In search of my last great loudspeaker

[NorthStar]

As we have to fill the thread with OT subjects, while we wait for the revelation, I will refer to the interesting work of Charles E. Zeilig and Jay Clawson, published in the last issue of HifiCritic, that can be downloaded at the HifiCritic site http://www.hificritic.com/flac-wav-sound-quality-research.html. The authors have developed a method to quantify small differences in sound reproduction "Height Method for Quantifying Subjective Sound Quality", and published it in several articles in TheAbsoluteSound. I have only looked superficially, and have no firm opinion on this technique, but it looks quite interesting. If the method was sensitive enough to show the differences due to computer activity in playback of files, probably it could also show differences due to power plug metals.

Although this work was presented more than a year ago, I could not find any discussion on it in the net.

Good reading Francisco; I'm still on the first installment of this two-part article...almost done...then part two.

1. http://www.hificritic.com/uploads/2...ifferences_between_wav_and_flac_formats.p df
2. http://www.hificritic.com/uploads/2...ifferences_between_wav_and_flac_formats.p df
__________

Maybe record albums (LP vinyls) still have a greater emotional music affair with audiophiles?

I have no relationship with any of those members of that discussion (from the video); the sole purpose is for people to listen to the music the way they feel more connected to it, and that is all. It's for serious people who take the time. After all, it's what we're here for; sweet music to our soul. :b

I know where my connection is; in both analog and digital music mediums.

 

[fas42]

Microstrip phrased it well when he suggested the better term is a difference in "perceived" response, which most likely has a qualitative rather than quantitative basis. However even that assumption may be wrong and opens up another can of words entirely. If there is a qualitative change rather than a quantitative change, shouldn't it be measurable? My science hat says yes, probably. However not every aspect of perception is yet measurable and what to measure to explain these perceptual differences may elude us for years to come. Another way of saying this, is that this has been a topic of great debate and discussion in the forum for years and probably will be for years to come without that the data that any empirical thinker would like to have to prove or disprove the hypotheses such as you have thoughtfully raised. I look forward to the data you anticipate obtaining nonetheless.

IMO, and experience, it is qualitative - on another website, in a blog area, someone mentioned switching on a decent system in the morning, for background listening, and had the volume just right, at the turn on, for this type of listening. But he was very perplexed - after an hour or so the volume rose, every time, so he would march down to the gear, and lower the volume by a certain amount, so that it was "just right", again. Every morning, the pattern repeated. "What's going on with my amplifier, it's faulty, the volume rises of its own accord - this is a nuisance!" he asked in a post. I suggested that, measurably, not a single thing was changing - but his subjective reaction to the sound had altered, because the quality of the playback was altering, getting better, as the system stabilised each morning ... a couple of days later, he posted, "You were spot on!" - he measured the actual SPLs at the switch on, and then the couple of hours later; of course, they were identical ...

 

[YashN]

There's a lot to read there, Marty and I did spend some time, but this is what I think:

There is way too much 'patching', too many complications. Aim for simplicity and 'directness'.

Here are some radical things to try:

Instead of getting manufacturers of speakers to get you a way to tweak components like resistors, radically change the approach, and first work on your room.

Use REW and a good condenser microphone and make a proper diagnosis of the room. Then correct the room response appropriately using organic, acoustic means. Use DIY acoustic panels as well as DIY bass traps.

Get rid of the piano in the listening room: it has strings and has a very good resonant body, the glass behind your listening position is not very good either (this judging from a linked-to post from this Original Post, so I have read partially two or three threads already with the walls of texts - not that this post is any shorter, haha! ).

Any resonant body will seriously affect what you're listening to, including your floorboard for instance.

Once this is done and the room response is appropriately smoothed, you can consider clean power, all the more important if you're bent on using DSP correction.

DSP correction for me isn't a panacea and should be avoided at all costs, unless there are pesky peaks and troughs that haven't been corrected by the organic, acoustic room treatment. It's a last resort.

It's a last resort for a couple of reasons: each DSP component will have its own signature. Grounding issues with the added component can complicate matters

Now, for clean power and other components, you will also want to consider vibration isolation especially of the seismic kind. And when it comes to the ultimate, you will want to combine that with any AC filter box.

I read that you are already very aware of tubes, and like the mid-range, so here a SET tube is all that is needed, nothing complex, just the sweet spot.

Additionally, and more importantly for me: I am not in agreement about the importance of frequency response in a system. For me, this is a solved problem and has been for many years. It is far, far more important to get a proper time accurate response, especially for timbral, rhythmic and spatial accuracy.

In short: Attack Transient Response > Frequency Response.

Now, having said that, continuing on the approach of finding boxed speakers + crossovers and pursuing manufacturers to provide flexibility for tweaking may not be the best - they won't agree: they spend time and money to tweak the components for their brand sound. It won't happen.

This requires a different approach.

The different approach is to use a crossover before amplification and then amplify with the best for the job (frequency-wise for you), and that also includes having amplifier outputs connecting directly to the drivers.

Instead of boxed drivers, try open-baffles, where usually you are able to tweak or integrate your own crossovers. Have someone build one for you if you don't want to build yourself anymore.

Now, seeing your mention of DSP processors it makes me think maybe you haven't heard high-rate DSD?

High-rate DSD (like DSD256), combined with a native DSD DAC and SET tube amp (or a combo of SET Tube + SS for low ends) is what gives me the purest sound.

Explore this route: use existing PCM if needed, up-convert in real-time to high-rate DSD into SET Tube amps (or a Tube + SS combo) and if processing is needed, use HQ Player as it can process native DSD natively or it can up-convert PCM to high-rate DSD (you'd need a native DSD DAC). Best thing is just to play native high-rate DSD in a native DSD DAC with the room acoustically treated.

Crossovers in software form on a computer can do wonders too if you bi-amp (you can tweak the time delays if necessary there too).

Nowadays, there is no need to spend an inordinate amount to get fantastic sound. Yes, you can always spend a lot to get whatever you want, but if what you want to approach is live players in a hall, you don't need DSP to correct the response: that's like applying an ointment on a pimple: it can hide it for a while but the pimple is still there as is the underlying cause.

In short:

0. Attack transient response first rather than frequency response (the latter is a side-effect of your room and the piano in it as well as potentially other things like the windows behind the listening position)
1. Clean Power (and grounding)
2. Isolate Vibrations
3. Diagnose and treat the room first organically and acoustically - lose the piano in the listening room. Only resort to DSP in case there are big peaks or troughs remaining.
4. Use Tubes, crossovers before the amps, connecting to the Open Baffle drivers directly. Have you tried Marchand cross-overs?
5. Try high-rate DSD + SET Tubes (or PCM, but with HQ Player ideally) + bi-amplication for the lower freq 'slam' if necessary

Think for instance of having a great acoustic musician at your place (Yo Yo Ma or Al Di Meola for instance). Would you tell him: "Hey, instead of listening directly to you, I'd like to 'mic' you and then DSP process it to 196kHz and listen to that instead.

Would that fly?

 

[Al M.]

Marty already made his decision, we are just waiting for the 'reveal'.

 

[Ron Resnick]

Marty, I loved reading your opening report! As someone who is on the same "last speaker" journey I was fascinated to read your progress and conclusions along the way!

Why wouldn't your Gothams be able to supplement the low frequency reponse of Rich's Divas in a way that would be satisfactory to you?

 

[Zero000]

211, first of all, its no embarrassment to be insane in this hobby. Most of are and its sort of a badge of honor. OK, seriously, nobody ever said that the sound of one's system does not change with rhodium vs gold, vs bronze plugs. That's not in question as this has been discussed extensively in this forum. (and as usual, without consensus; some prefer one and some prefer the other). What I am suggesting is that I find it difficult to believe that one can find a quantitative change when measuring FR to pink noise at the listening position that is dependent on the metal surface of the contacts in an AC outlet. Nobody disputes that there are surely qualitative changes that can be heard with these various outlets. Microstrip phrased it well when he suggested the better term is a difference in "perceived" response, which most likely has a qualitative rather than quantitative basis. However even that assumption may be wrong and opens up another can of words entirely. If there is a qualitative change rather than a quantitative change, shouldn't it be measurable? My science hat says yes, probably. However not every aspect of perception is yet measurable and what to measure to explain these perceptual differences may elude us for years to come. Another way of saying this, is that this has been a topic of great debate and discussion in the forum for years and probably will be for years to come without that the data that any empirical thinker would like to have to prove or disprove the hypotheses such as you have thoughtfully raised. I look forward to the data you anticipate obtaining nonetheless.

I find it extremely difficult to believe. Which is why I am going to do it. Because I can hear it.

Marty, what you need to do to detect this stuff is use very little smoothing, and keep zooming on in to an FR plot, preferably with a PC and a good sized screen. If there's something there, I find you can see patterns between the two items under test that are definitely repeatable.

It is a bit like using a very high power microscope.

 

[KeithR]

Yes, they are:

http://www.yg-acoustics.com/userfiles/files/PDF/Technologies/YG_DualCoherent_Des14.pdf

Alex - I guess it depends on how you define it. YG clearly specifies only at the crossover point which I wouldn't consider a truly time/phase coherent design. That said, very few companies do this these days.

Sorry for the interruption - probably a good thread topic for another sub-forum!

 

[bonzo75]

I asked another speaker designer a question, whether underpowering a speaker in a massively large room could lead to a bass shy response. What I understand from him, (admittedly without really understanding, is)

"Possibly, because as the test is usually done at no more than 4-8 watts max rms, power is not an issue , but it could show a roll off due to the output transformers. Also you can’t test like that at the listening position, to get a proper window you should do 3 plots , avg across the listening plane L/C/R then avg, to see what it actually looks like , a center seat only plots, say, nothing.


Could the output transformer (of the amp) limit the bass response .. ? Yes … "

 

[YashN]

Marty already made his decision, we are just waiting for the 'reveal'.

Curious nevertheless to see if he keeps the same approach of buying yet another boxed speaker set.

A while ago, Lukasz Ficus experimented with a cross-over rig so that he could easily swap components and compare. More recently, perhaps 1 year ago, like Marty, I asked similar questions about the crossover in the Totem Mites (they're in my city) and they were adamant they don't recommend changing anything - their voicing, their brand - much like the late Dr. Gizmo said several years back about how manufacturers react.

Makes more sense to go the Open-Baffle route firstly as everything is in the open, and further than that: bi-amp with active cross-overs before the amp and tweak there if necessary. In any case I don't like the cross-over in the cabinet approach anymore.

Also, total deadening of the room isn't the goal.

 

[YashN]

Yes, they are:

http://www.yg-acoustics.com/userfiles/files/PDF/Technologies/YG_DualCoherent_Des14.pdf

I like the approach (also 'preserve transient and spatial cues').

 

[bonzo75]

Curious nevertheless to see if he keeps the same approach of buying yet another boxed speaker set.

A while ago, Lukasz Ficus experimented with a cross-over rig so that he could easily swap components and compare. More recently, perhaps 1 year ago, like Marty, I asked similar questions about the crossover in the Totem Mites (they're in my city) and they were adamant they don't recommend changing anything - their voicing, their brand - much like the late Dr. Gizmo said several years back about how manufacturers react.

Makes more sense to go the Open-Baffle route firstly as everything is in the open, and further than that: bi-amp with active cross-overs before the amp and tweak there if necessary. In any case I don't like the cross-over in the cabinet approach anymore.

Also, total deadening of the room isn't the goal.

All sounds good except his room is not totally dead. I have been to a Rives Audio room and 3 complete SMT rooms and many treated rooms, and his 25*35*14 room, planned and built and treated over years, transported from Dallas to NJ, is the best around. In some ways, after doing the hard work upfront, he now has it easy. Put any speaker in that room with giant JL Gothams and Spectrals, the bass and slams and jumps and rise are unleashed like in no other room.

...(except the Apogee rooms )

 

[Steve williams]

All sounds good except his room is not totally dead. I have been to a Rives Audio room and 3 complete SMT rooms and many treated rooms, and his 25*35*14 room, planned and built and treated over years, transported from Dallas to NJ, is the best around. In some ways, after doing the hard work upfront, he now has it easy. Put any speaker in that room with giant JL Gothams and Spectrals, the bass and slams and jumps and rise are unleashed like in no other room.

...(except the Apogee rooms )

it is a very special room indeed. I agree

 

[YashN]

All sounds good except his room is not totally dead.

I didn't say it was: it actually is a very 'live' room if the glass windows and the piano are still there behind the listening position, the latter with its resonant body and strings that can vibrate sympathetically as well, thus affecting response, whereby you'll find yourself trying to patch this with DSP.

Much better to fix things in the room beforehand.

 

[Mike Lavigne]

the one room variable might be how the current room ideally works for the Pipedreams compared to how it might ideally work for a conventional dynamic driver speaker. but this is not really a lowest octave question.

 

[Zero000]

I asked another speaker designer a question, whether underpowering a speaker in a massively large room could lead to a bass shy response. What I understand from him, (admittedly without really understanding, is)

"Possibly, because as the test is usually done at no more than 4-8 watts max rms, power is not an issue , but it could show a roll off due to the output transformers. Also you can’t test like that at the listening position, to get a proper window you should do 3 plots , avg across the listening plane L/C/R then avg, to see what it actually looks like , a center seat only plots, say, nothing.


Could the output transformer (of the amp) limit the bass response .. ? Yes … "

All sorts of things affect frequency response.

But not much affects frequency response more than the position of the mic, bar using deliberate EQ.

For this reason, I take all plots with a pinch of salt. No question the mic at the listening position is the most meaningful in the room of the listener's speakers.

I once showed a loudspeaker designer how to take frequency response plots. I kid you not! He was using an SPL meter and a signal generator. After telling him where he was going wrong, he was being remarkably stubborn about defending his position.

But when he actually saw it done, and how simple it was, he ordered a Dayton, thank Christ! But I knew he was upset, as the results were not encouraging, and the variance with small mic positional changes was very marked indeed.

I also saw a plot of the Analysis Omega taken at 1m in a respectable magazine. It looked dreadful. You cannot measure a speaker like that at 1 metre and get a reasonable result.

Unless precise positional information is supplied for Rich's Diva FR measurement, and a proper calibrated mic is used, I would ignore the results. Really. You cannot deduce much of any meaning from it.

 

[Believe High Fidelity]

As every audiophile knows the first thing any audiophile has to come to terms with for any purchase is a budget. In my experience, there are 3 kinds of audiophile budgets; the actual cost, the cost he tells his wife, and the cost he tells his friends. In my case, I have the good fortune to be in a situation where the first two are identical. That does not mean that this comes without some sort of quid pro quo. You can bet your ass that it does. Now, as far as the cost you share with your friends, that is determined by what kind of friends you are referring to. To the friends that come to the house and see some big fancy speakers and electronics that glow nicely in the dark, I always tell them that no matter what piece of gear they point to, the cost is about thousand dollars, tops. Hey, we have to live with these people and that figure doesn’t bend them out of shape very much. Moreover, it computes well with them as they nod quietly and realize it’s more than they spent on their Bose system so it all makes sense, even if they think we’re crazy. (If they only knew!). However, to fellow audiophiles and folks who read blogs such as WTF, there is no need to disclose the cost- you already know it. But still, there has to be a budget. A real number budget. For this exercise, it was about 50-60K. So let’s get to work.

I’ll begin with a list of speakers that made my initial short list. These included Raidho D4.1s, Sonus Faber Aidas and Apogee Diva’s. Regretfully, there are many superb speakers I just could not afford to consider such as the Wilson Alexx or the top of the line models from Tidal or Evolution Acoustics as well as several others. But the objective of this exercise is not to have the world's best or most expensive stereo. Rather at this stage of my life and at my age (and probably good guidance for any age), it's all about having a highly enjoyable one that provides great sound regardless of source or content and most importantly, puts a smile on your face when you listen to music. And if you can't do that on 50 grand, I suggest something is very wrong with this picture

This is a very interesting game to guess the speaker. Lot of clues in this message so lets try to narrow the reveal. (hope I guess right)

Budget - 50-60k real world price

Ruled out:

Sonus Faber
Raidho
Apogees
Wilson
Tidal
EE

Former line array owner and have subs in place for bottom end (but still wants bottom octave from main speaker)
Traits taken
-Dynamics
-Soundstage
-Wide Dispersion
-Time Alignment


Given the above I am going to guess the following

-Rockport Avior/Cygnus
-Von Schweikert VR44/55
-Focal Utopia series

Look forward to seeing your reveal!

 

[microstrip]

(...) Why wouldn't your Gothams be able to supplement the low frequency reponse of Rich's Divas in a way that would be satisfactory to you?

In principle it more difficult to integrate a box subwoofer with a dipole than with conventional box speakers. I have now got a JL Audio CR-1. I f I find that its insertion is acceptable in my system I will look for some subwoofers for my SoundLabs - first the Martin Logan Descent I already own, then a pair of F113's .

 

[bonzo75]

In principle it more difficult to integrate a box subwoofer with a dipole than with conventional box speakers. I have now got a JL Audio CR-1. I f I find that its insertion is acceptable in my system I will look for some subwoofers for my SoundLabs - first the Martin Logan Descent I already own, then a pair of F113's .

Have you checked with Duke lejeune, he is a soundlabs veteran who matches subs with them

 

[LL21]

In principle it more difficult to integrate a box subwoofer with a dipole than with conventional box speakers. I have now got a JL Audio CR-1. I f I find that its insertion is acceptable in my system I will look for some subwoofers for my SoundLabs - first the Martin Logan Descent I already own, then a pair of F113's .

Very interesting...look forward to hearing how you get on with the JL CR-1 and of course some big fast subs. Good luck and please continue to post as appropriate.

 

[KeithR]

Given the above I am going to guess the following

-Rockport Avior/Cygnus
-Von Schweikert VR44/55
-Focal Utopia series

Look forward to seeing your reveal!

Don't think Rockport would be the one as they are not configurable to a person's room which appears a high priority.

I'm going to join you though and guess VSA.