Another go at DSS filter ...
OK. So I'm willing to give the DSS shelving filter another go. This time with adustment and longer period.
I start with my flat dipoles. Flat here is my definition strictly, which is measurement from 1m both outdoor and indoor, ungated.
With this setup I'm happy. Sure they may sound bright on some material as many people would point out. But many times they are simply perfect. I find it also that with FM radio station, those "pop" stations would sound bright, but the "classical station" like my favourite 3MBS in melbourne they just sound perfect and realistic.
First I tried to tune down the tweeter as many speaker builders normally do. I turned down the level about 2-3db. I've done this before and know what to expect. They sound alright, not bright at all, but on many other materials they would sound dull. These are purely subjective observation. I did not take long. I listened for 2 days and to memorise my impression.
Then I put in the 3.3db shelving filter as in Orion 3.2. I immediately did not like it. The sound seems muted. This is what SL may meant by "non-descript gray" perhaps? But what if I change the attenuation level?
Well ... I found out that changing just few db makes big difference (yeah just like audiophiles say when they change something hey? ...🙂). But indeed at -2db the sound is almost like unattenuated level (flat) and 2.5 sounds "just right". But it is still different from simply turning down the tweeter level.
So will try to live with it a week or so and report.
OK. So I'm willing to give the DSS shelving filter another go. This time with adustment and longer period.
I start with my flat dipoles. Flat here is my definition strictly, which is measurement from 1m both outdoor and indoor, ungated.
With this setup I'm happy. Sure they may sound bright on some material as many people would point out. But many times they are simply perfect. I find it also that with FM radio station, those "pop" stations would sound bright, but the "classical station" like my favourite 3MBS in melbourne they just sound perfect and realistic.
First I tried to tune down the tweeter as many speaker builders normally do. I turned down the level about 2-3db. I've done this before and know what to expect. They sound alright, not bright at all, but on many other materials they would sound dull. These are purely subjective observation. I did not take long. I listened for 2 days and to memorise my impression.
Then I put in the 3.3db shelving filter as in Orion 3.2. I immediately did not like it. The sound seems muted. This is what SL may meant by "non-descript gray" perhaps? But what if I change the attenuation level?
Well ... I found out that changing just few db makes big difference (yeah just like audiophiles say when they change something hey? ...🙂). But indeed at -2db the sound is almost like unattenuated level (flat) and 2.5 sounds "just right". But it is still different from simply turning down the tweeter level.
So will try to live with it a week or so and report.
I am still very surprised at the way this flat is not correct thing has been received. To me it make very little sense. The thing is that what you hear at the listing position is a function of the direct sound and the reflected sound, and the ratio of the two. The direct sound is controlled by the on axis response or what ever axis you are listening on. The reflected sound is controlled by the power response of the speaker and the room absorption characteristics. The power response is related to the on axis response by the variation of the directivity index with frequency.
Regardless of the uniformity, or lack there of, of the radiated power with frequency, if we place a speaker in a room where the ratio of direct to reflected sound is optimum at the listening position, what ever that is, then altering the high frequency response will not alter the ratio of direct to reflected sound at any given frequency because the directivity is fixed by the speaker design. The spectrum of the total sound field will, however, change. If the same speaker is now placed in a room which is more reverberant at high frequency, less high frequency energy will be absorbed and the ratio of direct to reflected sound will change from the optimum. There will be a greater contribution to the reflected sound at high frequency since a lower percentage of the radiated power will be absorbed. If we cut the high frequency using a shelving filter we may restore the spectral balance of the total sound field, but the ratio of direct to reflected sound will remain unaffected and will still be different than the optimum. For example suppose that in the optimum room the direct sound has an amplitude of 1.0 and the reflected sound is also 1.0. We have a ratio of direct to reflected of 1 and the total sound level is 2. Now we move to the non-optimum room. The direct sound remains at a level of 1 but the reflected sound is now 2. The ratio is 1/2, the sum is 3. If we restore the correct sum we must cut on axis response and hence the radiated power by 2/3. This yields total sound level of 2, as in the optimum room, but the ratio of direct to reflected remains 1/2.
If, however, we choose to listen in a highly reverberant room, or at a distance well past the critical distance, where the reflected sound is dominate, then shelving down the high frequency, or any other equalization that restores the correct spectral balance to the sound field, will be beneficial. (The critical distance is the distance where the direct and reflected sound are of equal level. As the distance increases beyond the critical distance the reflected sound becomes dominant.) But if the listening position is such that the reverberant field is dominant then a great deal of the information from direct sound (spatial clues, transient attacks, understanding of spoken words, etc) is lost. If that is what you are after, fine. It follows Bose theory.
However, there is still a catch, what if we choose speaker A over speaker B and the directivity of speaker A is very different from speaker B? Both may have perfectly flat or otherwise identical axial response but they will have very different power responses. As a result regardless of which room they are placed in, they will produce different direct to reflected sound ratios. The solution is not to simply adjust the on axis response but the design the room to have the correct absorption characteristics for each speaker. The absorption characteristics will determine the ratio of direct to reflected sound, and must be different for each speaker. Once the room characteristics are set, we can then choose to alter the total sound field through equalization of the axial response while retaining the targed ratio of direct to reflected sound.
I think it should be apparent that we have a sliding scale here between room characteristics, on axis response and directivity. Obviously if we listen in an anechoic chamber all that matters is the direct sound. The radiate power (hence directivity) are not relevant. If we listen in a highly reverberant room, or at a distance well past the critical distance it is the direct sound (or axial response) that becomes irrelevant. Thus we are free to alter the on axis response to correct the power response so as to provide the desired total sound field in the given room. In between is the realm of reality where we must couple the room characteristics to the speakers directivity. If we have a room with fixed characteristics we must adjust the directivity of the speaker to suit. If the speakers directivity is fixed, it is the room which must be adapted. Once this is achieved the axial response can be manipulated to what ever is desired w/o altering the D/R ratio.
All in all, the proposed shelving function may well provide a satisfactory correction for the speaker to which it is applied provided the environment is appropriate and remains within some specified range of reverberation. I believe that in the case of the speaker for which it was developed the designer indicated an RT60 on the order of 600 msec. That is pretty reverberant compared to the 300 msec the BBC considered in their studies on controlled directivity, or so I recall.
It would seem that the best we can do is to design speakers with smooth on axis response and comparably smooth power response vs frequency. From there it is a matter of coupling the room to the speaker to achieve the desired direct to reflected sound ratio at the listen position and from there, apply what ever equalization to the axial response is required to make any given recording sound balanced. It should also be apparent that the shelving function proposed is basically the reintroduction of a treble level control. And the implication that it could possible be correct for all speakers in all rooms for all recording is simply unacceptable.
Regardless of the uniformity, or lack there of, of the radiated power with frequency, if we place a speaker in a room where the ratio of direct to reflected sound is optimum at the listening position, what ever that is, then altering the high frequency response will not alter the ratio of direct to reflected sound at any given frequency because the directivity is fixed by the speaker design. The spectrum of the total sound field will, however, change. If the same speaker is now placed in a room which is more reverberant at high frequency, less high frequency energy will be absorbed and the ratio of direct to reflected sound will change from the optimum. There will be a greater contribution to the reflected sound at high frequency since a lower percentage of the radiated power will be absorbed. If we cut the high frequency using a shelving filter we may restore the spectral balance of the total sound field, but the ratio of direct to reflected sound will remain unaffected and will still be different than the optimum. For example suppose that in the optimum room the direct sound has an amplitude of 1.0 and the reflected sound is also 1.0. We have a ratio of direct to reflected of 1 and the total sound level is 2. Now we move to the non-optimum room. The direct sound remains at a level of 1 but the reflected sound is now 2. The ratio is 1/2, the sum is 3. If we restore the correct sum we must cut on axis response and hence the radiated power by 2/3. This yields total sound level of 2, as in the optimum room, but the ratio of direct to reflected remains 1/2.
If, however, we choose to listen in a highly reverberant room, or at a distance well past the critical distance, where the reflected sound is dominate, then shelving down the high frequency, or any other equalization that restores the correct spectral balance to the sound field, will be beneficial. (The critical distance is the distance where the direct and reflected sound are of equal level. As the distance increases beyond the critical distance the reflected sound becomes dominant.) But if the listening position is such that the reverberant field is dominant then a great deal of the information from direct sound (spatial clues, transient attacks, understanding of spoken words, etc) is lost. If that is what you are after, fine. It follows Bose theory.
However, there is still a catch, what if we choose speaker A over speaker B and the directivity of speaker A is very different from speaker B? Both may have perfectly flat or otherwise identical axial response but they will have very different power responses. As a result regardless of which room they are placed in, they will produce different direct to reflected sound ratios. The solution is not to simply adjust the on axis response but the design the room to have the correct absorption characteristics for each speaker. The absorption characteristics will determine the ratio of direct to reflected sound, and must be different for each speaker. Once the room characteristics are set, we can then choose to alter the total sound field through equalization of the axial response while retaining the targed ratio of direct to reflected sound.
I think it should be apparent that we have a sliding scale here between room characteristics, on axis response and directivity. Obviously if we listen in an anechoic chamber all that matters is the direct sound. The radiate power (hence directivity) are not relevant. If we listen in a highly reverberant room, or at a distance well past the critical distance it is the direct sound (or axial response) that becomes irrelevant. Thus we are free to alter the on axis response to correct the power response so as to provide the desired total sound field in the given room. In between is the realm of reality where we must couple the room characteristics to the speakers directivity. If we have a room with fixed characteristics we must adjust the directivity of the speaker to suit. If the speakers directivity is fixed, it is the room which must be adapted. Once this is achieved the axial response can be manipulated to what ever is desired w/o altering the D/R ratio.
All in all, the proposed shelving function may well provide a satisfactory correction for the speaker to which it is applied provided the environment is appropriate and remains within some specified range of reverberation. I believe that in the case of the speaker for which it was developed the designer indicated an RT60 on the order of 600 msec. That is pretty reverberant compared to the 300 msec the BBC considered in their studies on controlled directivity, or so I recall.
It would seem that the best we can do is to design speakers with smooth on axis response and comparably smooth power response vs frequency. From there it is a matter of coupling the room to the speaker to achieve the desired direct to reflected sound ratio at the listen position and from there, apply what ever equalization to the axial response is required to make any given recording sound balanced. It should also be apparent that the shelving function proposed is basically the reintroduction of a treble level control. And the implication that it could possible be correct for all speakers in all rooms for all recording is simply unacceptable.
In my opinion the complete lack of recording standards in the industry is another culprit that doesn't allow for "one curve fits all". Many pop recordings are so high-pitched that I can't listen to them with flat response. Many classical recordings on the contrary need flat highs to become detailed and airy. Even if I got the direct response, the directivity and the room right, I am still taken hostage by the recording engineer and his peculiarities.
Rudolf
Rudolf
I wonder why I never touch my substantial variety of tone adjustments?
Probably 20 knobs I could twist or buttons I could press. I use a lot of them in setting up but then don't touch them after. BTW, that includes a sophisticated "loudness" compensator that tracks fairly correctly with the setting of my volume control. I leve this "on" all the time and, YES, it compensates for equal-loudness hearing automatically.
I bet that is true of almost everybody out there except most people don't even have the simple good sense to use their tone controls for setting up - as if God had ordained "Let there be flat" and tonal compensation were a sin. (Granted, the tone controls are lousy EQ from a human-factors perspective, and better to install a better EQ.)
I never listen seriously to anything but serious music, so I am not much exposed to wanton commercially-driven distortion. But I never feel an urge to "re-engineer" recordings. Maybe I should tweak old Pablo Casals recordings? Maybe I should tweak lots of recordings. But I never feel much need to do so. Maybe my automatic loudness compensation goes almost the whole way to adjusting tonal balance as music loudness varies.
If others share this operational life-style, then we are saying that recording engineers are not introducing tonal distortion we need to fix. Or that the different direct-reflected ratio of the mic set-ups matters much.
Footnote: for years I wondered why wise people said that a recording has to be played back at the "right" volume. If this is an observation shared by others, it can mean only one thing - equal-loudness compensation is essential. Any other explanation?
Probably 20 knobs I could twist or buttons I could press. I use a lot of them in setting up but then don't touch them after. BTW, that includes a sophisticated "loudness" compensator that tracks fairly correctly with the setting of my volume control. I leve this "on" all the time and, YES, it compensates for equal-loudness hearing automatically.
I bet that is true of almost everybody out there except most people don't even have the simple good sense to use their tone controls for setting up - as if God had ordained "Let there be flat" and tonal compensation were a sin. (Granted, the tone controls are lousy EQ from a human-factors perspective, and better to install a better EQ.)
I never listen seriously to anything but serious music, so I am not much exposed to wanton commercially-driven distortion. But I never feel an urge to "re-engineer" recordings. Maybe I should tweak old Pablo Casals recordings? Maybe I should tweak lots of recordings. But I never feel much need to do so. Maybe my automatic loudness compensation goes almost the whole way to adjusting tonal balance as music loudness varies.
If others share this operational life-style, then we are saying that recording engineers are not introducing tonal distortion we need to fix. Or that the different direct-reflected ratio of the mic set-ups matters much.
Footnote: for years I wondered why wise people said that a recording has to be played back at the "right" volume. If this is an observation shared by others, it can mean only one thing - equal-loudness compensation is essential. Any other explanation?
Last edited:
but lots of times not the final say, even the best are still at the mercy of the band/producer/record company...
I have to say I agree with Rudolf's analysis. But since musical enjoyment is subjective, it is no wonder that individual tastes so often vary.
I believe that the discussion about the Perfect Frequency Response Curve often fails miserably because, in the case of JohnK's indepth analysis, he failed to illustrate the type of music he prefers and the volume level he normally enjoys.
I also think Bentoronto hits the nail on the head. If I had an auto loudness compensation based on volume I would probably be much happier to not feel like I have to tweak my FR slope to attenuate higher frequencies.
DS21 also makes a very astute recognition that the musicians in the band often have an significant impact on the final recording's relative loudness.... many of the musicians I know have significant hearing loss in the higher frequencies and thus it does not come as much of a surprise to me that many (dare I say most) recordings are to bright to my ears when listening at realistic volume levels.
Gainphile, when you report back would you please give us some specifics on the types of music you evaluated and the relative volume levels you used most frequently? 🙂
Cheers,
Josh
John, your reasoning is very strong. In fact I had read your website in regards to power response deviation of OB + typical dome tweeter.
I still want to keep an open mind of SL's finding. I really don't think he's an average tinkerer and had presented his finding and reasoning in few forums like the Burning Amp.
There was something interesting that I found last night. Out of curiosity, having the DSS I disconnected my rear tweeter. Now, I am familiar with it's subjective effects but with the DSS in place I found not having rear tweeter the overall presentation are still great (???). I had in the past done this many times and without the rear tweeter (at that time) the speaker sounds disjointed. What's happening here?
Could it be the shelving filter had provided smoother transition between midrange and (now) forward-firing tweeters hence it's more acceptable? This has got me thinking now to replace the dome with waveguide for better pattern matching....
Anyway.. still few more day/week to observe and we'll see. So at the moment I am using DSS with no rear tweeter.
Josh, my program material varies and I like anything acoustics. I don't have a particular "test track" to say. I have few screenshot here. I also listen alot to FM radio station (3MBS in Melbourne - great volunteer run classical station) and pay attention to the host voice. I found their voice when talking is more realistic in my car stereo than my living room.
I don't need to remind you that all of these are subjective observations.
I still want to keep an open mind of SL's finding. I really don't think he's an average tinkerer and had presented his finding and reasoning in few forums like the Burning Amp.
There was something interesting that I found last night. Out of curiosity, having the DSS I disconnected my rear tweeter. Now, I am familiar with it's subjective effects but with the DSS in place I found not having rear tweeter the overall presentation are still great (???). I had in the past done this many times and without the rear tweeter (at that time) the speaker sounds disjointed. What's happening here?
Could it be the shelving filter had provided smoother transition between midrange and (now) forward-firing tweeters hence it's more acceptable? This has got me thinking now to replace the dome with waveguide for better pattern matching....
Anyway.. still few more day/week to observe and we'll see. So at the moment I am using DSS with no rear tweeter.
Josh, my program material varies and I like anything acoustics. I don't have a particular "test track" to say. I have few screenshot here. I also listen alot to FM radio station (3MBS in Melbourne - great volunteer run classical station) and pay attention to the host voice. I found their voice when talking is more realistic in my car stereo than my living room.
I don't need to remind you that all of these are subjective observations.
All this presumes that the direct field and reverberent field are equally important, or that the steady state response (the combination) is what we hear. This just isn't so. The direct (or early) sound has much more importance than the later reverberent field in determining our perception of frequency balance.
A lot of studies and papers point to this. In equalizing PA systems researchers found that setting the direct response to flat gave best results even if the direct field was 15dB lower than the reverberent field. Lipshitz and Vanderkooy did a study where they independently manipulated the direct and reverberent fields and found that large holes in the reverberent field were generally unnoticed. They also found that flat power response was undesirable both in combination with rising direct response (the easiest way to achieve it) or even with flat direct response. If you look at the speakers that did well in Floyd Toole's studies, they generally had falling power response and a variety of power response holes. They overwhelmingly had smooth and flat axial response.
Its easy to measure the steady state response in a room but there is no evidence that it is the subjectivly right thing to measure.
David S.
It is my contention that all SL is really doing is correcting the tweeter level for the power response issues. So any speaker that has such issues may benefit. But as I indicated earlier in the thread (i think it was here) I think the better path is to attempt to correct the power response problem as I have done with the Note and/or as Earl Geddes has done with his speakers. I would not say that even my and Earl's approaches provide the correct relation ship between on axis response and the variation of directivity with frequency, but I do think that speakers which are free distorted power response probably represent a better test bed than one which has irregular behavior, particularly in a critical area like the upper midrange, lower treble. I guess that I am in a unique position to compare the effect of smoothing out the power response between the transition of mids to tweeter. Even in my room which has RT60 closer to 300 msec, which I prefer, compared to the 600 msec SL indicates is what he recommends, the difference between the Note and NaO II is apparent. I have described the difference as the II having an ELS like sizzle where as the Note does not. Both speakers have very flat axial response. In fact, the Note is flatter than the II in the sense that the II has somewhat of a BBC dip in it the way I have it set up. Trust me, you would not want to put a shelving filter on the Note, at least not in my room.
I have been looking into this myself ever since I design the original NaO and indicate so on the web page you refer to just under Figure 4, "This would clearly be an ideal result: flat power response to 15K and uniform directionality as well. I am currently researching tweeters which might be used in such an application." The Note is the result of that research. I also played around with shelving down the tweeter in the II. It and the ability to switch off the rear tweeter have always been a design feature of the II. I never found the shelving of the tweeters to be universally satisfactory. On material where the II tends to sound a little aggressive sometime it is better to switch off the rear tweeter and leave the response flat. It is very dependent on source material.
I just don't think that a treble control which is something that has been around since the dawn of Hi Fi represents any kind of break through. I also think that the reason this has garnered so much attention here and on other discussion groups is largely because of who presented the idea. I guess I'm a little different because I consider only what is said and attach no emphasis or relevance on who said it.
It may well be that the shelf works well with the speaker it was applied to, but even so, based on what I posted previously about direct and reflected sound and the relationship between the reflected sound, power response and room characteristics, a single fixed response can not be universally correct for even one speaker in all rooms. Form my point of view the problem comes down to defining the correct ratio of direct to reflected sound vs frequency and then that will set the relationship between room characteristics and the directivity of the speaker. It's not exactly that straight forward, but this is a more consistent argument.
I find this truly surprising. Having spent a fair amount of my life in front of a P.A. at the mixing desk it doesn't ring true. And having hauled a P.A. rig into hockey rinks, aircraft hangers, museums, old and new theaters, outdoors, etc. I can tell you they did not sound anywhere near the same. Nor was the final house EQ the same. And that was ALL the reverberant field. Everything else was identical.
Would love to read those studies. I just don't understand what they mean. 😕
The 15dB number (direct field predominates even if 15dB below the level of the reverberent sound) comes from an AES paper: Theoretical and Practical Considerations in the Equalization of Sound Systems by William Connor, April 1967. Queen and Bridges had similar conclusions in other AES papers.
Since you have experience with PA I'm sure you are well aware of the notion of "House Curves". In large venues the PAs are never EQed to flat but generally to some falling treble response that makes the system sound balanced. This, of course, is the essence of this thread but magnified by the larger dimensions of an auditorium: Should we equalize a speaker to flat and how should we be measuring it when we do?
For cinemas they have come up with a very non-flat "X curve" and even have a variable curve with different amounts of treble roll-off based on number of seat or size. Nobody gives a good reason why flat would be wrong or even why the ideal target is a moving function based on theater size, but it is easy to see that if the direct field were flat, the reverberent field will get more downward tipped the larger and deader the venue.
At PSB I did some measurements where a speaker, very flat in the direct field, was measured in several rooms at ever increasing distances. At typical listener locations this resulted in a treble down shelf of about 2dB. I'd suggest that that was a good "house curve" for a typical home listener-position EQ, although measuring the direct field will always be more accurate than adjusting the reverberent field.
David S.
speaker dave often makes good sense to me.
There's a chapter missing from Toole which is about aural perception. People have lots of vocabulary about vision and are able to talk about visual phenomena but not so much about the other senses.
Mountains in the distance aren't "photo flat." There are all kinds of physical shifts in the signal (besides the basic geometric ones) which tells us they are far away (color shift towards blue, for example) and are called by psychologists (AKA neural scientists) perceptual "cues" to distance.
Something similar in sound. With crisp Martin Logan treble, you can hear the rosin on Gil Shahan's violin bow and the little sharp noises on guitars. And that means "I must be up close to the player." Take away some of the crispness and you feel further away. (A lot of people don't want a speaker that makes them feel up-close all the time. Some do like the magnifying-glass perspective.)
So there are sound cues of various sorts in music room, PA, or theater set-ups that tell the listener various things about the environment. Some make the environment coherent and some the opposite. Painters make all kinds of decisions/choices when painting (pictures of) mountains and likewise we should be aware of these cues in "painting" our music reproduction. After all, we are not actually sitting in Carnegie Hall and we are trying to re-create another place in our music room. In other words, we are obliged to make choices, just as the painter is obliged to make choices. So what makes coherent reproduction?
Last edited:
Hello David
Could you clarify steady state response?? I am thinking either an ungated impulse response or an RTA measurement. I am think length of time of the measurement is important in this context.
Thanks Rob🙂
Could you clarify steady state response?? I am thinking either an ungated impulse response or an RTA measurement. I am think length of time of the measurement is important in this context.
Thanks Rob🙂
Yes, I am aware of Toole's work and I agree. The conclusion I draw from Toole's work is that flat direct sound with basically a power response that falls off at higher frequency seems best. He also noted that the very best speakers had a power response that was monotonic, without peaks of dips, but smoothly falls off as the frequency rises above some point. The point that I think was missing was the roll of the room in this. The room absorption characteristics should be important in defining where and at what rate the power response should roll off. Ultimately we do not hear power response, we hear a combination of the direct and reflected sound, the reflected sound being a function of power and room absorption.
As far as what is more important, it depends on what you are trying to recreate in you room. If you are trying to recreate a front row center listening experience then the direct sound is of course very important and probable dominates what you hear. On the other hand if the illusion is to be 5th row, 2nd balcony, the importance of the direct sound is diminished. But I would not necessarily agree that because the reverberant field is dominate that it is ok to whack the hell out of the direct response to achieve the desired reverberant field. My preference would still be to retain flat response and have the reverberant field established by the correct matching of the room absorption with the speaker's power response. It is like I said, there is a sliding scale here. If you are in an anechoic room power response is irrelevant. As the room moves toward being more reflective, power response enters the picture.
[edit] I wanted to add that if it is the natural attenuation of high frequency with distance that is to be emulated then there may be some justification to rolling off the highs slightly. But if the highes in the direct sound roll off due to the dissipation in the air, then the reflected sound will suffer this same, or greater attenuation since it necessarily propagates a greater distance to reach the listener.
Last edited:
Funny you should mention that as I was just thinking about it. I used to do a lot of panoramic photography (thus the name) and noticed the mountains in my photos always looked short. After a good bit of experimentation I found that stretching the image to 133% in the vertical brought proportions back to were they looked "like what I saw". I tested this on many viewers without telling them what was going on. Those familiar with the view always chose the stretched version as more accurate. Even after I told them it had been stretched, they still preferred it as more realistic. This didn't work well with photos of people or houses, however!
So in this case, flat may have been accurate, but it didn't look "real". My guess is that painters use the same trick. In fact I now know they do, having worked with many. They tend to go way past 133%.
Perception is a funny thing.
Steady state is a general term for viewing anything that doesn't reach its ultimate value instantly. You don't complete the measurement until the the parameter hits a "steady state". For sound in a live room SPL increases after the source is turned on (more and more reflections arrive at your ears) until the sound decay from absorption and the sound being "replenished" are in balance.
Any measurement using pink noise and a long time window would be considered a steady state measurement. Until the advent of impulse measurements with FFT, or mlssa chirps and such, all measurements were steady state. This means there was no way to distinguish between the frequency content of the direct sound, the early reflections or the later reverberation.
Yes, an ungated impulse measurement would be steady state also.
Regards,
David
Except that mastering engineers tend to work alone without anybody else present while recording and mix engineers have the artist, producer and record company officials constantly hanging around.
The best ME are completely autonomous.
In the parlance of Toole's metric, power response as "Sound Power" is derived from anechoic measurements. The spatial averaging is not simple arithmetic, but it is fundamentally a fixed loudspeaker parameter, as opposed to spatially averaged in-room response measurements. What is the significance of this distinction?
Last edited:
- Status
- Not open for further replies.