Beige,
I just looked at your measurements from one of your previous postings. Essentially, you are done. Do not worry about the dip around 103 Hz, it is immaterial as it spans a very small range. If you apply 1/12th or 1/6th octave smoothing, you will see it will disappear, and as you have correctly surmised, this is related to the listening position and particularly a modal issue/modal region most likely. In my post here, you will see I have one around 50Hz, and again, it is immaterial and it does disappear if I move my listening position/mic just a few inches forward or backward.
Whatever you do, don't apply EQ to correct this at all, it's no use and most likely you cannot make it disappear without moving your listening position at which point you will have to determine if that is sonically pleasing as your imaging/staging from your main speakers may sound a little different.
I don't know how much EQ you have applied overall, but in my setup with 4 subs, the need for EQ was exclusively at 30Hz or below which makes acoustical sense. I only manipulated phase, XO slopes, XO points in the area above 30Hz. I didn't have to manipulate 'delay.' I didn't have to use PEQ's since I have enough LF damping in the room and have enough subs as well. I also tried hard to keep my XO slopes at 2nd order and not higher order (this took the most time, as it involved moving the subs to different locations, remeasuring). As Earl says, you basically use what works and gives you essentially a flat FR with perhaps a little bit of an upward tilt below 100Hz (i.e. my target curve).
Essentially I have only one formal listening seat, and that's where I measured from. After finally arriving at my configuration, I did remeasure at different listening positions, and they were quite good as well which confirms that there is uniformity of the bass response when you go with a multisub configuration.
Best,
Anand.
I just looked at your measurements from one of your previous postings. Essentially, you are done. Do not worry about the dip around 103 Hz, it is immaterial as it spans a very small range. If you apply 1/12th or 1/6th octave smoothing, you will see it will disappear, and as you have correctly surmised, this is related to the listening position and particularly a modal issue/modal region most likely. In my post here, you will see I have one around 50Hz, and again, it is immaterial and it does disappear if I move my listening position/mic just a few inches forward or backward.
Whatever you do, don't apply EQ to correct this at all, it's no use and most likely you cannot make it disappear without moving your listening position at which point you will have to determine if that is sonically pleasing as your imaging/staging from your main speakers may sound a little different.
I don't know how much EQ you have applied overall, but in my setup with 4 subs, the need for EQ was exclusively at 30Hz or below which makes acoustical sense. I only manipulated phase, XO slopes, XO points in the area above 30Hz. I didn't have to manipulate 'delay.' I didn't have to use PEQ's since I have enough LF damping in the room and have enough subs as well. I also tried hard to keep my XO slopes at 2nd order and not higher order (this took the most time, as it involved moving the subs to different locations, remeasuring). As Earl says, you basically use what works and gives you essentially a flat FR with perhaps a little bit of an upward tilt below 100Hz (i.e. my target curve).
Essentially I have only one formal listening seat, and that's where I measured from. After finally arriving at my configuration, I did remeasure at different listening positions, and they were quite good as well which confirms that there is uniformity of the bass response when you go with a multisub configuration.
Best,
Anand.
Last edited:
poseidonsvoice,
this curve is already corrected by the auto calibration system in my AVR. It's fine for movies but it's not good for music. The bass has no punch. There is clearly something going on in the room because auto calibration system is always giving me phase errors.
And because of that I'd like to try the Geddes approach and do the placement and possible EQ manually. I can keep the manual settings for music and use the auto room correction system for movies. That seems wise to me since there are 11 full range speakers in there and both, timber and time domain, are more crucial. What do you think?
this curve is already corrected by the auto calibration system in my AVR. It's fine for movies but it's not good for music. The bass has no punch. There is clearly something going on in the room because auto calibration system is always giving me phase errors.
And because of that I'd like to try the Geddes approach and do the placement and possible EQ manually. I can keep the manual settings for music and use the auto room correction system for movies. That seems wise to me since there are 11 full range speakers in there and both, timber and time domain, are more crucial. What do you think?
Last edited:
Beige,
I apologize, I had thought you had manually set everything. The problem with autocalibration is exactly the problems you are hearing. I don't mind global EQ AFTER you have set the subs in appropriate locations, and manually set parameters as I suggested in my previous post. You will find that in many situations, particularly with 4 subs, and enough LF damping that any global EQ is in and of itself, useless.
I think you should manually set things as I have done. Pretend that you do not have an autocalibration system in your AVR. And I do recommend 3 or more subs. I think you should read this as well:
http://www.hifizine.com/2011/06/bass-integration-guide-part-1/
I would read Parts 1, 2 and 3. Parts 2 and 3 are embedded as links in Part 1.
And for movie watching, I use my manually set up subs. I don't use any autocalibration which would have boosted that 50Hz area I had mentioned earlier. DIRAC, however, is getting much smarter.
My $0.02.
Best,
Anand.
I apologize, I had thought you had manually set everything. The problem with autocalibration is exactly the problems you are hearing. I don't mind global EQ AFTER you have set the subs in appropriate locations, and manually set parameters as I suggested in my previous post. You will find that in many situations, particularly with 4 subs, and enough LF damping that any global EQ is in and of itself, useless.
I think you should manually set things as I have done. Pretend that you do not have an autocalibration system in your AVR. And I do recommend 3 or more subs. I think you should read this as well:
http://www.hifizine.com/2011/06/bass-integration-guide-part-1/
I would read Parts 1, 2 and 3. Parts 2 and 3 are embedded as links in Part 1.
And for movie watching, I use my manually set up subs. I don't use any autocalibration which would have boosted that 50Hz area I had mentioned earlier. DIRAC, however, is getting much smarter.
My $0.02.
Best,
Anand.
Last edited:
Thank you for that. I will get into the reading right away!
I don't see a problem using uncalibrated (or manually adjusted) sub signal when watching movies but when it comes to full range speakers I think some EQ can help with the timber and time alignment.
I'm using Audyssey (if anything it should boost the lower area but it does not!) at this time but am eagerly awaiting for Dirac which will be capable of integrating 4 individual subs.
I don't see a problem using uncalibrated (or manually adjusted) sub signal when watching movies but when it comes to full range speakers I think some EQ can help with the timber and time alignment.
I'm using Audyssey (if anything it should boost the lower area but it does not!) at this time but am eagerly awaiting for Dirac which will be capable of integrating 4 individual subs.
Perhaps the issue is because it is "flat" and that is not the target response according to Sean Olive and Floyd Toole.
Have a look at Sean Olive's, The Subjective and Objective Evaluation of Room Correction Products. There is also a PDF download that you may want to look at.
The point is that a flat in-room measured response is not the preferred target. If you look at slide 24 you can see a downward tilting response as the most preferred by test listeners:
See the top red curve/tilt. Roughly about 20 Hz to -10 dB at 20 kHz is the slope.
If you look at slide 26, to our ears perception, the downward tilting slope is perceived by our ears as flat or neutral sounding - see red trace again:
If you eq for flat in-room response at the listening position, it is going to sound way too bright and no bass...
Can you change the target curve in the auto calibration system? I believe there is an Audyssey app for that.
Merry Xmas!
Attachments
Thank you too.
I, too, like the lower end raised a little bit and that's the beauty of manual adjustment. Altough one can always run the subs a bit hotter in case of automated room correction systems.
I had such curve in my previous room, no treatment, no EQ, no nothing. Pure luck, I guess. Then I moved the gear, added more subs and now I see how lucky I was before. So more hard work for me!
Oh, right! Merry Christmas to all!
I, too, like the lower end raised a little bit and that's the beauty of manual adjustment. Altough one can always run the subs a bit hotter in case of automated room correction systems.
I had such curve in my previous room, no treatment, no EQ, no nothing. Pure luck, I guess. Then I moved the gear, added more subs and now I see how lucky I was before. So more hard work for me!
Oh, right! Merry Christmas to all!
+1
Someday scientists might be able to explain why so few folks in the subwoofer forum seem to know this, preferring instead to stick with the naive textbook notion of flat measured FR (assuming there actually is any standard way to measure FR in a room at home).
There is also a continuation of that slope in the treble. Not sure if it has been studied with the same care (including care in pre-testing the hearing of experiment participants).
B.
Some people on the DIYMA forum were discussing the use of multiple midbasses in a car.
The concepts are the same as multiple midbasses in a room, but the frequencies are "shifted" up an octave.
For instance, in a room that measures five meters x 2.5 meters, multiple subs will smooth out the response of the octave from about 50Hz to 100Hz. In a car that measures 2.5 meters x 1.25 meters, multiple midbasses will smooth out the response of the octave from about 100hz to 200Hz.
Some observations and simulations are here:
YouTube
The concepts are the same as multiple midbasses in a room, but the frequencies are "shifted" up an octave.
For instance, in a room that measures five meters x 2.5 meters, multiple subs will smooth out the response of the octave from about 50Hz to 100Hz. In a car that measures 2.5 meters x 1.25 meters, multiple midbasses will smooth out the response of the octave from about 100hz to 200Hz.
Some observations and simulations are here:
YouTube
It's been a few years since I set up my multisub array. Currently I have 3 subs each with their own plate amps. I'm only running in stereo now so I summed the left and right channels at each sub by feeding the plate amps the speaker levels signals. This doesn't work for multi-channel, or at least it means neglecting the other channels when summing.
Anyway, I'm interested to see what options might have popped up in the last few years to simplify running a multisub array from line level signals. I'm looking at constructing some new subs fed from an amp rather than plate amps. At one point I was looking at using something like a Behringer DEQ2496 or miniDSP to sum channels and do EQ/filtering and using a pro audio amp with multiple output channels to feed the amps. Is there anything that has popped up in the last few years to do this in a more, how shall we say, dedicated fashion? Most of the sub amps I've looked at only have one or two channels of output. I guess what I’m envisioning is something that could take at least the left, right, center, and LFE channel line level signals as inputs, sum them, does some eq and filtering, and have enough outputs for 3-4 subs. It would be cool if there was an all in one box with amplification for each of the channels as well, though running something like a pro-audio amp wouldn't be too much of a problem.
edit: Maybe a little more background on this. I've kept ground loops mostly under control, although it has mean running extension cords to one or two of the plate amps since the subs are "scattered" around the room. This should be less of a concern running all the subs off a single multi channel amp and cut down some of the extra cables running around the room I'm hoping.
Dan
Anyway, I'm interested to see what options might have popped up in the last few years to simplify running a multisub array from line level signals. I'm looking at constructing some new subs fed from an amp rather than plate amps. At one point I was looking at using something like a Behringer DEQ2496 or miniDSP to sum channels and do EQ/filtering and using a pro audio amp with multiple output channels to feed the amps. Is there anything that has popped up in the last few years to do this in a more, how shall we say, dedicated fashion? Most of the sub amps I've looked at only have one or two channels of output. I guess what I’m envisioning is something that could take at least the left, right, center, and LFE channel line level signals as inputs, sum them, does some eq and filtering, and have enough outputs for 3-4 subs. It would be cool if there was an all in one box with amplification for each of the channels as well, though running something like a pro-audio amp wouldn't be too much of a problem.
edit: Maybe a little more background on this. I've kept ground loops mostly under control, although it has mean running extension cords to one or two of the plate amps since the subs are "scattered" around the room. This should be less of a concern running all the subs off a single multi channel amp and cut down some of the extra cables running around the room I'm hoping.
Dan
Last edited:
Would you guys care to take a look at these stereo measurements?
Which one would you choose for start adding the subs?
Easyupload.io - Download mdat files
Which one would you choose for start adding the subs?
Easyupload.io - Download mdat files
What's your budget? How much power per channel?
Sounds like ideally, you want a 4x4 dsp, and of course 4 amp channels.
Of 4 channel prosound amps, i think qsc's PLD series is the most affordable.
It can route any or all of the inputs to each output channel.
The outputs are also configurable, singly, bridged pairs, parallel 2,3, or 4 channels.
Presets, PC connectivity/control, rms & peak limiting, crossovers, delay, etc, but only 5 band EQ (which should be fine for subs)
Models range from 400 to 1150 W per channel @ 8 ohms.
About the only downer with them imo, is the fan..kinda loud....but about in the same league as a behringer 4 channel i owned
Sounds like ideally, you want a 4x4 dsp, and of course 4 amp channels.
Of 4 channel prosound amps, i think qsc's PLD series is the most affordable.
It can route any or all of the inputs to each output channel.
The outputs are also configurable, singly, bridged pairs, parallel 2,3, or 4 channels.
Presets, PC connectivity/control, rms & peak limiting, crossovers, delay, etc, but only 5 band EQ (which should be fine for subs)
Models range from 400 to 1150 W per channel @ 8 ohms.
About the only downer with them imo, is the fan..kinda loud....but about in the same league as a behringer 4 channel i owned
What are your measurements showing, what are the differences?
Attachments
Hi,
Thank you for looking the plots.
If you look at the plot descriptions there are more details.
1. 30 cm from the front wall, speaker ports CLOSED, left speaker distance set to 1,1 m
2. 60 cm from the front wall, speaker ports CLOSED, left speaker distance set to 1,1 m
3. 60 cm from the front wall, speaker ports OPEN, left speaker distance set to 1,1 m
4. 3. 60 cm from the front wall, speaker ports OPEN, left speaker distance set to 1,1 m, crossover at 60 Hz
5. 3. 60 cm from the front wall, speaker ports CLOSED, left speaker distance set to 1,1 m, crossover at 60 Hz
Thank you for looking the plots.
If you look at the plot descriptions there are more details.
1. 30 cm from the front wall, speaker ports CLOSED, left speaker distance set to 1,1 m
2. 60 cm from the front wall, speaker ports CLOSED, left speaker distance set to 1,1 m
3. 60 cm from the front wall, speaker ports OPEN, left speaker distance set to 1,1 m
4. 3. 60 cm from the front wall, speaker ports OPEN, left speaker distance set to 1,1 m, crossover at 60 Hz
5. 3. 60 cm from the front wall, speaker ports CLOSED, left speaker distance set to 1,1 m, crossover at 60 Hz
While Dr Geddes may be the best person to answer your question, I will say that measurement 1 and 2 are interesting. Number 1 has fewer variations in the above 100Hz region, and number 2 varies less below 100Hz. To me this suggests a more wholesome coupling of the sub and the listening position through the room. Have you considered the advice here to average your measurements?