[OTL]

.... A loudspeaker is supposed to have no sound of its own, while 'simply' reproducing the source signal. A loudspeaker can be 'understood'. An instrument made a few hundred years ago is likely to be a different matter.

A good point, but the loudspeaker is not only reproducing the "tone" of the Strad.  It is simultaneously reproducing the "entire group" as well as the microphones, the venue, the mixdown EQ including the reproduction source, the interconnections and the amplification components in the chain.  To say nothing of launching the waveform into an "unknown" space.  This is so complex it cannot presently be appoximated, never mind modeled with anything approaching accuracy.

So, what is "accuate" in regards to a loudspeaker?

You could say that we understand the interaction of the final amplification device with the loudspeaker.  But which amplifier?  Which loudspeaker? In which room? With what mixdown EQ?  Hmmm....I can change the overall sound with a set of cables......?  Now only my audiophile recordings sound right and wouldn't that suck?

So, that leaves us with a craftsman's interpretation of what a loudspeaker should "sound" like.  Frankly, if a speaker can introduce even an approximation of the original event in sombody's living room, it's a damn miracle!  Most reproducers fall far short of this request.

In my experience several speakers come somewhat close to reproducing the ambiance, the palbability and the impact, but nothing can touch a pair of Apogee's that have been properly sourced, positioned and amplified.  NOTHING!  But then again, even Apogee's can't hold a candle to the live event.

With Apogee no longer in business I'm about to give VMPS a shot.

If they suck, I'll let 'em go cheap and chalk it up to a lesson learned.  If not, I'm grinnin' from ear to ear.

Either way, no big loss and I learn a ton.

Just my 2 cents.....

Mike B.

[JoshK]

I can assure you that VMPS speakers don't 'suck'.  That has never been what this thread was about, for the record.

Anyway, OTL, I can assure you what is and is not known about launching waveforms into a space isn't as mystical as you make it sound.

[OTL]

I can assure you that VMPS speakers don't 'suck'.  That has never been what this thread was about, for the record.

Anyway, OTL, I can assure you what is and is not known about launching waveforms into a space isn't as mystical as you make it sound.

Well, originally this thread was about how Brian tunes his speakers, implying that his xover changes could not measurably contibute to an audible change.  I'll leave that interpretation open, because....

Secondly, this thread discussed science's present understanding of the sound of a Strad and how it produces it's "tone" and the inability of today's scientists and artisans to produce anything closely approximating that "tone".  

As for me, I've been implying that if if you can measure it but can't reproduce it, either your measurements are faulty, or the goal just simply can't be materialized today due to the complexity of all the contributing factors.  Likely both are true for both examples.

All I'm attempting to put forward is that all this crap (both Brians's xover methodology and the Strad's tone) is not "well understood" by present science and that an artisan with the skills and an ear is very well capable of producing something that our present science cannot measure and/or reproduce.

There are no absolutes here.  Let's all ask more questions (me included) in pursuit of what we don't know instead of banging our chests and screaming about the very limited amount of valid information we do know.

Listen, share and enjoy.

Mike B.

[diy_freak]

Well, originally this thread was about how Brian tunes his speakers, implying that his xover changes could not measurably contibute to an audible change. I'll leave that interpretation open, because....

Secondly, this thread discussed science's present understanding of the sound of a Strad and how it produces it's "tone" and the inability of today's scientists and artisans to produce anything closely approximating that "tone".

As for me, I've been implying that if if you can measure it but can't reproduce it, either your measurements are faulty, or the goal just simply can't be materialized today due to the complexity of all the contributing factors. Likely both are true for both examples.

All I'm attempting to put forward is that all this crap (both Brians's xover methodology and the Strad's tone) is not "well understood" by present science and that an artisan with the skills and an ear is very well capable of producing something that our present science cannot measure and/or reproduce.

Hi Mike,

I appreciate your attitude. At the same time you are making a lot of assumptions. Let me try to point out a few out:
- this thread has to do with how the VMPS speakers sound
- this thread has to do with xover methodology or speaker design philosophy
- the tiny steps in tolerance/trimcaps discussed here are audible
...
- Instruments crafted in the late 17th century are comparible with loudspeaker manufactured yesterday
- These instruments have the same sound now as in the days they were made
- Stradivarius' violins are not well understood and not replicatable

Except for the last one, which I don't know enough about to comment, I'm certain none of them are true.

There are no absolutes here. Let's all ask more questions (me included) in pursuit of what we don't know instead of banging our chests and screaming about the very limited amount of valid information we do know.

Listen, share and enjoy.

Mike B.

I totally agree with asking more questions. I also feel strongly about keeping an open mind for 'the unknown'. But there's a limit....

Do I need to understand the origin of gravity, to calculate where and at what speed the apple is going to hit the ground? Of course not. The same goes for the issue of component tolerance which we discussed here. It's really basic engineering, which is also the way that Brian presented it.

I feel completely confident when I state that no human being can hear a 0.01dB difference. That was the issue here. There were many, many issues presented that contradict Brian's claims. Let's not wipe them all from the table by going into the land of mysticism.

Edit: many typo's and then some...

[Ethan Winer]

Mike,

> Secondly, this thread discussed science's present understanding of the sound of a Strad and how it produces it's "tone" and the inability of today's scientists and artisans to produce anything closely approximating that "tone". <

With all due respect, you haven't a clue what you're talking about. If you follow the link I posted, and do a little Googling yourself, you will find that many modern luthiers are making truly excellent instruments. Understand that every piece of wood is different, so it's impossible to make one violin sound exactly like another. But no two Strads are identical either! The only "consistent" instruments I'm aware of are those now being made from carbon fiber. They're even making bows from that material, and some of those are pretty darn good too.

I attended a conference a few years ago where a famous string quartet played the same pieces four times - once with their own Strads, and once each for new instruments made by three modern luthiers. Carleen Hutchins was there, as were Bob and Deena Spear. I forget who the third modern luthier was. Bob and Deena are friends of mine, which is how I got to be invited.

Anyway, the bottom line is all four sets of instruments sounded very good. To the point that you couldn't proclaim one set "better" than another. Yes you could hear differences. But all were truly excellent.

The reason Strads sell for millions of dollars has little to do with their sound, and everything to do with their value as antiques. My friend Bob Spear has sold his instruments to a lot of very famous people, including Rostropovich and many famous symphony principal players. I assure you Slava can afford a Strad (and he owns at least one), yet he bought one of Bob's cellos anyway.

If you prefer to believe in magic and the unknown - and that's really the issue here - that's your right. But when you step into the realm of science and acoustics and musical instrument properites, you have to expect science-based answers from guys like me. Again, all of this stuff is well understood - violins and violas and cellos all function exactly as I explained (multiple resonances evenly spaced, having a high Q, etc) whether you care to accept it or not.

--Ethan

[diy_freak]

I'm fairly sure that if you ask any of these people about the audibility of differences in loudness, none of them will mention numbers lower than 0.1dB, not even under lab conditions. Dr. Klippel is the world's foremost authority on measurements and how to interprete them and would be the first person one would want to ask, which I did a few minutes ago via the Klippel site. Hopefully the email will come through.

It did   I can't help but be pleasantly surprised when people of this stature are willing to answer little questions asked by little minds. It seems that the most knowledgable people I encounter are also the most modest... I had a similar experience when I had a chance to meet Thiel and Small a few weeks ago at a dutch AES meeting. Here's the question and the thoughtful reply by Dr. Ing. Klippel:

Question:

I have a specific question which hopefully can be answered by Dr. Ing. Wolfgang Klippel... There are a few researches done and papers published on this simple matter, so I'm already in the right direction, but I'd like to hear the views of the world's foremost authority on measurements and their meaning. The question is: What is the smallest audible difference in loudness? So far I've seen numbers down to 0.2dB, which agrees with my own experience. Sometimes I fool myself in thinking I can detect smaller differences, but I can't be sure at all. Obviously, no two set of ears are the same, but I would think you would have an accurate and broad view on this matter.

Answer:

I am not an expert for this question ( I  have done no experiments in this matter). However, you find in the book Psychoacoustics by Zwicker on page 194 under 4.5 Models of Just-Noticeable Changes the approximative threshold factor of 1 dB. This is an approximative number and I believe under certain measurement conditions (e.g. measuring the amplitude modulation) you might come down to 0.2 dB. On the other side if you are interested in using this threshold for the interpretation of the amplitude response of an audio system I think you should work with 1 dB within a third octave band.

[OTL]

Ah yes.  But this is a one dimensional response to a 2 or three or 4 or more dimensional question.........

Amplitude response can be measured....phase response can be somewhat measured, but when both (and untold other) considerations are applied across a driver, driven by an amplifier of unknown reactance with feedback attempting to correct non-linear distortions without any respect to the original source signal's dynamic range, frequency response or acoustic venue...what are the chances of the faithful reproduciton of the original event in your home?  I'd wager that your local lottery has exponentially better odds.

WTF can possibly be done within the reproduction chain to improve upon the "accuracy" of the original signal in such a manner that makes it "more accurate" when reproduced at home?

In the meantime listen, share and enjoy.

Mike B.

[Brian Cheney]

OTL:
you're in a position to check your own perception of level differences with the RM30.

The L-pad between its 12 o'clock and 5 o'clock setting has an adjustment range of 4dB and about forty spaces between the windings in that area.  The characteristic is logarithmic but let's say each movement of the wiper from winding space to winding space is 0.1dB.  If you adjust one speaker only, that's 1/20th dB change in the band the Lpad influences.

Why not try to hear how small a change in midband level you can determine?  If you have a helper you can perform the test single blind.  Use a small flatblade screwdriver to change settings and go the smallest possible distance on the Lpad, one winding to the next.  Do this on one speaker only.  

I have trained myself to hear (in my quiet ideal acoustic environment, a 31'x13' LEDE room) a 12 o'clock to 12:05 o'clock difference in midband level, via the variable L pad on the back of the speaker.   You can be the subject or the tester (if another listener is present). Avoid communication until a full run of 20 trials is complete.

Remember a trial does not have to include an actual level change, but it is less confusing if you actually make a change up or down during each trial. Take a break between runs and use familiar program material.

This test is uncontrolled, but you may find it semi-useful. Good luck!

[diy_freak]

The L-pad between its 12 o'clock and 5 o'clock setting has an adjustment range of 4dB and about forty spaces between the windings in that area. The characteristic is logarithmic but let's say each movement of the wiper from winding space to winding space is 0.1dB.

Logarithmic is rather different from linear... It would be helpful for OTL to know what the changes are at the beginning, middle and end of the pot position.

I have trained myself to hear (in my quiet ideal acoustic environment, a 31'x13' LEDE room) a 12 o'clock to 12:05 o'clock difference in midband level, via the variable L pad on the back of the speaker.

Doesn't that present a practical problem: how to consistently turning a tiny pot only 2.5 degrees at a time? How do you do that? What dB difference does this represent?

[Danny Richie]

Doesn't that present a practical problem: how to consistently turning a tiny pot only 2.5 degrees at a time? How do you do that? What dB difference does this represent?

The first issue I have with the L-pad pots is that the sound quality through them is as bad as bad gets.

http://www.partsexpress.com/pe/pshowdetl.cfm?&DID=7&Partnumber=260-248&CFID=797133&CFTOKEN=24469829

Even the cheap sand caste resistors sound better.

http://www.partsexpress.com/webpage.cfm?&DID=7&WebPage_ID=322

Aside from that issue there are some of the other problems that I have seen with using the variable L-pads.

I worked on a set of speakers once that used them. The brand is not important.

What I noticed is that they were mounted from the inside and that there is no consistent clocking of the body of it.



In other words the body of it could vary by several degrees just by the way it is mounted. So regardless of the position of the flat headed, screw driver, slotted knob on the outside of the speaker, it could actually be a different clocking than one might think. You could adjust the outside slot so that it is perfectly vertical on two speakers but the body of the pot might be several degrees off by its rotation on the inside.

Then of course they are rated for 300 degrees of rotation +/- 5 degrees. So there is some variance to their range as well.

http://www.solen.ca/lpad.htm

Plus there is also a tolerance level between the two legs. For instance, you can take two units, and using a meter adjust them so that you get identical readings on the series leg, but then there could be as much as a 5% variance in the tolerances of the parallel leg, or visa versa.

With the pair of speakers that I was working with there were several challenges to balancing the pair of speakers to be the same, and I don't think it would have been possible to have achieved it without my measuring system. Getting it close by ear would have been a long and difficult process.

The first problem with this pair of speakers was finding a microphone height that gave the least amount of driver cancellation. The drivers were acoustically out of phase and cancelled one another all up and down as microphone height changed.

Changes in height by several inches cause big swings in response to the tune of 15 to 20db. Even a 2" change in microphone height caused a 14db swing in response.

Once a measuring height was found with the least amount of cancellation then I began adjusting the pots on the mid and tweeter to balance the response to as close to accurate as possible.

The next challenge was setting the second speaker in exactly the same place and trying to get it to match the first speaker. The result is pictured below.



Even with the adjustable pots I could not compensate for the fact that the tweeters were off by as much as 3.5db in the top end. Through the main range though, I was able to get the two speakers to be within 1db of each other.

Now the reason I posted these measurements was to show that even adjusting this pair of speakers to be as close to identical as possible, they are still very different. One speaker is 1db louder in one area than the other, while the other speaker might be 1db louder than the first in a different area.

Now this being said, how much would one speaker need to be louder in a given area in order for there to be a perceived difference in loudness from one speaker to the other?

I still think that a speaker will need to be a full db louder or more for any difference in a given area to be noticed under the best listening conditions. For the average person to notice a difference in output level, in a given area, then there would need to be a difference in the 3db range.

For this particular pair of speakers one could simply change there listening height with a simple tilt of the head and there would be a variance of a much greater magnitude than any minor changes in the position of the pots.

If a person cannot notice an amplitude shift in the order of 10 to 15db by small changes in listening height then how can a .1db difference be heard?

If I were to take a measurement of a speaker and then let someone take down my microphone, and readjust the stand height, then have them hand it back to me and ask me to put it back where it was so that I got the exact same reading to within .1db, then it would take me 5 or 6 difference measurements to find a spot where I was within .1db of the original measurement if I was lucky.

[mjosef]

So if the pot is a 'crappy pot', is there a higher quality pot available? Just wondering, since there are 'crappy' volume pots and better quality ones.

[diy_freak]

So if the pot is a 'crappy pot', is there a higher quality pot available? Just wondering, since there are 'crappy' volume pots and better quality ones.

I know one thing for sure: there aren't pots in existence with tolerances that come anywhere near the tolerances Brian claims to apply.

Brian, are you still reading this? How can you claim that you 'trim with 0.01uF steps by ear to the exact needed capacitance' while using pots which have tolerances that are many, many times higher, no matter what quality you might have found? Yet another reason why your claim is not improbable, but simply impossible. The only thing I'm wondering about somewhat is wether you actually believe it yourself. And if so, how did you manage to do that?

Of course, I'm not expecting an answer from you here. But in case you do have something to say, do it here instead of some other thread with a 'vague' reference to our 'discussion'. That would be great, heck, even mature. Thanks.

[JohnR]

Hans, Brian wasn't talking to you, he offered a suggestion to someone else. He's not obliged to respond to your bait.

[Danny Richie]

So if the pot is a 'crappy pot', is there a higher quality pot available? Just wondering, since there are 'crappy' volume pots and better quality ones.

I have had application for adjustable resistance to control the amount of bass management in the Alpha LS design.

I used a high quality Ohmite stepped rotary switch. It is a 15 amp switch with Silver coated contacts. I think they were $60. each.



I then used bundles of 100 ohm and 80 ohm Mills resistors on each leg to get the desired resistance levels to choose from.

Seams like I had resistance levels of 16, 20, 25, and 33 ohms and one blank for zero attenuation.

Anyway, there are options out there.

[Skynyrd]

My take on the whole speaker compnonet tolerance in VMPS speakers is:  If the speakers don't sound great, it's your fault due to your inability to tune the putty and/or the midrange-tweets properly.

This can never be proved.

Skynyrd

[OTL]

OTL:
you're in a position to check your own perception of level differences with the RM30.

The L-pad between its 12 o'clock and 5 o'clock setting has an adjustment range of 4dB and about forty spaces between the windings in that area.  The characteristic is logarithmic but let's say each movement of the wiper from winding space to winding space is 0.1dB.  If you adjust one speaker only, that's 1/20th dB change in the band the Lpad influences.

Why not try to hear how small a change in midband level you can determine?  If you have a helper you can perform the test single blind.  Use a small flatblade screwdriver to change settings and go the smallest possible distance on the Lpad, one winding to the next.  Do this on one speaker only.  

I have trained myself to hear (in my quiet ideal acoustic environment, a 31'x13' LEDE room) a 12 o'clock to 12:05 o'clock difference in midband level, via the variable L pad on the back of the speaker.   You can be the subject or the tester (if another listener is present). Avoid communication until a full run of 20 trials is complete.

Remember a trial does not have to include an actual level change, but it is less confusing if you actually make a change up or down during each trial. Take a break between runs and use familiar program material.

This test is uncontrolled, but you may find it semi-useful. Good luck!

Apologies for the late response, but some of us make a living in disiplines that don't give a rat's ass about audio reproduction.  For me, intenet design and Enterprise security have no concerns about the faithfull reproduction of an audio waveform.  But, thankfully all of us here do.

Ahhh....got the RM30's playing a bit of Sade in the background......

Unfortunately, what is below is a tease because my VMPS pair continue to change as they run-in...making any judgement at this point, well....almost pointless.

What I will contribute is that:
- I (and my SO) can easily hear the difference between daytime and nightime power
- Although not quite up to "cinema standards", the RM30's offer a quick and palpable bass that will let go of a tune, but not too soon.
- I (and my neighbors) are enjoying my vinyl again, and again and again....
- I (almost) have more volume than required.  (Ask my neighbors) 
- There's some considerable work to come to maximize what I'm hearing, and I welcome it.

Listen, share and enjoy!

OTL

[diy_freak]

What I will contribute is that:
- I (and my SO) can easily hear the difference between daytime and nightime power
- Although not quite up to "cinema standards", the RM30's offer a quick and palpable bass that will let go of a tune, but not too soon.
- I (and my neighbors) are enjoying my vinyl again, and again and again....
- I (almost) have more volume than required.  (Ask my neighbors) 
- There's some considerable work to come to maximize what I'm hearing, and I welcome it.

I'm glad you like your speakers, but that's not what the topic is about... 

[SCooper]

DIY_ Freak

If that is not what the topic is about, WHY do YOU continue to mention Brian ?? 

[diy_freak]

DIY_ Freak

If that is not what the topic is about, WHY do YOU continue to mention Brian ?? 

Another VMPS owner in distress...

I guess I need to point out the obvious: this topic started in a direct discussion with Brian about, IMO rediculously, small tolerances. When presented with a few simple facts, he made very clear that he did not want to discuss his claims and went even further by not allowing me to post in his circle. So I started this thread to hear opinions from other speaker designers.  Surprisingly, he did enter this thread with a few posts, but I'm afraid their content does not substantiate in any way the wild claims he made ealier, in fact, it only contradicts them even more, as pointed out by several posters. Still, this thread has had pieces of meaningfull information exchange. A 'VMPS review' is not part of that.

I still welcome some feedback from OTL about the L-pad settings, eventhough as mentioned before, logarithmic is not linear so we can't be exact about the actual dB changes.

[ctviggen]

I have to say that much of what's being here doesn't make sense.  People are willing to attack Brian because he likes to adjust his tolerances to a very high degree, which people argue is immeaserable although apparently Brian is of the opposite opinion.  Yet those very same people espouse different amplifiers and use of power cords, interconnects, etc., none of which produces any measureable deviation whatsoever.  The very same people laughing at Brian for his apparently ridiculous beliefs about tolerances for filters have the same unsupportable beliefs regarding break in of cords, the "helpful" use of power cords, amplifiers have similar THD and intermodulation distortion readings sounding different, interconnects with no measureable differences, etc.  You can't have it both ways -- either you believe that mathematics and measurements solve these problems, or you don't.  I can't see how 2 percent tolerance on filters doesn't matter but amps that measure exactly the same sound different.