Product reviews

[Moog]

There is a review of the paradigm reference stuidio 12 subwoofer on page 42 of the March edition of the AVSA. In the " vital statisctics",  section ,the claimed frequency response is not even stated.  This is for a sub that costs R22 500. One can argue that the most the important statisctic for a sub is the low frequency response and the amplifier power. Why is this specification not published/disclosed, especially for a product that claims to be a "reference" product?

[joel]

The claimed response isn't part of the information we regularly put into our sub reviews, but it certainly could be included.
However the actual importance of claimed response can often be of dubious value as I've found response claims down to extremely low frequencies only to find when digging into the (very) fine print that this was at -16dB.

This being said, a few lines into the review I did mention that paradigm claim response down to 16Hz and amplifier power was listed so there was disclosure from the magazines part.

Amplifier power too is extremely misleading as more often than not the important specification of driver efficiency is omitted and without knowing this there is no way of telling if a 1000 watt sub will actually play louder than a 100 watt sub. Big power does however look great on a spec sheet.

Hope this answers your question, and if not please feel free to post again and I'll try again (it's getting late and the brain went to bed a long time ago). Anyone is also welcome to PM me or as long as it's not too early/late call me directly on 082 772 6223, and I'll answer any question as honestly as I can.

Joel

 

[Moog]

Your comments are noted Joel. Why not disclose the claimed specifications by the  manufacturer so that a reader can  make their own conclusions.
(even when the db limits are specified. Many manufacturers are guilty of this.
Again, when you go to page 50 of the same publication, where a review of the Dali Concept 10 speaker is published, you did disclose  the frequency response of the speaker as 25hZ to 25 kHz ( no decibel limits are specified) , which i presume is a claim from the manufacturer, thus contradicting your statement that that you will refrain from publishing manufacturers claims regarding power and frequency response. In addition to this, if you go to Dali's website( http://www.dali.eu/display_content.php/INT/speakers.html/150/1104) the claimed freqency response is 35hZ to 25kHz +- 3dB . Some consistency will also help.

[joel]

Hi Moog

I re-read my post and couldn't find where I said that we will refrain from listing power and frequency response . What i did say was that frequency response isn't part of the information that we regularly put in for SUB reviews. I also added that it Could be added. Power claims on the other hand are listed.

As far as the accuracy of the specs goes, we list the specs we find (and tolerance if given) and the distributors are then asked to check the specs for accuracy.

If you like i can send you our spec list and you are welcome to recommend any changes or additions. Just bear in mind that space is a concern so not every spec can be listed.

[Vaughan]

Frequency response specs are made almost entirely redundant without knowing the SPL measurements (since SPL varies with frequency and all that). The problem is that those measurements usually are never disclosed and if they were it would still be meaningless unless they revealed how the test was conducted (read : unlikely).

Heck, even if the testing was fully disclosed it wouldn't mean a thing because your room (size, shape, acoustics) will determine the response. Even a 'puny' Jamo Sub200 subwoofer which is low cost has measured response down to 20 Hz in-room (in my room). However output at 20 Hz is insignificant and hence made redundant. But it's nice seeing 20 Hz on the response graph. 

Just my 0.2c.

Regards,

[Moog]

Frequency response figures can not be redundundant if the decibel limits are specified, even if room variations will have an impact on the freqency response. This spec will inform the user about how flat the freqquency response is claimed to be, before taking into account room effects. If the manufacturer has disclosed that spec, why should a reviewer not pass on that information so that readers can judge for themselves.

[Ampdog]

I agree with both viewpoints. Nobody can get into trouble by quoting manufacturer's specs.  If on the other hand in practice they are pretty well meaningless as Goneten suggested (and I agree), the quoting of specs can itself be misleading. 'Heads you win, tails I lose' sort of thing. (In amplifiers something similar occurs with given thd figures.) I did not see how often subs are being reviewed - it might be useful to state somewhere in the review why frequency specs are not given, as a standard sentence. I would slightly edge toward this practice. For most readers these days it is almost easier to Google the make anyway and get more 'specs' that way.

[Vaughan]

Frequency response figures can not be redundundant if the decibel limits are specified, even if room variations will have an impact on the freqency response.

It's a numbers game. The room will skew the results to such a degree you'll wonder what the fuss was all about in the first place. That impressive sub with a claimed -3dB 14 Hz-80 Hz spec will, in all likelihood, turn into -25 (to 35 dB or worse) spec in a room. Remember, this is all assuming that the claimed response figures are accurate and that's the problem; not all claims have a measure of truth to them.

This spec will inform the user about how flat the freqquency response is claimed to be, before taking into account room effects.

Sure. The spec will inform how flat it is in an anechoic chamber. No one listens to a speaker or subwoofer in an anechoic chamber. But it gets worse. If you go to HomeTheaterShack you should find a number of independent subwoofer measurements conducted by Illka (who is one of the most knowledgeable subwoofer testers I know and is now a top engineer at JL Audio) who measured a number of subwoofers ground plane, outside away from any reflecting structures and more than a few subwoofers were no where near as flat as originally claimed by the manufacturers. 

Regards,

[Moog]

Well, let us look at this situation like this. Let us assume speaker A has been tested and found to exhibit a - 3db response of 25 hZ in the anechoic chamber and speaker B has a response of 35 hZ at with the same input voltage/current and both  have the same sensitivity across the frequency range, the same number of speaker drivers of the same size. Can you explain how will the superior low freqency response of speaker A will deteriorate when compared with speaker B?

Even if it a numbers game, as a reviewer , one should publish the specs in a similar way.  If you look at my second post on this thread you will notice that a review of the subwoofer does not include the frequency response spec but a review of a  certain speaker does include a frequency response spec, despite the fact that a less optimistic spec  is disclosed at the manufacturers website.

To me these are two different issues and i can summarize them  as follows:

1) Consistency in a  review (so far i do not think my forumites (is there such a word) have addressed this and:
2) The need to disclose all manufacturers specs ( My view is that a reviwer should disclose this since no testing has been done to prove that  the specs are met or not). Surely it is not practical for a reviewer to take into account romm effects unless he or she can measure a speakers's response in a speciic room and also disclose the room dimensions  for the sake of informing readers.

[Vaughan]

Let us assume speaker A has been tested and found to exhibit a - 3db response of 25 hZ in the anechoic chamber and speaker B has a response of 35 hZ at with the same input voltage/current and both  have the same sensitivity across the frequency range, the same number of speaker drivers of the same size. Can you explain how will the superior low freqency response of speaker A will deteriorate when compared with speaker B?

Assuming both speakers were tested in the same anechoic chamber using identical testing methodologies then speaker A would exhibit a superior bass response. Unless, of course, speaker A suffered thermal compression earlier at higher levels which would alter the response somewhat. But hey, that's nitpicking.

My view is that no matter how flat the sub is measured to be, it will always be at the mercy of the room and the rooms influence on the response is large enough to render anechoic or ground plane measurements to be useless. Even in the above example, if your room had a wide depression let's say around 25-50 Hz then what good would the anechoic measurements serve ? Your room would have taken a near perfect  response and butcher it to the nth degree. That's what almost all rooms do; they butcher the response in every conceivable way.

As to why the published specs weren't published, who knows. I see no reason why they couldn't have been published. Personally, I take sub specs with a grain of salt unless independently measured by a trustworthy source. But hey, that's just me.

Regards,

[Shonver]

I see Goneten also mentioned SPL earlier on. I guess he means maximum SPL (right, Vaughn?). Apart from compression effects, there is also the question of headroom: some subs are equalised (especially those that are sealed); equalisation can chow up heaps of amplifier headroom. So, at a specific output level, things may look flat as a pancake, but then it tells us nothing about whether it can sustain that performance at higher output levels. The design might even include a limiter/compressor that limits maximum output at the lower frequencies (that is usually where it runs out of juice), whilst still allowing the upper frequencies to increase in output level. So, across its claimed bandwidth it does not maintain the same dynamic range.

OK, that's all technical... So, Goneten, what is a good place to find reliable specs for subwoofers?

[Vaughan]

I see Goneten also mentioned SPL earlier on. I guess he means maximum SPL (right, Vaughn?).

Yes.

what is a good place to find reliable specs for subwoofers?

That's an excellent question, but I don't think there is an easy answer to that one unfortunately. Part of the issue is that it's difficult to find a place that uses consistent testing methodologies. Heck, I'm not sure of anyone who does ground plane measurements these days. Although admittedly I haven't done much research into this as of late so for all I know there are several websites that do but I think it's unlikely.

HomeTheaterShack was an invaluable source of information when Illka was still measuring but he has since joined the ranks of JL Audio. His measurements are still available to look at.

Regards,

[Vaughan]

Just to touch on the above, I did find this :

http://avtalk.co.uk/showthread.php?t=10999

They seem to be the only ones I could find that perform ground plane measurements.

Regards,

[alternativeroute]

My view is that no matter how flat the sub is measured to be, it will always be at the mercy of the room and the rooms influence on the response is large enough to render anechoic or ground plane measurements to be useless.

I am just discoving the role rooms play now... It seems there are two options: either tweak the room to hell with loads of absorbants and reflectants or get an equaliser... Sorry off topic but worth noting if your are shopping according to FR graphs...

[joel]

Firstly thank you to Goneten for some usefull insights.

i can assure everyone that there is no conspiracy around printing manufacturer based claims.

On consistency, speaker reviews have always had claimed frequency response listed. Sub not, but as already stated this can be added.

Let us assume speaker A has been tested and found to exhibit a - 3db response of 25 Hz in the anechoic chamber and speaker B has a response of 35 Hz at wwith the same input voltage/current and both  have the same sensitivity across the frequency range, the same number of speaker drivers of the same size

We're talking about active subs here right?
This means that we may be able to measure  the same output level (let say 90db) and we may know the input voltage to the amplifer section (if this is a THX spec), but this is unlikely.
We never know sub gain, or level setting (important as on full gain setting there may be a 50Hz hum that would just be measured as a possible peak, and harmonics), we also don't know how much power is actually being delivered to the driver, we don't know if the subs internal crossover or equalisation was used (and these could artificially skew results) or bypassed and we often don't know what kind of measurement was taken either (anechoic, in room, etc).

So many unknowns and the only way for us to resolve the issue of which sub is best for you or me is to actually listen.
Good specs mean nothing if they don't translate into an enjoyable experience.

As per usual the offer to pop around to the AVSA studio to have a look  and listen is open, and I'll even go so far as to say that if you have a sub that you want to bring along, bring this too. If you ask nice enough I may even be persuaded to bring along one of the real time analysers I have or possibly even find/make a new cable for my Linear X LMS.

Heck I'm even willing (if the boss agrees) to have everyone who has a sub that they would like measured over to the AVSA office on a weekend. Who knows it could even make for a decent article.
 
Don't think I can do any more than stating my position and making the offer that I have