Best sounding sub for $1,000

[WmAx]

I just have to interject....

For about $1000, you can DIY a sub that is better than any commercial unit at any price....

Or spend $1000 and get a 'decent' commercial unit.

Just an example....

JL Audio W7 12": about $500 from a discount retailer.

Behringer EP2500 amplifier(2000 clean watts, mono): $350

Behringer BFD parametric EQ: $100

Building material plus finish: $50-$100

Use a sealed design, and it will compare closely with the Fathom F113, a very respective high end commercial design. Build using a properly engineered ported system, it will have superior output to 2 x F113 units, and just about anything else on the market. I will glady help with proper cabinet design to extract full performance if requested.

You could even have the enclosure made by a local cabinet maker, and then just install the subwoofer in the completed cabinet.

-Chris

 

[Inertia]

gus6464:
What is group delay exactly? I have never heard this term!

 

[bandphan]

With a eq/crossover unit like the BFD or DCX from berhinger (insert other brands) you can tame the subs response to meet your liking and rooms needs.

 

[avaserfi]

gus6464:
What is group delay exactly? I have never heard this term!

Group delay is actually related to frequency response. It is the negative derivative of phase (which correlates with frequency response) hence the correlation between group delay and frequency response.

Simplified: It is how delayed some frequencies are relative to others when being reproduced, in this case by a subwoofer.

In the end, group delay rarely has real effect on audibility, rather the frequency response that shapes group delay is what typically causes audible differences. Due to this there are far more important measurements to look at such as driver linearity and power compression as well as frequency response.

 

[croseiv]

I must admit that I feel educated. This thread has some good information in it! I've seen the group delay before, but was not clear on its meaning.

 

[gus6464]

The fact of the matter is that when it comes to group delay some will say it doesn't mean anything (asaverfi), and some will say it does (me). In reality if group delay really went hand in hand with type of box used (ported, sealed, passive radiator), then why are there sealed subs out there with worse group delay measurements than ported ones. Now you always hear about people preferring sealed subs for music than ported because they find them to be more accurate and punchy. As you can see that is not always the case as a sealed sub can have just as high of a group delay than ported ones. But in the end it is up to you to decide if it really matters to you or not. All I know is that the subs with some of the lowest group delays (Rhythmic Servo 12, JL F112) are the ones that most people always refer to as being very musical. I for one believe that group delay is the closest thing we have in order to measure accuracy and speed of a sub so I always care about that measurement.

 

[haraldo]

I just have to interject....

For about $1000, you can DIY a sub that is better than any commercial unit at any price....

Or spend $1000 and get a 'decent' commercial unit.

Just an example....

JL Audio W7 12": about $500 from a discount retailer.

Behringer EP2500 amplifier(2000 clean watts, mono): $350

Behringer BFD parametric EQ: $100

Building material plus finish: $50-$100

Use a sealed design, and it will compare closely with the Fathom F113, a very respective high end commercial design. Build using a properly engineered ported system, it will have superior output to 2 x F113 units, and just about anything else on the market. I will glady help with proper cabinet design to extract full performance if requested.

You could even have the enclosure made by a local cabinet maker, and then just install the subwoofer in the completed cabinet.

-Chris

I just have to say.... I admire your dedication to help people in this forum.....

 

[haraldo]

The fact of the matter is that when it comes to group delay some will say it doesn't mean anything (asaverfi), and some will say it does (me). In reality if group delay really went hand in hand with type of box used (ported, sealed, passive radiator), then why are there sealed subs out there with worse group delay measurements than ported ones. Now you always hear about people preferring sealed subs for music than ported because they find them to be more accurate and punchy. As you can see that is not always the case as a sealed sub can have just as high of a group delay than ported ones. But in the end it is up to you to decide if it really matters to you or not. All I know is that the subs with some of the lowest group delays (Rhythmic Servo 12, JL F112) are the ones that most people always refer to as being very musical. I for one believe that group delay is the closest thing we have in order to measure accuracy and speed of a sub so I always care about that measurement.

Very often I have found that I prefer performance of sealed sub-bass system, because to my experience I have found the bass to be better more controlled and much more like real life music. I have always believed that this is because of the physical properties of how a sealed enclosure performs as opposed to ported enclosures, in general....
And also how group delay and phase shifts are different in sealed vs ported systems.....

Now through many other posts in this forum by WmAx and Avaserhifi I understand that this is totally uncorrect??? yes???.... and that these observations only relate to improperly tuned ported systems, and that a properly tuned cabinet will outperform any sealed system.

So I'm now to the point that.... I thought I understood, but now I don't understand anything anymore, really.....

By reading more posts here, hopefully I will understand toooooo in the end.....

Guess all of us will not agree anyways, but we will hopefully learn along the way.....

 

[WmAx]

The fact of the matter is that when it comes to group delay some will say it doesn't mean anything (asaverfi), and some will say it does (me). In reality if group delay really went hand in hand with type of box used (ported, sealed, passive radiator), then why are there sealed subs out there with worse group delay measurements than ported ones.

The difference between avaserfi(or me if you prefer) and you, is that avaserfi has used blinded testing under bias reduced situations to make his claim, where as you are depending on low reliablity sighted listening and speculative evidence.

I have put together, myself, careful ABXed control samples with group delay of 0 to the high levels typical of a ported 4th order system at 30Hz. When one bothers to reduce the variables to such an isolated incidence, and then do the listening tests with no knowledge of what is being heard(ABX randomization software), then a little bit of useful information can be gained.

My conclusion, gained with use of controlled listening tests, is that group delay has little if any relevance on bass sound quality in itself. Look to other variables to blame.

-Chris

 

[avaserfi]

The fact of the matter is that when it comes to group delay some will say it doesn't mean anything (asaverfi), and some will say it does (me). In reality if group delay really went hand in hand with type of box used (ported, sealed, passive radiator), then why are there sealed subs out there with worse group delay measurements than ported ones. Now you always hear about people preferring sealed subs for music than ported because they find them to be more accurate and punchy. As you can see that is not always the case as a sealed sub can have just as high of a group delay than ported ones. But in the end it is up to you to decide if it really matters to you or not. All I know is that the subs with some of the lowest group delays (Rhythmic Servo 12, JL F112) are the ones that most people always refer to as being very musical. I for one believe that group delay is the closest thing we have in order to measure accuracy and speed of a sub so I always care about that measurement.

I have no interest in endlessly debating this topic. There are perceptual papers that address the audibility of group delay while not definitive at such low frequencies the seemingly accepted value for group delay audibility (which in all honest seems fairly stringent) is one cycle. Virtually all subwoofers have a group delay within this range, especially at the price points concerned within this thread.

As I previously mentioned in post 24 group delay is correlated to frequency response. The natural musicality found in the subwoofers you listed can be attributed directly to their frequency response and natural roll off slope. Such aspects can be easily recreated via proper use of equalization.

Let us take, for example, the SVS SB12-Plus and SVS PB12-NSD discussed previously in this thread.

Frequency response is shown followed by group delay:

If the group delay is compared between the two units it certainly seems as if the ported box has a superior measurement in certain areas, specifically 30Hz-50Hz range. Now if the frequency response plots are analyzed at these points it is clear the ported box has a more flat response in this region. This flatter response directly effects phase which is where group delay is derived. The group delay itself plays no part in the perception of musicality it is just a poor, indirect, measure of frequency response.

The reason ported subs often, not always, have higher group delay than sealed subs is because of their extended bass shelf and different natural roll off relative to sealed subwoofers. These factors also contribute to perceived boomy nature of sealed subs (although likely not as much as room interaction) and can be mitigated through proper use of a single shelving filter.

 

[MinusTheBear]

A question about group delay was posed on another forum which I read (coincides with avaserfi)

I posed the question to Tom Cumberland, Axiom's chief R & D engineer, who answered me this way:

"The issue of group delay can best be explained by moving a sub one to two feet in a given direction. This being said, there is group delay in everything that relates to room placement. Whether a sub is ported or sealed only affects its overall frequency response. Of course, a smooth linear response is the optimal performance from any loudspeaker. Simply moving a subwoofer around the room creates group delay. This movement, of course, helps remove the nulls and peaks in the frequency response in the room whether it is one sub or multiple subs."

 

[gus6464]

Then explain to me this. Here are 2 extremely linear subs up to 20Hz. Rhythmic Servo 12" and PB13-Ultra.

FR

Power compression

Group Delay

As you can see both subs are pretty much flat between 20-50Hz and within their limits perform perform pretty much the same. Yet the group delay on the sealed sub is considerably lower than the ported PB13. If group delay was a true derivative of FR then they should for the most part match. Now you can say the PB13 has way more output which I do not deny but I am specifying that I say within the limits of both.

 

[mike c]

i'm surprised nobody checked the "how to interpret these tests" from the testers themselves.

AVTALK

Group Delay.

This is a measure of how the sub delays certain frequencies with respect to others. Typically the extreme bass will arrive at your ears (or our test mic) a few milliseconds after the rest of the frequencies and the extent to which this happens varies from sub to sub. The smaller and more consistent the group delay is across the frequency range, the better the sub should sound. If group delay becomes excessive it may be audible as 'slowness' in the bass or muddiness. For example, if the bass associated with a bass drum beat arrives after the 'thwack', it may sound ponderous and slow. If they arrive more or less in sync then it will sound 'tight'. 8)

How much group delay is audible? Opinions are divided, but a good rule of thumb is that a time delay of between 1 and 1.5 cycles is just audible. So at 50Hz, group delay should be kept below 20ms (1000/50) and at 20Hz, it should be below 50ms (1000/20) to guarantee inaudibility. 8)

HTS

Group delay

Mathematically group delay is the negative derivative of the phase response. Phase response was measured using the ETF-5 program which calculates it from the measured impulse response. Since frequency response and phase response correlate with each other, also FR and GD correlate with each other. Meaning flat or gently sloping FR results in low GD. That's why the final in-room frequency response should be as flat as possible. There are no studies defining the audibility of GD at low frequencies, but the threshold is suggested to be in the range of 1-1.5 cycles. Therefore I have included the 1 cycle curve on the graphs.

Ideally GD should be as low as possible, or at least gradually/gently rising.

basically, 1 cycle is the black line. anything below the black line is "ok"

 

[Sheep]

Then explain to me this. Here are 2 extremely linear subs up to 20Hz. Rhythmic Servo 12" and PB13-Ultra.

FR

Power compression

Group Delay

As you can see both subs are pretty much flat between 20-50Hz and within their limits perform perform pretty much the same. Yet the group delay on the sealed sub is considerably lower than the ported PB13. If group delay was a true derivative of FR then they should for the most part match. Now you can say the PB13 has way more output which I do not deny but I am specifying that I say within the limits of both.

Interesting, you left out Distortion. Don't worry, I've got your back.

PB-13Ultra:

Rythmik 12:

Extremely linear for the SVS, not the Rythmik. The distortion spike is more problematic then the acceptable GD curve.

SheepStar

 

[gus6464]

i'm surprised nobody checked the "how to interpret these tests" from the testers themselves.

AVTALK


HTS


basically, 1 cycle is the black line. anything below the black line is "ok"

So mike since you said anything below the black is just "ok", if you were looking for a sub for 100% music listening so accuracy and speed is most important, which would you choose between the rhythmic and PB13 ultra?

 

[gus6464]

Interesting, you left out Distortion. Don't worry, I've got your back.

PB-13Ultra:

Rythmik 12:

Extremely linear for the SVS, not the Rythmik. The distortion spike is more problematic then the acceptable GD curve.

SheepStar

Once again I did say comparing the subs within their capable limits. At 90dB output which is about the norm for most people, both subs will have no audible distortion.

 

[Sheep]

Once again I did say comparing the subs within their capable limits. At 90dB output which is about the norm for most people, both subs will have no audible distortion.

The SVS was still in it's capable limits. It had little to no power compression unlike the Rythmik, and it is had less distortion. Isn't that what a servo is for anyway? Looks like LLT subwoofers are still the best way to go. I once heard someone say this about Servo's: "They only benefit poorly designed drivers, as a properly designed driver would gain nothing from servo feedback."

SheepStar

 

[mike c]

out of all the subs i have, i probably would NEVER be able to pick one out in a blind test playing music. well there's one i could - my JBL HTIB sub.

so honestly IMO, looking for SQ differences in subs is really about as easy as finding SQ differences between amps and cables.

to answer your question: ultra. based purely on the power compression and THD charts.

edit: i define accuracy to be ... linear FR and no audible THD in adequate SPL levels

speed is in the same category as chocolatey to me.

 

[avaserfi]

It seems there is a misunderstanding of how ported subwoofers actually work. In the case of a sealed aligment all the sound is coming from a single source, the driver itself. With a ported alignment the sound emanates from the driver to point (above tuning) and the port picks up where the driver leaves off, when acting together in a properly designed system this results in a flat response. These two slopes driver/port interaction are not linear when looked at alone. This creates non-linear phase and thus non-linear group delay (as noted all based on these slopes in frequency response where group delay and phase are derived!).

Here is an example of a nearfield measurement of a driver (purple) and a nearfield measurement of a port from the same subwoofer (orange) in room response is flat well below 20Hz:

Despite the subwoofers overall flat response the subwoofer/port interaction individually is not flat creating a higher group delay than typical sealed subwoofers as shown. Despite this the group delay is not audible, but the differences in frequency response between sealed and ported subwoofers are.

I will make it simple for anyone who is interested in verifiable correlation of perception to measurements on the subject of subwoofer speed: Group delay has virtually no bearing on the subject especially with regard to music playback, rather it is frequency response (with special focus on extension and linear response) as well as roll off rate of the subwoofer.

 

[Gimpy Ric]

Will there be a quiz on all the material covered ? Andrews getting pretty deep. But I appreciate the information.