Anechoically "flat to 20Hz" vs a 35-40 Hz roll-off

[KEW]

I am one of those who generally believe that sealed is better for music. This is from my experience more than from any science.
The best ported subs I have had a chance to listen to were dual SVS PC-12pluses. IMHO, while they were excellent for HT, they were not close to the quality of my sealed subs (Rythmik E15HP or PSA XS15se) for music!

After researching posts from Josh Ricci, it seems that the perception of "good tight musical bass" is actually most often associated with a roll-off (which is more common to sealed designs). IOW, a roll-off at 30Hz will sound "tighter" than a sub that is flat to 20Hz. I am still wrapping my head around this, and hope to set up some good tests to prove it out; but it does seem that could fit my experiences. Another factor here is that without additional processing, any sub which is flat to 20 in an anechoic chamber is pretty much assured to have overblown low bass once you put it in a room (especially if near a wall or corner) which I would assume to only make matters worse!

My personal experiences with the two following subs seem to support the idea of preference for roll-off vs flat instead of sealed vs ported.

I am a fan of the ported Dayton Audio SUB-1200 as a high value (which starts to roll off at around 40Hz), and for music would rather listen to them than the PC-12pluses!

I am not a fan of the sealed JLAudio E112's (which are flat to the low-mid 20's).

With some very rare exceptions, the music I listen to does not involve anything below 35Hz, so I am unclear on why there is such an easily noticeable difference. IOW, if the sub is not used below 35Hz why should it matter what it does at 25Hz???

 

[shadyJ]

Maybe it's worth it for you to investigate why you prefer the sound of music of with flat-to-20Hz subs to conventional sealed slopes, given that most music does not have content that low. What I would do if make a list of some tracks that you think sound better with a sealed slope rolloff, and then play them with a frequency curve that is flat vs a frequency curve that rolls off. While you play them, take screen shots of spectragrams taken from a mic recording what is happening in your room, or you could also just take some FFT recordings. Compare them and see what the difference is. One easy way to do this is use a PC-12 Plus if you still have it, and take readings of it with ports plugged vs ports open.

I would prefer subwoofers to have a default flat response and don't think manufacturers should make presumptions about the size of the user's room. The best thing to do is have the sub have a range of frequency response curves to choose from, like SVS's Ultra subs or Hsu's subs. Let the user decide what sounds best to them.

 

[KEW]

Maybe it's worth it for you to investigate why you prefer the sound of music of with flat-to-20Hz subs to conventional sealed slopes, given that most music does not have content that low. What I would do if make a list of some tracks that you think sound better with a sealed slope rolloff, and then play them with a frequency curve that is flat vs a frequency curve that rolls off. While you play them, take screen shots of spectragrams taken from a mic recording what is happening in your room, or you could also just take some FFT recordings. Compare them and see what the difference is. One easy way to do this is use a PC-12 Plus if you still have it, and take readings of it with ports plugged vs ports open.

The first thing I have in mind is to listen to a system with measured flat "in room" response (tuned via mini-DSP) to see if I find this offensive. Then add a little roll-off to see how that compares. There is some reasonable consideration that my sealed subs are actually closer to flat in-room than the ported ones tuned to be flat in an anechoic chamber, and I need to reach an understanding on that first. Josh Ricci made more or less that exact comment about the PSA XS15se in the DataBass database:

The basic frequency response shape with the low pass filter bypassed shows a response ... with a gently sloping low end that corners at 30Hz and appears to enter a sealed systems natural 12dB/octave roll off below that point. It should be a good match with the boost often seen in the low bass once placed in room.

http://www.data-bass.com/data?page=system&id=105&mset=117
(Based on his comments, I changed the OP to 30Hz roll-off (instead of 40) because my XS15se's are the best I have heard for music!)

I would prefer subwoofers to have a default flat response and don't think manufacturers should make presumptions about the size of the user's room. The best thing to do is have the sub have a range of frequency response curves to choose from, like SVS's Ultra subs or Hsu's subs. Let the user decide what sounds best to them.

Ultimately, I agree with you that that is a very good approach. Where I would part company from you is considering budget constraints. A flat response in an anechoic chamber costs money/size, and the ability to tailor the response curve is adding even more cost. The subwoofer(s) is/are often the most expensive component in a system.
If we are paying more to get un-needed bass response, then paying yet again for the ability to tame it, there is a lot to be saved by finding a sub that bypasses those costs.

What is the least expensive option you know of to fit your approach for a 16' X 22' X 9' room (my best guess at a common sized living room)?

 

[William Lemmerhirt]

Hi Ken. What kind of music are you listening to? Something purely anecdotal is that I believe some recordings contain lower sub harmonics of instruments, and also environments. This may add to the realism and overall presence. Thats all I got, not awake yet...

 

[Halon451]

With some very rare exceptions, the music I listen to does not involve anything below 35Hz, so I am unclear on why there is such an easily noticeable difference. IOW, if the sub is not used below 35Hz why should it matter what it does at 25Hz???

This. ^^ It depends on the type of music, but I believe that the reason people tend to gravitate towards sealed subs (in-home mind you, we're not talking about car audio here) is that much of the music that's out there doesn't really contain huge amounts of content at those lower audible frequencies (20Hz-ish).. so it's rather moot in most cases to have a sub that digs that deep (and could even be detrimental to some extent). I mix recorded music sometimes (I'm not a pro, I've been doing this for years as an amateur mostly for my own band's material), and I typically roll off frequencies for the lower end instruments well above the 20Hz mark because those frequencies often tend to overpower areas I need to bring forward in the mix. This is common in music production actually. Where people tend to "feel" music is a bit different than where they "feel" it for HT/movies; for example, a standard kick drum typically has its biggest punch and impact beginning in the 50-60Hz region up to maybe 80Hz depending on how the drum is tuned. This is obviously well within the limits of a well-designed sealed sub, and the sealed sub will naturally deliver that punch in a bit more of a defined, clear manner than a ported. Even the bass guitar I will place a LPF at around the 40Hz mark, sometimes even higher. For movies, the "feel" usually comes in at those really low frequencies for LFE, and things like gunshots, etc., may still hover around the same regions as the aforementioned kick drum. Now, all THAT being said - it does depend on the music.

I'll give you another example. If anyone is familiar with the Nine Inch Nails track "Discipline" this is a song that has a very tight punchy low end sound baked right into the mix. It's a phenomenal song to play on a good system. I recently cued it up on my HT system because someone else here on the forum mentioned it as a test track and I thought "duh! Forgot about that song what a great one to play!" There is a crazy low frequency sound effect sweep thing that occurs midway in the song to transition back into the final chorus. Since I do most of my music listening in my truck running a dual 10" ported sub enclosure that part of my song was always my favorite - I love hearing it dig down and sweep up back into the main chorus. Anyway, in my HT I'm running an SVS 12" ported sub - most of the song played very well, the kick drum was nice, tight, punch and it delivered. But that sweep... holy hell.. it was the first time in the hundreds of times I've heard that song that I realized that sweep actually starts waaaay sooner than I previously thought. And it just added that much more power to the transition in the song. It dug way low to get that sound, and I felt it before I actually heard it indicating the sub managed to hit a brief sub-20Hz output. I never would have experienced that with a smaller sealed sub.

Me, personally I would always opt for the ported variety for this very reason. You can often manage to tame a ported sub to deliver tighter punch for music while keeping the ability to dig that deep, and I like having that flexibility in my system when it's called upon. Granted it's sometimes a bit more challenging to find that balance and content and mixes vary so wildly it's almost total trial and error at times to find the right combination of placement/settings/EQ that sounds best for most content (unless you just like fiddling around with the controls between every source, I do not).

*Edit: Granted my above examples are entirely anecdotal to be fair. Your experience may vary of course. So in the spirit of the original thread you started under the 'General A/V' section.. "it depends."

 

[TLS Guy]

This is I think all due to issues of time and transient response. These time errors interact with rooms.

Unfortunately there is far too much reliance on frequency response alone, and no where near enough attention to time. There has been far too much nonsense that time in audio systems matters little. In the bass decade I think this is far from true. Remember you could produce a perfect frequency plot with a monstrous time delay, that would render the system totally unlistenable and not know it from the frequency plot.

There has been a lot of discussion about bass and subs lately, so this has prompted me to study this issue in more depth.

I will post when I have made more observations. What I do know is that I have no trouble getting realistic deep bass in a variety of rooms without getting issue of boom and a boat load of speaker room interaction.

I'm pretty sure a big part of the problem is the massive abuse and misuse of all types of subs. Then the whole issue of subs covering up for less than adequate speakers is a totally different can of worms.

 

[TLS Guy]

I should amplify my remarks a little more to get people thinking about the issue.

There is no such thing as fast bass. In other words a low long wave low frequency will never have a sharp upstroke. To get that implies the addition of high frequency content.

The bottom line is that the system must have an excellent transient response to have natural realistic reproduction.

So having speakers of different types from different manufacturers, with a one fits all crossover order (bass management) and subwoofers placed in many cases in a steady state, it is small wonder there is chaos.

I have said this before, and I'll say it again, that an off the self crossover is unlikely to do the trick, unless the system is built for that crossover. Bass management is a crossover and almost always a combination of acoustic and electrical roll off orders.

So it is small wonder there are problems. Once you start moving any sound sources around the room (subs) then you are going to get differing time paths between fundamentals and harmonics all over the room.

Priority has been given to steady state frequency response and time relationships have been ignored.

Small wonder we have a boat load of problems with results falling short of expectations.

 

[highfigh]

My personal experiences with the two following subs seem to support the idea of preference for roll-off vs flat instead of sealed vs ported.

I am a fan of the ported Dayton Audio SUB-1200 as a high value (which starts to roll off at around 40Hz), and for music would rather listen to them than the PC-12pluses!

I am not a fan of the sealed JLAudio E112's (which are flat to the low-mid 20's).

With some very rare exceptions, the music I listen to does not involve anything below 35Hz, so I am unclear on why there is such an easily noticeable difference. IOW, if the sub is not used below 35Hz why should it matter what it does at 25Hz???

There's no brick wall below the tuning frequency and if you make something vibrate at one frequency, it will usually vibrate at others, too. It's a mechanical device, not an electronic oscillator. Also, you may have forgotten about 'beat frequencies', which occur when two notes are played and they interact- the beat frequency is the difference between them, so it can be any frequency, but the speaker needs to be able to react to it. Below the port tuning frequency, the cone's motion isn't well controlled and that's one of the reasons so many subwoofers fall flat on their face- bad design.

A sealed box does what it can, a vented box has a port that's really doing one note well, but it's not only going to do one note because the characteristics of a resonator doesn't dictate that behavior.

http://hyperphysics.phy-astr.gsu.edu/hbase/Audio/basref.html

 

[shadyJ]

This is I think all due to issues of time and transient response. These time errors interact with rooms.

Unfortunately there is far too much reliance on frequency response alone, and no where near enough attention to time. There has been far too much nonsense that time in audio systems matters little. In the bass decade I think this is far from true. Remember you could produce a perfect frequency plot with a monstrous time delay, that would render the system totally unlistenable and not know it from the frequency plot.

There has been a lot of discussion about bass and subs lately, so this has prompted me to study this issue in more depth.

I will post when I have made more observations. What I do know is that I have no trouble getting realistic deep bass in a variety of rooms without getting issue of boom and a boat load of speaker room interaction.

I'm pretty sure a big part of the problem is the massive abuse and misuse of all types of subs. Then the whole issue of subs covering up for less than adequate speakers is a totally different can of worms.

At data-bass, there is a guy investigating the importance of the effects of first-time arrival response on low-frequency sound. It is an interesting line of questioning, but it needs some actual research. It sounds to me like you are thinking along the same lines as that guy. I think there could be something to it, but I do think getting a flat response is paramount. I think frequency response deviations will be more audible than very slight delays in time arrivals in subwoofer band frequencies.

 

[Irvrobinson]

I'm pretty sure a big part of the problem is the massive abuse and misuse of all types of subs. Then the whole issue of subs covering up for less than adequate speakers is a totally different can of worms.

What do you think about using subs to increase bass smoothness in a room? Use or abuse?

 

[Pogre]

I'm too much of a newb to have enough experience to weigh in. My only experience with ported were a pair of Klipsch 10" subwoofers. My sealed SVS' blow them away. Not really a fair comparison though.

I am in the market for some new subs (got a diy project list together) and decided to go with a pair of 15" sealed subs, though ported isn't out of the question if the price is right. Either way I'm keeping an eye on this thread.

 

[shadyJ]

If we are paying more to get un-needed bass response, then paying yet again for the ability to tame it, there is a lot to be saved by finding a sub that bypasses those costs.

The problem here is that you can't really be certain what the room will do to the bass. Sure, if you are in a small sealed room, you can count on some low frequency gain, and in that case, get a sealed sub- if a flat response in room is what you are after. But, of course, such a room is not always the case, and there is also those who might prefer a boost in deep bass anyway. With respect to low frequency headroom, I would rather have it and not need it than need it and not have it.

What is the least expensive option you know of to fit your approach for a 16' X 22' X 9' room (my best guess at a common sized living room)?

Pressure vessel gain typically starts where the longest dimension of the room meets one half the frequency wavelength, so for a 22' long room, the 1/2 wavelength is about 25 Hz, so maybe that is where you can expect pressure vessel gain to begin to compound low frequency output. What option will suffice depends on what the listener wants, like how loud, how low, how flat of a response. You will have to give me more specific criteria.

 

[KEW]

There has been a lot of discussion about bass and subs lately, so this has prompted me to study this issue in more depth.

I will post when I have made more observations.

Fantastic!

I know this is not an overnight process, but I also know you have the expertise and resources to do this well.

Thank you Dr. Mark!!!

 

[ski2xblack]

TLS's thoughts remind me of something Barry Ober told me years ago regarding bass reproduction, subs, and mains. He was at M&K at the time, has since worked at JL as well as doing other things. He was a stickler for paying attention to the time as well as frequency domain, and a strong proponent of sealed systems (both subs and mains) for their superiority in that area. I tend to agree, there seem to be fewer opportunities for mistakes that way. I don't tend to care much for the deep stuff as much as getting the crossover to the mains right, which is tough enough.

 

[TLS Guy]

At data-bass, there is a guy investigating the importance of the effects of first-time arrival response on low-frequency sound. It is an interesting line of questioning, but it needs some actual research. It sounds to me like you are thinking along the same lines as that guy. I think there could be something to it, but I do think getting a flat response is paramount. I think frequency response deviations will be more audible than very slight delays in time arrivals in subwoofer band frequencies.

Not so in the last two octaves. The wavelengths are long. So you can have a smooth freqeuncy response and excite room modes repeatedly where otherwise they would have dies away.

The question must be repeatedly asked as to why most systems recording a piano and then playing it back, the bass decade will be far form the real thing. I can assure that does not have to be the case but usually is.

 

[TLS Guy]

What do you think about using subs to increase bass smoothness in a room? Use or abuse?

Can't tell. It might help, but I think you can't rely on smooth response as the only benchmark. I'm pretty sure response and time have to be considered together. There are just so many areas in the bass where huge wavelength discrepancies arise. and that is time.

Crossover time shifts.

Placement time shifts. (Remember when you move a sub away from other sources of sound there will be different delays for pretty much every listener in the room.)

Group delay in the speaker design.
Those are the main ones. The other thing is that phase shift and time shift are identical ways of expressing the same thing.

I'm leaving for our Eagan residence now and so will be off "the string" for a while.

 

[KEW]

The problem here is that you can't really be certain what the room will do to the bass. Sure, if you are in a small sealed room, you can count on some low frequency gain, and in that case, get a sealed sub- if a flat response in room is what you are after. But, of course, such a room is not always the case, and there is also those who might prefer a boost in deep bass anyway. With respect to low frequency headroom, I would rather have it and not need it than need it and not have it.

Pressure vessel gain typically starts where the longest dimension of the room meets one half the frequency wavelength, so for a 22' long room, the 1/2 wavelength is about 25 Hz, so maybe that is where you can expect pressure vessel gain to begin to compound low frequency output. What option will suffice depends on what the listener wants, like how loud, how low, how flat of a response. You will have to give me more specific criteria.

Let's take our discussion to the link below to avoid diluting the more generalized emphasis occurring here:
http://forums.audioholics.com/forums/threads/room-gain-and-subwoofers.106690/

 

[yepimonfire]

I am one of those who generally believe that sealed is better for music. This is from my experience more than from any science.
The best ported subs I have had a chance to listen to were dual SVS PC-12pluses. IMHO, while they were excellent for HT, they were not close to the quality of my sealed subs (Rythmik E15HP or PSA XS15se) for music!

After researching posts from Josh Ricci, it seems that the perception of "good tight musical bass" is actually most often associated with a roll-off (which is more common to sealed designs). IOW, a roll-off at 30Hz will sound "tighter" than a sub that is flat to 20Hz. I am still wrapping my head around this, and hope to set up some good tests to prove it out; but it does seem that could fit my experiences. Another factor here is that without additional processing, any sub which is flat to 20 in an anechoic chamber is pretty much assured to have overblown low bass once you put it in a room (especially if near a wall or corner) which I would assume to only make matters worse!

My personal experiences with the two following subs seem to support the idea of preference for roll-off vs flat instead of sealed vs ported.

I am a fan of the ported Dayton Audio SUB-1200 as a high value (which starts to roll off at around 40Hz), and for music would rather listen to them than the PC-12pluses!

I am not a fan of the sealed JLAudio E112's (which are flat to the low-mid 20's).

With some very rare exceptions, the music I listen to does not involve anything below 35Hz, so I am unclear on why there is such an easily noticeable difference. IOW, if the sub is not used below 35Hz why should it matter what it does at 25Hz???

The dayton 1200 doesn't roll off that high. If you've measured and you're getting a roll off like that it's likely that 40hz is your room mode and that's making it appear to roll off higher. Even my sub 1000 is flat to 30hz. I had the same problem in a small 12x12 room, two dominate room modes were 40 and 58hz, because of the size I had a 15dB spike centering around 50hz, making everything, including the sub 1500, appear to roll off steeply 45hz. I moved the sub to a larger room that was much longer than it was wide and voila, I get a flat response down to 23hz.

I've never felt like sealed subs were more musical than ported. The only reason they would be is if the ported sub was designed poorly, for example, too high of a port tuning or too small of a cabinet causing one note bass, or too small of a port inducing port wind noise.

A sealed sub may roll off gentler, but keep in mind humans are not very sensitive to bass at lower volumes, unless someone is listening at something like 90dB or above. If a sub rolls off at the 40hz but has 20hz 12dB down, it's likely going to be below the threshold of hearing. The whole fast vs slow sub argument is a waste of time. A well designed sub with a stiff driver will not have resonance issues and long decay times, regardless of whether it's ported or not. Subwoofer are intentionally designed to be slow. A stiff suspension and thin lightweight driver would both increase the -3dB point, and induce ringing and cone breakup at low frequencies, that's why regular speakers have greater distortion at low frequencies, regardless of what their f3 is.

Sent from my SM-G360T1 using Tapatalk

 

[KEW]

The dayton 1200 doesn't roll off that high. If you've measured and you're getting a roll off like that it's likely that 40hz is your room mode and that's making it appear to roll off higher. Even my sub 1000 is flat to 30hz. I had the same problem in a small 12x12 room, two dominate room modes were 40 and 58hz, because of the size I had a 15dB spike centering around 50hz, making everything, including the sub 1500, appear to roll off steeply 45hz. I moved the sub to a larger room that was much longer than it was wide and voila, I get a flat response down to 23hz.

I've never felt like sealed subs were more musical than ported. The only reason they would be is if the ported sub was designed poorly, for example, too high of a port tuning or too small of a cabinet causing one note bass, or too small of a port inducing port wind noise.

A sealed sub may roll off gentler, but keep in mind humans are not very sensitive to bass at lower volumes, unless someone is listening at something like 90dB or above. If a sub rolls off at the 40hz but has 20hz 12dB down, it's likely going to be below the threshold of hearing. The whole fast vs slow sub argument is a waste of time. A well designed sub with a stiff driver will not have resonance issues and long decay times, regardless of whether it's ported or not. Subwoofer are intentionally designed to be slow. A stiff suspension and thin lightweight driver would both increase the -3dB point, and induce ringing and cone breakup at low frequencies, that's why regular speakers have greater distortion at low frequencies, regardless of what their f3 is.

Sent from my SM-G360T1 using Tapatalk

Interesting, I had not really thought much of in-room peaks causing an early roll-off. Makes sense!

I was sloppy to say 40Hz (35Hz is better), but used Jman's measurements for the Dayton SUB-1200 (below). This is before he was combining the port and the driver measurements into one curve, so we have to make some inferences, but the port starts to roll-off ~35Hz and the driver is falling fast at ~35Hz.

I have only heard the 1200. Do you feel the 1000 and the 1500 are equivalent designs, just with more (or less) extension/output, as expected from the size of the driver?

 

[William Lemmerhirt]

Interesting, I had not really thought much of in-room peaks causing an early roll-off. Makes sense!

I was sloppy to say 40Hz (35Hz is better), but used Jman's measurements for the Dayton SUB-1200 (below). This is before he was combining the port and the driver measurements into one curve, so we have to make some inferences, but the port starts to roll-off ~35Hz and the driver is falling fast at ~35Hz.

I have only heard the 1200. Do you feel the 1000 and the 1500 are equivalent designs, just with more (or less) extension/output, as expected from the size of the driver?

I doubt that peaks cause early roll off. A peak in a higher freq range may just give the illusion of an earlier roll off. I don't see how a peak would change the lower response.