyester Capacitor
C5,C9_______________1µF 63V Electrolytic or Polyester Capacitors
C6,C8______________47µF 63V Electrolytic Capacitors
C7________________100pF 63V Ceramic Capacitor
IC1_______________TL072 Dual BIFET Op-Amp
SW1________________DPDT Switch (four poles for stereo)
Attachments
**** El Greco Fallacy Alert ****
hint: why not boost the range 20,000-25,000 Hz because human hearing is so poor there.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.