Jump to content
Electrodynamic

SI 4" coil DS4-18 info and discussion

Recommended Posts

Wow...............moving along.................As Bosso asked, and I am very curious to know also, will there be a 15" version of the DS4 Nick?

Share this post


Link to post
Share on other sites

Yeah but you said many people would sim the 24 to have more Spl than normal when using the wrong parameters. What people modify WinIsd to get them close? Surely not many. The point is It looks like to get an accurate model from data base one needs the Le xmax and Bl xmax for all frequencies to be accurate. Why use the Le when the driver is an ULF monster? So your beef is Klippel verified yet Ricci explained to you there can be discrepancies in usage and operation. You may have to call many companies to satisfy your needs.

Share this post


Link to post
Share on other sites

Oh, I would not even know what the hell was going on until you posted. People jest about Klippel but I never knew what they were talking about, now I do.

Share this post


Link to post
Share on other sites

Yeah but you said many people would sim the 24 to have more Spl than normal when using the wrong parameters. What people modify WinIsd to get them close? Surely not many. The point is It looks like to get an accurate model from data base one needs the Le xmax and Bl xmax for all frequencies to be accurate. Why use the Le when the driver is an ULF monster? So your beef is Klippel verified yet Ricci explained to you there can be discrepancies in usage and operation. You may have to call many companies to satisfy your needs.

For this particular driver:

If you want an idea of spl at 20 percent distortion you should sim using the klippel verified xmax.

If you want an idea of max usable spl you should sim using the published spec (and maybe even a bit more than the published spec).

If you want the sim to be accurate you should adjust for the large coil effects.

 

Usually you won't have both a klippel verified xmax and a Bl only based xmax.  As long as you know how the xmax was specified you can get some useful information from a sim.  Otherwise you don't know if the sim is predicting spl at 20 percent distortion or 100+ percent distortion or something in between.

Share this post


Link to post
Share on other sites

SI never claimed klippel verified on the website, only in forum posts, of which I linked a few in a recent avs thread.  And just a few weeks ago he told a customer that one SI 24 = three HST 18s, so I don't think this point has sunk in for him just yet.

 

But yeah, if he doesn't make any false claims or provocations I would never respond to anything he says.  Why would I?

 

The problem recently hasn't been so much Nick but misinformation from others (usually very non technical people) like Enders in the most recent avs thread.  Usually I pop in with a quick post to clear things up and then get attacked for it - Enders posted a purely provocative post that he even admitted he fully expected to be banned for in response to my technical response.  This is not the exception, this is how it usually goes.

 

So that's where we're at.  I said a long time ago that if no one was posting false information or provocations I wouldn't ever say a word.  And I mean it.  This isn't fun and I get nothing out of it regardless of the crazy motivations people attribute to me.  And it's not just about Si either - Bill Fitzmaurice is probably by far the most unethical person I've ever seen abuse the concept of open forums for personal gain.  When unethical stuff happens I'm not going to just shut up and ignore it.  I know you talk about Definitive and some people talk about Chase(?) but I don't even know who they are.  Like I said I don't go around looking for scandals.  But if I see a problem I talk about it.

 

SI makes good products and SI customers and fans are good people.  They are just a bit too liberal with the claims sometimes and really don't like being corrected.

 

I fully admit I myself got too confrontational and pushed things further than necessary at times, but that's what happens when you face a barrage of character attacks and bizarre questioning of my motivations and the responses are completely devoid of any technical reasoning at all.. 

 

Very simply though - yes, I would love it if all this would stop.  But again, I never start it, I just respond to false information and provocations.

 

Lets end this right here. Thankfully AVS doesn't delete older posts so I can dig up the proof if needed. I have said it before on AVS and I will say it again here on DB. I use Xmax as 70% BL as per the Klippel BL graph. As in the BL plot is Klippel verified. I also said that I never listed Xmax as being within the 10% shift in Le that Klippel uses. Again, I can go back and screen shot it if that makes diyspeaker feel as if he was wrapped in a snuggy blanket...but I have said the latter on AVS so I'm not sure why he keeps clinging onto choice things I have said and is using them out of context. Hence the jokes about "Klippel verified". 

 

IMO this is beating a dead horse and at this point it is obvious there is nothing I can do to inform diyspeaker of the latter multiple stated information. If he chooses to misconstrue my posts into "NOT KLIPPEL VERIFIED" then he is entitled to do so and to also drag threads down into the dirt. Nothing I can do at this point. 

Share this post


Link to post
Share on other sites

I am not technical, at all!  I just know how shit(oops, bad word) works in my room.  I don't invent anything and just copy what great DIYers have done already.  I purchased drivers based on the community and data-bass, not SIMS.  So I have a question for you guys, if the SI 24 is klippel verified at 20mm, and the Dayton UM18 is 22 mm, the Dayton being an 18 and the SI a 24 should be less than 6 dB in output over the Dayton right?  I am basing this off a 21 being 3 dB over the same 18, and the 24 6 dB over the same 18.  Just a generalization.  The 24 is actually closer to 10 dB more than the Dayton in the ULF region and would indicate a greater than 20mm klippel, no?  Why use the Le Klippel when it is not really the bandwidth the 24 is used for, it is an ULF monster, no?  This is my perspective who just likes to have a great HT experience.  20mm 24 should have less displacement than dual 34mm 18's, no? Again, all this is not making sense as the 24 is concerned unless some are looking at 20hz and above in which I hardly even notice, I look for under 20hz because all these drivers will rock above with multiples.   

 

This is a very good question.  Assuming the Klippel figures for the two drivers are accurate, it would be easy to assume the HS-24 with roughly double the surface area and approximately the same "Xmax" would beat the Dayton by only 6 dB.  And herein lies yet another "problem" with Klippel: people don't know how to interpret the data.  Mind you, this isn't really a problem of Klippel, but it does relate closely to the problem of people running sims (WinISD or whatever) without understanding how to interpret the results.

 

But of course, the Klippel Xmax is simply the minimum of three different Xmax figures.  Each figure is defined as the excursion threshold beyond which a particular T/S parameter deviates from its baseline value by a certain percentage.  IIRC, +/- 20-25% may be typically used.  Note that the actual analysis also yields curves showing how the key parameters vary throughout the driver stroke.  This data is much more useful than the Xmax figures because it allows you to anticipate how the driver will behave when pushed beyond those thresholds.  AFAIUI, the considered T/S parameters are motor force (BL), suspension stiffness (Kms), and inductance (Le).  The other physical parameters of the system don't change much with stroke like these do.  However, other effects can modulate other parameters like heat increasing DC resistance or altering inductance, and so on.  Klippel testing ignores these and only considers parameters that modulate with the cone motion.

 

Distortion occurs when excursion increases because these T/S parameters change as the driver moves.  Typically, BL drops as the coil moves out of the magnetic gap.  Suspension stiffness increases as the spider and surround are stretched.  And inductance changes as the location of the coil changes relative to the magnet assembly and other metal parts.  The key point is that the amount of distortion introduced by modulations of these parameters *at a particular frequency* depends on how sensitive the response of the system is to these parameters *at that frequency*.

 

Note that in a sealed box, the air acts as a spring and its stiffness merely adds to that of the driver suspension.  We can call this the overall stiffness and treat it just like we did the stiffness for the driver suspension.  In terms of percentages, the greater the driver stiffness, the less the air spring actually matters.  So when the driver moves far enough to hit the limits of its suspension, the air spring stiffness isn't going to be doing much unless it was really high to begin with.  In any case, a sealed enclosure will definitely alter the Kms curve but it won't do a whole lot to the Kms Xmax, which is where the driver Kms usually shoots up real high.

 

A driver's response is sensitive to motor force (BL) throughout its bandwidth.  As such, the driver is unlikely to sound good when played beyond its BL Xmax under any circumstances.  Suspension stiffness (Kms), however, only really effects the low end response of the driver, below the resonance point.  The higher you go, the less it matters until its contribution vanishes.  Lastly, a driver's response is sensitive to inductance (Le) only at the high end of its response.  The lower the frequency, the less inductance matters.  However!  There are a few important subtle points to consider with these facts:

  1. Excursion always increases with low frequencies.  If no strong low frequencies are present, the driver is not likely to move enough to matter anyway.
  2. The low frequency part of the signal that's causing high excursions *may or may not* be the only part of the signal that is actually distorting.  If you've exceeded the Kms limit, then you will likely see distortion products involving the low frequencies but any high frequencies present will not distort because the driver response at those frequencies doesn't really depend on suspension stiffness in the first place.  On the other hand, if you've exceeded the Le limit because of the presence of a low frequency signal, the high frequency part of that signal *will* distort because the driver response at those frequencies depends a lot more on Le, even though the low frequencies may not be distorting.
  3. Third, when the signal that is inducing the excursions is not the same signal that is distorting, the signal that is inducing the excursion modulates the signal that is distorting.  This modulation creates non-harmonic distortion products that are likely to be *a lot more audible* than any harmonic distortion products that are present.

The HS-24 beat the pants off of the Dayton by 10 dB, despite having only 6 dB more output at Klippel Xmax.  Why?  Recall that exceeding Le Xmax does not cause distortion of low frequencies.  Assuming the Klippel Xmax of the HS-24 is limited to 20 mm because of the Le Xmax, it may be able to play low frequencies at excursion levels much greater than 20 mm.  Most likely the Kms Xmax and especially the BL Xmax for the HS-24 are quite a bit higher than for the Dayton.

 

For purely low frequency use, a hypothetical driver with Le Xmax of 20 mm and BL and Kms Xms of 30 mm would probably play fine with excursions to 30 mm.  This is realistic, for example, if you are using beefy mains and a low crossover like 40 Hz.  For purely high frequency use, the same driver would likely never exceed 20 mm Xmax, so we don't care about this particular case.  However, if we want to use the driver both below the box resonance frequency and above the frequency where the impedance begins to rise again, we should be aware of that any high frequencies that are playing at the same time as low frequencies that bring the excursion past 20 mm are likely to sound bad, potentially quite bad.  Most of us use our subs up to 80 Hz or higher, so these are legitimate concerns, especially because distortion becomes more audible when it involves higher frequencies.

 

At the same time, you have to ask yourself how often you expect the drivers to be actually exceeding 20 mm Xmax.  For a 30 mm Xmax driver we're talking about a -3 dB difference.  How often will your drivers be playing at -3 dB from their excursion limits?  To be fair, BL and Le distortion are probably more objectionable because of their tendency to create IM products.  As such, that hypothetical driver with 20 mm Klippel Le Xmax might still produce "less than great" sounds with 10 mm excursion.  (Don't forget that you need to look at the actual Le(x) curve to see how much Le distortion you'll get at 10 mm.)  Also, is your goal really to "avoid objectionable sound"?  Or is it to "avoid masking of content that would have been heard on a better (or closer to reference) system"?  I wish we had better tools to quantify these effects on the actual subjective listening experience.  For now though, everyone has to decide how to weigh these things.

 

Just a few more notes:

 

Le is also modulated by the instantaneous current flowing through the coil.  This Le(i) is called flux modulation distortion.  It too affects high frequencies predominantly, but unlike the modulations involving cone motion, flux modulation distortion can be substantial even without a low frequency signal and high excursions.  All it really takes is a lot of power.

 

Last but not least, I'd like to come back to Kms.  While Kms is relatively unimportant at the resonance frequency and above, it can have a big impact on THD for the lowest frequencies, below the resonance in a sealed system.  That's not to say that BL is unimportant but, I would bet that a substantial number of drivers out there have lower Kms Xmax than BL Xmax.  I believe this is why the CEA numbers almost always drop off faster than 12 dB/octave for sealed systems below their resonance.  One exception that comes to mind with regard to this trend is the TC Sounds LMS-Ultra, which must have a very linear suspension out to its specified Xmax.  However, these drivers are notorious for being easy to bottom out.  The suspension doesn't stiffen enough at the end of its stroke to keep metal parts from bashing together.  There are always compromises.

  • Like 2

Share this post


Link to post
Share on other sites

Very nice analysis SME! One thing that you touched on that is worth emphasizing is, for all of the talk about distortion generated by various components of the driver in operation, what matters is the audible distortion, not the inaudible stuff. The linearity of excursion levels we should be concerned about should be tied to some kind of minimum perceptual threshold of distortion. The problem is the research in this area is fairly sparse. If we could channel some of the effort spent debating the finer points of driver specs and performance instead into finding out where distortion becomes audible, we could give driver manufacturers a more worthwhile goal to shoot for instead of wasting effort into reducing distortion where it may already be irrelevant. An awful lot of money and time is invested into lowering distortion by speaker and driver manufacturers every year, but if only a little bit of that was set aside for perceptual hearing research, they may be able to more sensibly direct their efforts into targeting the distortion that matters as opposed to the distortion that doesn't. 

Share this post


Link to post
Share on other sites

Lets end this right here. Thankfully AVS doesn't delete older posts so I can dig up the proof if needed. I have said it before on AVS and I will say it again here on DB. I use Xmax as 70% BL as per the Klippel BL graph. As in the BL plot is Klippel verified. I also said that I never listed Xmax as being within the 10% shift in Le that Klippel uses. Again, I can go back and screen shot it if that makes diyspeaker feel as if he was wrapped in a snuggy blanket...but I have said the latter on AVS so I'm not sure why he keeps clinging onto choice things I have said and is using them out of context. Hence the jokes about "Klippel verified". 

 

IMO this is beating a dead horse and at this point it is obvious there is nothing I can do to inform diyspeaker of the latter multiple stated information. If he chooses to misconstrue my posts into "NOT KLIPPEL VERIFIED" then he is entitled to do so and to also drag threads down into the dirt. Nothing I can do at this point. 

 

I really don't want to play this game but if you want to we can.  For every post you can find where you explained that your published xmax is based on only Bl I'll find two or three where you mentioned Klippel or Klippel verified without stating specifically it's based only on Bl.

 

I know you find the semantics of this ridiculous but there are implications.  There's a very good reason why people are surprised (shocked actually) that the Le xmax is 20 mm and a lot of the time flat out refuse to believe it even though there's a graph and an explanation from Jacob that clearly show it.  It's because of the semantics, not explaining it's only based on Bl MOST OF THE TIME.  Sure, you've explained it once or twice but most people haven't seen it.  They HAVE seen your published xmax, lots of posts where you reference klippel or klippel verified without an explanation, and the data-bass write up on the 18 that says it was klippel verified.

 

Instead of playing this game I'd much prefer to defer to technical information like SME's post.  Clearly you don't see the importance of being clear with semantics so let's just stick to the technical conversation, it's much more enlightening than arguing that once or twice you did actually explain things adequately but usually don't.

Share this post


Link to post
Share on other sites

Very nice analysis SME! One thing that you touched on that is worth emphasizing is, for all of the talk about distortion generated by various components of the driver in operation, what matters is the audible distortion, not the inaudible stuff. The linearity of excursion levels we should be concerned about should be tied to some kind of minimum perceptual threshold of distortion. The problem is the research in this area is fairly sparse. If we could channel some of the effort spent debating the finer points of driver specs and performance instead into finding out where distortion becomes audible, we could give driver manufacturers a more worthwhile goal to shoot for instead of wasting effort into reducing distortion where it may already be irrelevant. An awful lot of money and time is invested into lowering distortion by speaker and driver manufacturers every year, but if only a little bit of that was set aside for perceptual hearing research, they may be able to more sensibly direct their efforts into targeting the distortion that matters as opposed to the distortion that doesn't. 

 

Audibility of distortion is somewhat subjective so this can be tricky.  Even order distortion can sound pleasant, fuller, better than a clean signal, it's a big reason people like tube amps.  (Well, that and the bloated bass.)  

 

And with the odd order distortion it comes down to how much any given person finds acceptable.  Some people that crave max spl over all else will tolerate a lot of distortion if it means more spl while others don't want any audible odd order distortion at all.

 

It's almost impossible to define a perceptual threshold unless you have a clean signal to compare to.  This is where you run into trouble with stuff like subjective assessment at a GTG where there is no clean signal reference present.  And the fact that some people will actually prefer the less clean sound of very distorted bass muddies up the waters when trying to set a guideline about what's best in terms of distortion vs value.  It's very expensive to fix distortion issues at a driver design level, it can be cheaper, easier and better to set distortion levels at the system design level.  For example, if people buy two drivers instead of one they can get the benefit of lower distortion at any given spl level AND the ability to play a lot louder compared to a single driver.  So would you prefer to buy a single driver that has lower distortion and costs twice as much or would you prefer to buy two regular drivers that when played together have the same low distortion at any spl level but can also play much louder?

 

The klippel system is already capable of measuring more than just THD but it's not up to the system or the reviewer to subjectively assess the audibility of the issues it measures.  These IMD and MTD tests measure EXACTLY what SME talked about.  Multi tone and intermodulation issues with complex signals.

 

From here - http://techtalk.parts-express.com/forum/tech-talk-forum/49873-css-vwr126x-klippel-testing?p=728752#post728752

 

Then there are the 3 different distortion tests offered by the Klippel: THD (harmonic distortion and it's elements), IMD (intermodulated, where you fix one tone and sweep a range of others) and MTD (multitone, where you play multiple tones at once). Each has it's own benefit

Share this post


Link to post
Share on other sites

I really don't want to play this game but if you want to we can.  

 Coulda fooled me. I'm pretty sure you absolutely love this game.

 

 

 

It's almost impossible to define a perceptual threshold unless you have a clean signal to compare to.  This is where you run into trouble with stuff like subjective assessment at a GTG where there is no clean signal reference present.  

 

I sincerely hope you are not referencing my g2g with this, because now you are just making shit up.

Share this post


Link to post
Share on other sites

I was referencing ANY situation where you don't have a reference system set up with a measured clean (certain percentage distortion) output at given spl levels to compare to.

 

If you just run one subwoofer by itself and don't compare it to anything else and don't measure the distortion how can you possibly make any objective assessment of perception of audibility of distortion?

 

If you want to make an assessment about perception of distortion that means anything you have to have a reference system that has clearly defined and measured performance characteristics to compare to and/or at least measure the distortion with a tool somewhat more useful than SpecLab to correlate the amount of distortion present to what you are hearing.  You can't just beat the living hell out of the sub in your living room (the room itself presents problems like modal issues and decay issues so it's not the driver alone that's being assessed), look at SpecLab and say Wow that sounds great.  It doesn't mean anything at all objectively.  It could create a very strong SUBJECTIVE impression but stuff like data-bass measurements or klippel tests are much better at assessing an objective viewpoint.

 

There's also the whole issue of what type of signal is played, what type of enclosure the driver is in, what filters are used on the subwoofer (bandwidth), type of distortion that is most prevalent and the subjective assessment of how much is acceptable, all of which will heavily influence perception of audibility of distortion, so an objective study of perception of audibility of distortion is not something that can easily be done at ANY GTG or informal testing session.

Share this post


Link to post
Share on other sites

I was referencing ANY situation where you don't have a reference system set up with a measured clean (certain percentage distortion) output at given spl levels to compare to.

 

If you just run one subwoofer by itself and don't compare it to anything else and don't measure the distortion how can you possibly make any objective assessment of perception of audibility of distortion?

 

If you want to make an assessment about perception of distortion that means anything you have to have a reference system that has clearly defined and measured performance characteristics to compare to and/or at least measure the distortion with a tool somewhat more useful than SpecLab to correlate the amount of distortion present to what you are hearing.  You can't just beat the living hell out of the sub in your living room (the room itself presents problems like modal issues and decay issues so it's not the driver alone that's being assessed), look at SpecLab and say Wow that sounds great.  It doesn't mean anything at all objectively.  It could create a very strong SUBJECTIVE impression but stuff like data-bass measurements or klippel tests are much better at assessing an objective viewpoint.

 

There's also the whole issue of what type of signal is played, what type of enclosure the driver is in, what filters are used on the subwoofer (bandwidth), type of distortion that is most prevalent and the subjective assessment of how much is acceptable, all of which will heavily influence perception of audibility of distortion, so an objective study of perception of audibility of distortion is not something that can easily be done at ANY GTG or informal testing session.

 

 

Yea well, that's just like, your opinion man. 

 

If you are fully discounting spec-lab measurements that were compared to the true digital stream of the content that we were testing, then I don't really have much else to offer from that specific g2g, so we can just leave it there. I'll be sure to get on my anechoic chamber build when I get my tax refund...

 

We also didn't just run one sub "In my living room" that day. I have had at least a dozen different subs in my well treated dedicated HOME THEATER, of which I have done all kinds of measurements for, including decay times.

 

Sure DB and the Kword are likely better at measuring these things, but I don't have a DataBass and I don't have a Klippel. Sometimes you just gotta work with what is available to you, and a $3,000 measurement rig is what we had to work with.  

 

Hey Bosso, just throw that thing in the trash, it's worthless now. 

Share this post


Link to post
Share on other sites

Calm the sarcasm down.  Even klippel and data-bass measurement on their own are not sufficient to do a study on the perception of the audibility of distortion.  Neither of them have a subjective component at all (other than the text write up on the data-bass pages).  And perception is at least somewhat subjective.  But they are better than an in room SpecLab measurement.

 

Your SpecLab measurement just shows a picture with colors in which the room effects are just as dominant (if not more so) than the driver distortion.  (Unless there's components of SpecLab that I haven't seen, which is possible.  And depending on mic location and distance.)

 

Even a simple REW measurement is better than SpecLab for looking at distortion, it at least gives you a graph with numbers on it and you can see clearly the amplitude of each type of distortion so you know which type is predominant.  That give a much better picture of what you are actually hearing.  And obviously if you do the simple REW measurement outside the room is not in the equation.

 

The room really can't be discounted.  If you have a 15 db room mode and you eq that down then the distortion in those frequencies also goes way down but the decay time still isn't what it would be outside.  It's just a mess if the goal is to figure out the actual driver distortion.

 

The point is that if you don't know exactly how much or what type of distortion you are actually hearing you are in no position to judge the perception of audibility of that distortion.  This is really something that requires a full on professional study, not an informal test.

Share this post


Link to post
Share on other sites

Don't tell me to calm anything down when in a round about way you were essentially insulting me, and the process by which we ran that day. Sure the room can't be discounted, I never said it should be. We also had REW measurements from that day, and I have REW measurements from many other days and many other systems as well, but the measurements we saw through SL that day were surprisingly similar to distortion measurements that I have seen in other places with the same drivers. Who would have thought? 

 

All the spec lab colors are pretty though aren't they?

  • Like 1

Share this post


Link to post
Share on other sites

Don't tell me to calm anything down when in a round about way you were essentially insulting me, and the process by which we ran that day. Sure the room can't be discounted, I never said it should be. We also had REW measurements from that day, and I have REW measurements from many other days and many other systems as well, but the measurements we saw through SL that day were surprisingly similar to distortion measurements that I have seen in other places with the same drivers. Who would have thought? 

 

All the spec lab colors are pretty though aren't they?

 

I wasn't insulting you.  I just said that's not the way to do a study on perception of audibility of distortion.  I said a bunch of other stuff on the topic too but you took that particular part as an insult even though I was talking about informal testing in general with GTGs as an example, not your specific GTG.  I didn't even reference your GTG until you brought it up.  Perhaps i should have been more clear as this has been an issue of contention in the past and clearly continues to be.

 

I'm not sure how you correlate a picture with colors to the very specific information on a REW measurement graph and I'm pretty sure you can't, other than observing general trends, which is not nearly enough objective data for a study of the kind proposed by ShadyJ.

 

My head hurts too, yes please talk about DS4s.

Share this post


Link to post
Share on other sites

I guess the setups for all future GTG's need full peer review prior to the event to ensure scientific validity. All comments need to be subjected to peer review prior to public posting. Someone will a full technical background and experience, preferably at least a full professor at a leading academic institution, will have to serve as the editor.

 

All measurement equipment will have to be validated prior to event, the person running the equipment will require proof of training, minimally a professional certification for each piece of equipment.

 

To study the perceptual aspects, a different cadre of experts will be required. The disciplines of neuroscience and psychology will have to be engaged to ensure extraneous variables that were not accounted for in the validated study plan are dealt with. As the study involves human subjects, the entire GTG plan will have to be submitted to an institutional review board to ensure proper protection of human subjects.

 

Of course we will require a null hypothesis, identification of independent and depended variables, and an a priori statistical analysis plan. To ensure listener bias is not factored in the testing will need to be blind, preferably double blind.

 

Sounds like a lot of work....and no alcohol consumption by anyone at anytime during the event.

 

It's just like taking work on the road, count me out. I get enough science during the workweek, not on the agenda for weekends.

Share this post


Link to post
Share on other sites

I thought they were just about .... getting together.

 

Maybe I'm missing something. Didn't know a lab coat and hard work with forensic analysis were a requirement.

Share this post


Link to post
Share on other sites

Yes get togethers are just about getting together, that's the point.

 

Get togethers are not a good place to do a study on the audibility thresholds of distortion.  That's what I was trying to say, that's what I did say in response to ShadyJ.

 

About Beast's GTG in particular, now that it's been brought up, claims of low distortion were made.  I don't think the setting, the gear, the software or the manner of testing would be able to give any information to provide any objective input on that matter.  That's not the point of a GTG, it's to get together.  But then it should be known that any subjective claims made arising from the meeting are completely subjective.  That's not the way they were presented, even now in Beast's most recent posts.

Share this post


Link to post
Share on other sites

GTGs are not only about getting together and putting faces and personalities to screen names.

 

I've been to many GTGs over the past decade and a half. There were some very cool and well thought out presentations done to include several or more subs from different mannys and even several that had separate, dedicated 2-channel listening rooms for the golden ear folks.

 

Subjective jargon has always confused me. Partly because every brain is wired uniquely and partly because English simply does not have enough words to properly describe perceptions of human senses, including physical sensation, smell, sight and sound.

 

Over the years, we've grown more sophisticated with tech that allows us to 'see' what we're hearing. We have native FR, FR at the listening position in any given room, O scope and DMM analysis of the signal chain and spectrograph snapshots of the digits vs the presentation of same by whatever system at your ears.

 

Paul and I have focused on this metric for a number of years and he has acquired some very high resolution hardware to go with my ACO Pacific/Edirol measurement hardware. More importantly, The Wizard of Shred has devoured the operation manuals and theories associated with the various hardware and, simply put... knows what the hell he's doing.

 

Paul gathered A LOT of data from both GTGs at Brandon's. He formats and digests it and forwards some of it to me for scrutiny.

 

Here is a single, <one minute snapshot of the HS24 in action at Brandon's primary listening position reproducing a difficult passage from a well known Hipster-Dufus techno-instrumental tune.

 

The digits version is here compared to the HS24 presentation at the seats. Only the single HS24 was playing with a 100 Hz LPF in line. The input signal was bumped +15dBRL. To those who of the uninitiated, that's an ass load of input signal.

 

9032514d5aaa771f1a0800ce6f9f2198.gif

 

I am only going to post this animation. The signal was clean. The FR non-linearity (FR at the seats) will not be posted, but is reflected in the mic'd SL version. The digits version of the spectrograph was very accurately calibrated. The room gain profile, derived from the difference of the FR at the seats vs the native FR of the sub as tested, will also not be posted. Rest assured we have all of that data and more. Sufficient to quite accurately tell objectively what's happening at the listener's ears vs what was encoded on the CD.

 

Analyze the graphs to your hearts content. Enlarge specific areas and carefully match the recorded colors to the color scale, frequency-by-frequency. I have. Here's the bottom line... There is ZERO harmonic distortion, as a %, which is how HD is calculated, at the seat. ZEE, EEE, ARE, OH.

 

So, do I care what Klippel says? Do I care what WinISD, Hornresp, Bassbox or LSP Cad says? Do I care what the CEA2010 burst number @ 10 Hz, 2M, GP is? Although it's a real screamin' blast to read these, many times heated, theoretical back and forth, mostly among people who've never built a subwoofer, much less have the wherewithal to to verify any of the theory with accurate data, it means nothing compared to the above example.

 

The evidence overwhelmingly disputes the 20 mm claim. It's nonsense and whomever continues to try to pass it off as fact is wearing an egg mask.

 

James, nice posts on this subject and spot on. Nice to see you posting here. Start a running thread so we can keep up with your shenanigans, eh?

  • Like 2

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...

×
×
  • Create New...