• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Subwoofer Comparison

Kachda

Addicted to Fun and Learning
Forum Donor
Joined
May 31, 2020
Messages
903
Likes
1,578
Location
NY
Thanks.

Assuming all the tests are proper, there is not a big difference between the old PB/SB1000 and the new PRO versions, which is surprising.

1618946688798.png
 

Attachments

  • 1618946416996.png
    1618946416996.png
    40.2 KB · Views: 147

Sancus

Major Contributor
Forum Donor
Joined
Nov 30, 2018
Messages
2,923
Likes
7,616
Location
Canada
Assuming all the tests are proper, there is not a big difference between the old PB/SB1000 and the new PRO versions, which is surprising.

Is it surprising? I haven't seen subs to improve a lot over a single generation, unless there are size increases. For example if you compare the SVS SB12-NSD to the SB-2000, there was only maybe 1-2dB improvement(and the SB-2000 is a smidge larger). I'm ignoring Erin's SB-2000 Pro/SB-3000 test since it produced hard-to-believe numbers.
 

Kachda

Addicted to Fun and Learning
Forum Donor
Joined
May 31, 2020
Messages
903
Likes
1,578
Location
NY
Is it surprising? I haven't seen subs to improve a lot over a single generation, unless there are size increases. For example if you compare the SVS SB12-NSD to the SB-2000, there was only maybe 1-2dB improvement(and the SB-2000 is a smidge larger). I'm ignoring Erin's SB-2000 Pro/SB-3000 test since it produced hard-to-believe numbers.

given the similar driver size and amplifier, yeah it’s isn’t surprising. So is the only advantage of the new model the app based control and dsp engine ?
 

Sancus

Major Contributor
Forum Donor
Joined
Nov 30, 2018
Messages
2,923
Likes
7,616
Location
Canada
given the similar driver size and amplifier, yeah it’s isn’t surprising. So is the only advantage of the new model the app based control and dsp engine ?

Yeah, probably. Which, to be fair to SVS, is a useful feature for people who don't have an AVR or have low-end one that is poor at bass EQ.
 

Tom C

Major Contributor
Joined
Jun 16, 2019
Messages
1,501
Likes
1,370
Location
Wisconsin, USA
This is a super useful spreadsheet. Thank you Sweetchaos for your time and effort that benefits the whole community.
I have one question. Sorry if it's naive, but then, I'm naive.
The sheet doesn't have distortion measurements, right? Or maybe it does, but I don't fully understand what I'm looking at. Isn't that important? It seems like the way to use this to make a judgement on what you might like to purchase is to look at output versus price, and try to find a unit capable of filling your room with adequate spl. Or am I missing something?
 

Sancus

Major Contributor
Forum Donor
Joined
Nov 30, 2018
Messages
2,923
Likes
7,616
Location
Canada
The sheet doesn't have distortion measurements, right? Or maybe it does, but I don't fully understand what I'm looking at. Isn't that important? It seems like the way to use this to make a judgement on what you might like to purchase is to look at output versus price, and try to find a unit capable of filling your room with adequate spl. Or am I missing something?

The spreadsheet contains CEA-2010 results which is literally that. From the first post: "CEA-2010 is an industry standard first published in 2006 (updated as CEA-2010-A in 2012) that specifies a method for measuring subwoofer's maximum SPL (before subwoofer distorts) at various frequencies."

Erin has a good video explaining CEA-2010 testing and Audioholics also has a page on it.
 
OP
sweetchaos

sweetchaos

Major Contributor
The Curator
Joined
Nov 29, 2019
Messages
3,872
Likes
11,557
Location
BC, Canada
The sheet doesn't have distortion measurements, right?
That's correct. I'm not showing you the distortion at those measured frequencies, just the max SPL levels.

Here's how data-bass shows their CEA-2010-A measurements, for Rythmik F18:
https://data-bass.com/#/systems/5ad778f92d279a0004669a18?_k=hvix0w
1619211231013.png


All I'm doing is taking the max SPL (from 2nd column) (in decibels) and transferring it to my spreadsheet:
1619211278772.png


The 3rd column in the data-bass table shows distortion. Would these distortion figures tell you anything?
They don't mean anything to me.

Plus, not all sources show them, so it's an incomplete data set.

I know that the CEA-2010-A protocol (as explained in Erin's videos), tests each frequency to see whether the subwoofer is capable of playing at each frequency, before distortion takes over.
I guess if the distortion is too high, then the test will fail at that frequency, so as long as you see values for a certain frequency, that means the test was successful. If you see a blank value for a given frequency, it means either the test had failed or it was never tested at that frequency.
That's why I link to each source, so if you wanted, you could find out more information about that.
I personally don't care about putting 'Failed' or "Not tested" in my spreadsheet, so I left all those values blank.
 

Sancus

Major Contributor
Forum Donor
Joined
Nov 30, 2018
Messages
2,923
Likes
7,616
Location
Canada
I've seen a manufacturer make the argument that distortion measurements below CEA2010 limits are still relevant, and can be high in some subs... but they are selling a very expensive product so you know...

data-bass measures lower level distortion and even smaller, older designs like my SB12-NSD don't show distortion as high as 10% at "lower"(95dB, heh) levels until below 30hz, and you'd have to try pretty hard to convince me that was audible.

In general I think CEA-2010 is a good test and if you are worried about distortion, just buy bigger CEA2010 numbers than you think you need.
 

AdamG

Proving your point makes it “Science”.
Moderator
Forum Donor
Joined
Jan 3, 2021
Messages
4,637
Likes
14,928
Location
Reality
@sweetchaos, amazing and highly valuable work you have done Sir! Most excellent I commend your effort and the resulting product. Should help hundreds upon hundreds of ASR members refine their preferences and choices. :cool:
 

AdamG

Proving your point makes it “Science”.
Moderator
Forum Donor
Joined
Jan 3, 2021
Messages
4,637
Likes
14,928
Location
Reality
No need to thank me Sir. You truly have created a very valuable tool for ASR Members to employ in the Subwoofer selection process. We Thank you for your hard work.
 

Tom C

Major Contributor
Joined
Jun 16, 2019
Messages
1,501
Likes
1,370
Location
Wisconsin, USA
I appreciate the patience you folks have shown. I see now that what is implied in the standard is that as frequency gets very low, a progressively higher level of distortion is tolerated. When you think about it, can a person really sense distortion at all in a 10 or 15 Hz impulse? Is there any distortion in an earthquake, or a cement truck rumbling by? Probably not, but maybe some studies have shown it can, or maybe they just had to establish a definition somehow. I imagine at 10 Hz, I could tolerate 100% or multiples more. Things change a lot at 30Hz, where the standard becomes much stricter. At 125Hz, 2.7% seems like a lot of distortion to tolerate (more lenient than strict), but may be easier to accept when there are other high quality speakers present in the system reproducing the program material with less distortion (in which case the overall system SINAD would be something less than 2.7%). Is that an argument for the Geddes approach to subwoofer integration?
It seems an unfamiliar way to present the data. When a DAC or full-range speaker is tested, we’re used to seeing distortion level measured at a particular frequency with a sweep of power output levels, or a fixed output level and a frequency sweep. But in subwoofer world, you hold frequency constant, and increase the output until you hit the ceiling on a defined distortion level. It must work OK to do it that way, but it takes a little re-alignment of thought to internalize.
 

Sancus

Major Contributor
Forum Donor
Joined
Nov 30, 2018
Messages
2,923
Likes
7,616
Location
Canada
But in subwoofer world, you hold frequency constant, and increase the output until you hit the ceiling on a defined distortion level. It must work OK to do it that way, but it takes a little re-alignment of thought to internalize.

IMO this test is also the way it should work with speakers, well, in a perfect world where THD was replaced with a measure that had a solid connection to audibility anyway. S&R does testing to thresholds of distortion(10% for low bass, 3% for everything else, respectively) and I find this more useful than the fixed 86/96dB sweeps, since it actually gives you information on how loud a speaker will go before distortion becomes a big problem. However, this test requires far more data than the simplistic 86/96dB single sweeps. Truly, measuring subs is a heroic effort and we should be grateful for the data we have. %THD is an inherently weak form of information about distortion anyway so you can't use it very precisely.

8361-MAX-580x435.png


At 125Hz, 2.7% seems like a lot of distortion to tolerate (more lenient than strict)

I don't think it's a lot... There's little evidence it would be audible with music. With test tones, yes. CEA2010 isn't just about %THD either because it has different thresholds for different harmonics at different frequencies. And 118dB SPL ground plane with no room gain is absurdly loud. The Rythmik F18 only has about 1.3% distortion at 125hz/100dB SPL. Plus not many people are crossing subs that high.

The research into audibility of distortion is complicated and nobody has a good standard measure that can be related to hearing. So you kind of either take CEA2010 as good enough, or you only buy a sub off data-bass I guess because it's one of the few(only?) places with more detailed distortion measurements.

But again, you can very easily reduce risk of audible issues by buying more and bigger subs than you need. That's why I'm planning to replace my SB12-NSD with 2x Arendal 1723 2S, lol.
 
Top Bottom