Axiom Audio EP500 VS SVS PB12-Plus/2

[silversurfer]

OK, I was hoping maybe I was making some sense.

I agree with you though, in the business on how humans hear or perceive hearing, there is a lot that can not be explained with objective data.

 

[craigsub]

OK, I was hoping maybe I was making some sense.

I agree with you though, in the business on how humans hear or perceive hearing, there is a lot that can not be explained with objective data.

You were making sense ... it was just that the "on a roll" reminded me of Animal House ... so I could not resist.

And yes, there are people so wrapped up in measurements that they will go to great lengths to discredit those who value listening. They call listening "voo-doo", and "subjective nonsense" ... and they also will say that those who prefer listening tests to measurements are "measurement phobic" in order to discredit the listening test.

But, if you think about it, measuring is at least as subjective, because it is people deciding which measurements are important.

The most often quoted test result as given by the main on line "testers" is max SPL at 20 Hz with no more than 10% THD. That is a subjectively picked criteria.

Enough for now ... My Behringer arrived, and I need to play ...

 

[silversurfer]

But, if you think about it, measuring is at least as subjective, because it is people deciding which measurements are important.

I agree, which is why I posted that little piece earlier on why I think it is dangerous to believe that one "reviewer" is the end-all. I believe people have or will start engineering subs for his tests and biases, some can even argue that his tests are geared for a certain type/brand of sub. In either case, it is not healthy. When one group of people pushes one reviewer or train of thought, it limits advancement and minimizes other aspects than may be just as or even more important.

 

[---k---]

http://stereophile.com/features/306truth/

Just something to think about.

 

[craigsub]

Ryan ... That was an excellent read. Here are a couple of graphs, as promised, of the VTF-3HO from 10 to 50 Hz. This is the both ports open mode. I have included the 1/24th octave raw sweep, and the smoothed sweep also. I also included a 1/24th sweep switching to 1/6th using the menu.

 

[Steve42]

http://stereophile.com/features/306truth/

Just something to think about.

k,

As a long time music lover and audio equipment "nut", that is one of the absolute best articles I've ever read on this subject. I'm sure similar things have been written by others, but I like how it really gets into it and gives good examples and quotes from professionals on the subject.

It really should be a "sticky" at the top of every AV forum IMHO. I'm sure it would alleviate a lot of useless arguments and confrontations on these boards!

Thank you very much for posting it.

Steve

 

[Mark Seaton]

The issue is not whether THD is inaudible ... the issue is WHEN does THD become audible ? This whole 10% THD + N "brick wall" makes no sense, unless your subwoofers happen to be built to that spec.

The problem lies in that the "T" in THD leaves us way too many variables to hope for any strong correlation to THD in anything other than very similar systems. HD is audible beyond some level that is dependent on both harmonic order and fundamental frequency. As such, a THD measurement gives us no indication of the audibility of the distoriton, just the magnitude of all distortion produced.

This is hardly a new concept, although not often talked about since THD was a lot easier for manufacturers to denote, especially in the electronics world. Back in 1950 D. E. L. Shorter suggested that harmonics be weighted by n^2/4 where n = the order of the harmonic. I believe that works out to roughly a 12dB/octave weighting, which is not too far off of the more recent work by Earl Geddes and Lidia Lee, which they call the GedLee metric. This article includes a 3rd party look at these metrics and references for Shorters work and some before him and is posted on the GedLee website:
Weighting Up by Keith Howard

The exaggerated example/translation of the above is that when producing 60Hz, a 1200Hz harmonic doesn't have to be very loud to be noticable and annoying, while 120Hz 2nd harmonic can be much more significant before becoming objectionable. I whole heartedly agree that 10% THD should not define the maximum limits. Personally I still prefer the 1/3rd octave burst that Keele and Linkwitz first used which leaves it to the reviewer to determine at what level the sound of the "boink" changes. It is quite easy to discern when the distortion becomes clearly audible.

 

[Peter Marcks]

This is my response to Ilkka's test on IMD:

Certainly a valiant effort at quantification, Ilkka, but I am surprised that many are so quick to jump to conclusions about a topic that is more complex than it seems and that most have a very limited understanding of.

Let’s talk about the test. 120db at 4” from the woofer does not translate into very high excursion. Since both subs use 12” woofers and have the same port tuning frequency, both subs should have the about same level of excursion per a given SPL (all else equal). One would expect Inter-modulation distortion to be about equal between the two subwoofers (all else equal).

Perhaps a more relevant comparison to illustrate differences in IMD would be to compare a single driver sealed subwoofer to a multi-driver ported subwoofer.

Your results suggest that even with an excellent driver like the Ultra driver, IMD is on the same order of magnitude as THD. Harmonic distortion, being harmonically related, is usually less objectional than IMD which creates non-harmonically related tones. So I would not say that IMD is not important. In fact, I would say IMD distortion in the Ultra is the dominant source of objectionable noise (looking just at this data set)!

Please do not misinterpret our stance on IMD (and this is not directed at Ilkka, just at everyone in general). Our stance is that, rather than using one ultra high excursion woofer, we would rather use more cone area and lower excursion in order to achieve a desired SPL level. There are multiple reasons for this based on our subwoofer design philosophy, including lower IMD. Some companies will claim that IMD is BS, which is ironic and even puzzling because these same companies also use multi-driver and/or ported systems which both act to reduce elements of IMD!

The reality is that any one distortion component taken in isolation will not be enough to capture differences in performance (and I am not claiming that Ilkka said this, just trying to provide perspective). We have always believed in achieving a good balance of low dynamic, harmonic, and inter-modulation distortion instead of only focusing on one type of distortion.

Sincerely,

 

[jmprader]

The Stereophile article was a good one to bring into the mix.

I love graphs and measurements and hyperbole, but my ears are the final arbiter when it comes to hi-fi decisions. While consistent with my "near blind" sub decisions, I've been all over the map with speakers, amps and front end stuff until I got to what I consider sounds the best for what I can afford. Speakers have been the most problematic end of things.

I really enjoyed John Dunlavy's products with 1st order crossovers for many years but went to 4th order crossovers/TL designs because, to me, they sounded better overall (especially in the bottom end). Aside from extensive equipment testing and tuning individual crossovers (the IVa did a pretty good job producing 400hz square waves for a speaker), Dunlavy used to do an awful lot of blind testing with musicians in tow to get it right . The simple science seem logical to me that his product should be quite accurate at normal listening levels. However, listening (and professionals who used and recommended the PMC's that replace my Dunlavy's) told me different. Ultimately, I followed what my ears said was best...and the wife agreed when I abx'd her. Of course, my wife "can't hear Jimi", so she is suspect.

...of course, then there was the listening room...and no wall treatments...and...and...and.

I'm going to try to find the time to take the test tonight...see if my ears are made of tin. If they are, I can get an I-pod.

Nothing to say about the other thread that got this started. I'm dedicated to ignoring the anti-Bob's.

 

[bobgpsr]

The test does show that different people have varying levels of audibility of THD. I'm probably typical as I can't tell audible THD below -18db.

versus older quote where John does talk about hearing IM distortion:

...SPL in the deep bottom octaves and sound quality in the audible range are inversely related with high excursion drivers and therein lies my main problem with over-reliance on high excursion drivers. Its also why I am so enamoured with the HSU use with this new Adire motor topology. If you listen critically to different subs higher dynamic and IM distortion will show itself as a loss of detail and smearing in the mid and upper bass. That's why I have always looked cautiously at the THD+n measurements in the various reviews, they don't tell us the full picture about distortion and to the uninitiated those measurements can be misleading.

When I spoke to Dr. Hsu about distortion artifacts in drivers last November he talked at length about his reluctance to embrace high excursion drivers and how he was working on developing a solution to deal with these other distortions. Now we see what he was referring to...

 

[craigsub]

The problem lies in that the "T" in THD leaves us way too many variables to hope for any strong correlation to THD in anything other than very similar systems. HD is audible beyond some level that is dependent on both harmonic order and fundamental frequency. As such, a THD measurement gives us no indication of the audibility of the distoriton, just the magnitude of all distortion produced.

This is hardly a new concept, although not often talked about since THD was a lot easier for manufacturers to denote, especially in the electronics world. Back in 1950 D. E. L. Shorter suggested that harmonics be weighted by n^2/4 where n = the order of the harmonic. I believe that works out to roughly a 12dB/octave weighting, which is not too far off of the more recent work by Earl Geddes and Lidia Lee, which they call the GedLee metric. This article includes a 3rd party look at these metrics and references for Shorters work and some before him and is posted on the GedLee website:
Weighting Up by Keith Howard

The exaggerated example/translation of the above is that when producing 60Hz, a 1200Hz harmonic doesn't have to be very loud to be noticable and annoying, while 120Hz 2nd harmonic can be much more significant before becoming objectionable. I whole heartedly agree that 10% THD should not define the maximum limits. Personally I still prefer the 1/3rd octave burst that Keele and Linkwitz first used which leaves it to the reviewer to determine at what level the sound of the "boink" changes. It is quite easy to discern when the distortion becomes clearly audible.

Mark - Sometimes we have our differences. This is not one of those times ... totally agreed.

 

[Steve42]

versus older quote where John does talk about hearing IM distortion:

The sensitivity to THD tested with the Klippel test is different than testing sensitivity to IM distortion, no? The Klippel does not include IM does it?

 

[Buckeyefan 1]

http://stereophile.com/features/306truth/

Just something to think about.

Good article. There's nothing wrong with a little tube distortion .

 

[jakeman]

versus older quote where John does talk about hearing IM distortion:

Perhaps there was something you want clarified as I am missing your point. Why do you feel these two quotes are in conflict? In one I am reporting the results of my THD audability of the Klippel test and in the other I am talking about smearing of sound from dynamic and IM distortion. If you have a point to make then make it rather than suggest innuendo.

 

[jakeman]

I am surprised that many are so quick to jump to conclusions about a topic that is more complex than it seems and that most have a very limited understanding of.

Very true and it panders to the desire that most have to reduce complex physical pheneomena to one or two simple all encompassing numbers.

Let’s talk about the test. 120db at 4” from the woofer does not translate into very high excursion. Since both subs use 12” woofers and have the same port tuning frequency, both subs should have the about same level of excursion per a given SPL (all else equal). One would expect Inter-modulation distortion to be about equal between the two subwoofers (all else equal).

Indeed it should but that observation wasn't mentioned by the authors for some reason.

Your results suggest that even with an excellent driver like the Ultra driver, IMD is on the same order of magnitude as THD. Harmonic distortion, being harmonically related, is usually less objectional than IMD which creates non-harmonically related tones. So I would not say that IMD is not important. In fact, I would say IMD distortion in the Ultra is the dominant source of objectionable noise (looking just at this data set)!

Many users of those subs simply refuse to acknowledge that there is anything other than lowTHD in the Ultra. The smearing caused by IMD and dynamic distortion in the Ultra is clearly audible when compared to better sounding subs. It manifests itself as a loss of detail or articulation compared to say a EP500 as reported by the GTG participants and based on my own experience with these two subs in my HT. While I have not had the opportunity to listen to the VTF-3H0, Craig's impressions with music and comparing it to the Ultra suggests that the new HSU sub likely has less dynamic and IMD distortion as well, in keeping with HSU's stated balanced design criteria

Our stance is that, rather than using one ultra high excursion woofer, we would rather use more cone area and lower excursion in order to achieve a desired SPL level. There are multiple reasons for this based on our subwoofer design philosophy, including lower IMD. Some companies will claim that IMD is BS, which is ironic and even puzzling because these same companies also use multi-driver and/or ported systems which both act to reduce elements of IMD!

Yes. That stance puzzles me too since IMD and dynamic distortions IMO have a more detrimental effect on sound quality compared to THD depending on how sensitive one is to THD in the low frequency range. I guess BS does baffle some brains.

The reality is that any one distortion component taken in isolation will not be enough to capture differences in performance . We have always believed in achieving a good balance of low dynamic, harmonic, and inter-modulation distortion instead of only focusing on one type of distortion.

Amen. Great post and thanks for the well written insight.

 

[Steve42]

Good article. There's nothing wrong with a little tube distortion .

I think the article points to more than just that tube gear can sound better to some people. To me it was written to ecompass all audio gear not just one particular example.

 

[Steve42]

Looking forward to guys taking the Klippel test. It should be informative. Anyone doing so by tonight will have their % threshold published ...

Well, I don't know if it's my hearing and my lousy built-in sound card, but I know it's not my Sony CD3000 headphones. I took the test several times and averaged around -12dB.

Unfortunately, I'll probably have to blame my ears even though the sound floor coming out of the computer was relatively noisy when no music was playing. However when the music was playing it seemed pretty darn clear with no audible background "noise". I could play it pretty loud too as the CD3000's have a very sensitive rating.

Anyway, I'm begining to be even more of a believer that THD (at least in my case) is not that much of an issue. Maybe ignorance is bliss.

 

[craigsub]

Well, I don't know if it's my hearing and my lousy built-in sound card, but I know it's not my Sony CD3000 headphones. I took the test several times and averaged around -12dB.

Unfortunately, I'll probably have to blame my ears even though the sound floor coming out of the computer was relatively noisy when no music was playing. However when the music was playing it seemed pretty darn clear with no audible background "noise". I could play it pretty loud too as the CD3000's have a very sensitive rating.

Anyway, I'm begining to be even more of a believer that THD (at least in my case) is not that much of an issue. Maybe ignorance is bliss.

Let's also say you are really honest. As speakers go, Axiom's M80's are ruthlessly revealing. They are on a par with Klipschorns. I have watched musicians, speaker designers, and teenagers all fail to get past -18 dB.

I still chuckle over the guy who claimed to hear -45 dB on computer speakers.

 

[jmprader]

I managed a -15 on my kinda sorta decent desktop computer speakers during the homework hour (used to be my "home office", now, usurped). Hope to do it on the laptop routed through my mains sometime to see how much full range speakers help. My -18 was honestly a WAG and wrong.

Looking at the results pages, I'd be very interested to meet the guys who got to -45...I'll bet none of those guys never attended a Dead Concert in the 70's...

 

[Steve42]

Let's also say you are really honest. As speakers go, Axiom's M80's are ruthlessly revealing. They are on a par with Klipschorns. I have watched musicians, speaker designers, and teenagers all fail to get past -18 dB.

I still chuckle over the guy who claimed to hear -45 dB on computer speakers.

Yeah, I think people would be even more surprised if they took the test a few times one right after the other. I can see getting lucky a few times in a row within one test, but -45dB, yeah right........

I'd like to see a test done with any quality gear where someone consistantly scores even close to that high.