Inter Sample Overs vary with digital process ?

[Correction: Subsequent data has called most of the theoretical speculation below into question.  It turns out that SACD's do not have consistently low ISOs, they can possibly have the highest ISOs of all, as I detected with the the 'Poulenc' SACD released by Linn.  This is an orchestra with a large pipe organ, and if I had associated low ISOs with higher quality, let alone previously believing all SACD's to have low ISO, I would have believed this to require a minimal 2.0dB headroom for ISOs.  But in fact it set my current record...it required a a mind boggling 7.5dB headroom.  And it was somewhat inconsistent.  It might work with as little as 6dB headroom....or it might not.  It apparently produces a peak around the same size as the 96kHz sampling interval, and how high it registers depends on how much of the peak occurs in any one sampling interval.  So there is a large random element to how much headroom is required.  I've also noticed that on some DVD-Audio discs with 96kHz sampling rate.

It does still seem that Reference Recordings and 'Discipline' (the brand behind the 40th Anniversary King Crimson DVD-Audios) produce consistently low ISOs.  I would venture they seem to know what they are doing.  But the peak ISO level seems not to depend on DVD-Audio vs SACD.  Nor does it even have a fixed relation to music genre (rock vs classical) or does it necessarily have anything to do with the recording "Dynamic Range" or how much compression was obviously used---though it's possible some compression was slipped into the Poulenc at key moments and that's what gave it such high ISOs.]

I have begun fairly systematically copying all my SACD's and DVD-Audios into 24/96 copies for my hard drive.  And I'm noticing a few weird things.

Remember that I set my gain structure so that with an 880 Hz tone recorded at precisely 0dB (generated by Audacity) a +4 level would be the highest possible before clipping.

But for actual recordings, I have to set the level between -2.5dB and +2.5dB, giving them at least 1.5dB extra headroom and as much as 6.5dB appears to be needed in some cases.  This headroom is required for inter sample overs (ISOs) where in between the samples the signal peaks higher than 0dB.

Benchmark claimed that ISOs could be as high as 3.01dB and that many digital decoders failed to allow sufficient headroom.

BUT I am finding 3dB headroom is way insufficient.

EXCEPT, in some cases it isn't.  Some discs seem to cluster around 2dB headroom required, up to 2.5dB in some cases.  Which discs are these?

1) Reference Recordings HRx (174kHz/24bit)  seem to only require 2-2.5dB headroom.  These are some of the best sounding recordings ever.

2) SACD's generally only seem to require 2dB headroom  (I would have expected SACD to be the 'wildest,' but in actuality, it's the 'tamest.')

3) The DVD-Audios in the King Crimson 40th anniversary DVD-Audio boxes, which seem only to require 2.5dB headroom.  These are truly spectacular in audio quality (or in the case of Court of the Crimson King, merely way better than ever before).

On the other side?

1) Many hot sounding recordings on DVD-Audio, including Elton John, and especially Steely Dan.  (I've long noticed that.  I figured Steely Dan cranked up the compression so high it's bleeding out the ISOs). 

I wrote this off for awhile, but now I'm getting a pretty clear feeling that the recordings which require the least ISO headroom are the ones that sound the most natural, laid back, and 3D.  The recordings that require the most headroom sound highly processed.

You may be astonished to see some of the pictures when I post them.  Having the huge ISO's means the rest must be scaled back in a recording, losing dynamic range in the midrange.

I think the excessive ISO's occur when the the anti-alias filtering is insufficient, and full scale high frequency garbage gets into the digital encoding.  THAT's what's causing so much overshoot.

DSD has such a high rate it captures and controls all the HF crap by design.  So SACDs do not suffer from excessive ISO's.

Likewise, the PMI analog to digital encoders used by Reference Recordings, which must have superb filtering.

And whatever King Crimson was using in the 40th anniversary DVD-Audio set.

Now, it would be interesting to know what Steely Dan used in such things as the Everything Must Go DVD-Audio, that produces such high ISOs.

Perhaps it's not the digital process, but the amount of processing (including compression) used before digital encoding that is the culprit.  But the biggest ISOs look too big for just that, IMO.  Still, the sonic variation might just be coincidental, the least processed recording just happening to use the digital processes which produce the least ISOs.

This is also a function of the Oppo BDP-205, which is apparently not headroom constrained itself.  But I've found the height of the ISOs not to change with different reconstruction filters.  I think they are inherent in the data and and the degree of oversampling used, with higher amounts of oversampling 'revealing' the true ISO height (because you are filling in more of the points in between the points).

******

Update:

Now it appears that that the ISOs in DVD-Audio discs are all over the map.  They can be < 2.0dB, for example, in Queen's A Night At The Opera.  They can be < 2.5dB such as in every King Crimson 40th Anniversary DVD-Audio box.  Or, they can be as high as 6.5dB, as in Steely Dan's Everything Must Go, Elton John's Goodbye Yellow Brick Road, or Frank Zappa's QuAUDIOPHILIc.

SACD's still appear to be consistent at around 2.0dB, and everything from Reference Recordings is there too.

The Audio Hobby

The audio hobby means different things to different audiophiles, and isn't that wonderful?

Not always, perhaps.   It can be made awful in many ways.  People can fall victims to unfounded beliefs that cause them to waste time and/or money.  It can lead to the virtual inability to listen to music anymore, in extreme forms of audiophilia nervosa.

It can lead to endless bullying.  I think generally the best attitude is non-judgemental: live and let live.  However all the same I believe audio is filled with frauds of many kinds, both on the big and small levels.  You probably already know I side mostly with the audio objectivists as to what kinds of audio beliefs are founded and which are not.  All the same, I don't take it as my mission to change anyone's mind.  I suppose, in some cases, even I could be wrong too.

Anyway, nowadays I think generally it's a waste of time to compare good amplifiers, DACs, cables, power conditioners, or anything of that ilk.

All have been tested endlessly by audio objectivists getting only null results.  With my lack of patience, I'm unlikely to do better, if I follow all the proper procedures.  If I don't, then the result may be meaningless anyway.

My normal result with good amplifiers is this:  I start the auditory level matching process by matching the apparent loudness with both amplifiers to make it identical in fast A/B testing.  Once I have matched the level, I attempt to match the quality of the sound.  If one seems to have more bass, or highs, I assume that's because it's actually louder.  In this second phase I make only the smallest adjustments, 0.25dB at at time (I'm fortunate to have a first generation Emotiva Stealth DC-1 with 0.25dB adjustment.  That means, on average, the best case is within 0.125 dB of being exactly correct, which is close enough to pass the 0.1dB minimum in my experience.)  Ultimately, on every amplifier I've tested, I can make them sound identical simply by matching the level that closely.

Now I did not do such procedures when I thought for several years my then go-to amplifier, the Aragon 8008 BB, was sounding a bit harsh.  Back in those days, I found myself avoiding listening to the FM for very long.  It drove me insane.

I later found the distortion had risen to 0.7% because of low bias.  After bias adjustment, I got it back to 0.07% and sounding fine.

But I know that from years of experience and not A/B tests.  And measurements which make that experience believable.

And that's another thing.  One should often believe measurements, when they are meaningful and honest.  Not necessarily specifications.

Anyway, distortion can be a factor down to 0.1%, so products having higher than 0.1% should generally be avoided.  (In electronics, anyway, where it's easy to do better.  There's hardly any speakers that can do as good as that.)

In audiophile land, there are often electronic products with higher than 0.1% distortion.  And sometimes they sound better.  I think what's happening is that in some cases products with predominantly 2nd order distortion may fix recordings that were made with high amounts of 3rd order distortion.

Furthermore, boosting the amount of 2nd order distortion, which tends to occur with zero feedback designs (feedback works best at suppressing 2nd order distortion) can add additional "spaciousness" and air to recordings lacking those things because of poor production.

Things like this may work on some recordings and not others.  It may work best on recordings that are fairly simple, like a few instrumentalists.  Not on works of great complexity, like a full symphony orchestra.

Euphonic adjustments are like that.  Generally it's best to stick with low distortion, wide response, low noise, because that's the combination that works overall best on everything.  Basically what the objectivists say.

Other claimed magic requirements in design, however, are sold on the basis of faulty comparisons, typically failure to match levels very well.

Anyway, if feedback free amplifiers, and electrically charged cables, or whatever makes sense to you, go for it.

The best is when we're not bullying people over such things, one way or the other.

Even being forced to make a decision is a kind of bullying.

We are generally not designed to discriminate among audio reproduction systems.

We don't have a 'memory' that works very well for making such comparisons.  We don't store 'experience A' in anything like the raw form that would make for a good comparison with 'experience B.'

To be reliable at all requires, as the objectivists always say, instantaneous A/B switching.  Other than that, perhaps exhaustive training.

Furthermore, always being assigned to make the equipment comparison detracts from the process of having the most enjoyable and enlightening experience from the music, appreciating the music itself rather than arcana of possible sonic differences caused by different audio equipment.

Fine, the uber subjectivists say, just see which piece of gear gives you that most transcendent audio experience.

That's basically impossible, because each time you listen to the same piece of music you get a very different experience.  Symphony orchestras often like to prove this by playing a Premier (first ever) performance of some work twice, sometimes even without warning.  Few guess it was an identical repeat.  The identical music doesn't provoke an identical response.  And for a very important reason.

The you listening to any work the second time is now older and wiser, having already heard the music before.  The brain has already stored memories and made new connections.   Usually that opens up entirely new realms of experiences.  While closing down others.

(Curiously a work with more 'movements' can be heard more times without seeming repetitious.)

As many have often opined, that is what audio should be mostly about, the music, and quite often isn't.

(However in line with non-judgmentalism, I prefer to say that any mode of enjoying an obsession with audio reproduction is fine.  If your thing is building amplifiers that test conventional theories -- fine.  If your thing is arguing about whether such things can or do make any difference -- fine.  As long as I'm not to much detained by your obsessions.)

But along those line, I fear the excessive denigration of standard audio engineering practices, as often occurs in the writing of audio subjectivists (including the one I always loved to read anyway, Harry Pearson) tends to promote rather than discourage audiophilia nervosa and therefore lack of being able to enjoy music.

But some people swim in one thing or the other, so whatever works for you.  Doing 'comparisons' is also a way of just hearing things twice, which may itself be beneficial.

I have dubbed such demonstrations magic shows, and typically enjoy them even when (that is, all the time) my core beliefs are unshaken.  

I myself only drifted into near audio objectivism in my late 20's, after doing carefully constructed experiments I felt would 'confirm' my beliefs in tweakdom.  That's the path of many noted audio objectivists.  First they were fully taken in, then they decided to do some tests.

After the recording, DAC, and amplifier problems are 'solved,' is there anything left?  Two things, the loudspeaker/room (or headphone) interface, and the selection of the music itself.  Neither is a problem that will ever be 'solved.'

R128

Here is the EBU R128 standard for measuring "dynamic range" that Roon uses:

https://tech.ebu.ch/docs/tech/tech3342.pdf

Dynamic range can mean different things.  In the context of technical measurements of an amplifier or transmission system, dynamic range is pretty similar to "Signal to Noise."  What is range from the lowest signal that can barely be resolved to the maximum.  So a dynamic range of around 120dB is where many of the best units are (chips can be as good as 130dB).

But this has nothing to do with they "Dynamic Range" of program material.  But what has long bugged me is, what does this mean anyway, because every signal, no matter how high the peak, ultimately has to go back through zero again.  And how close it gets to zero depends on how finely you can measure it, assuming it's a continuous waveform.  So this is back to to the "dynamic range" of amplifiers again.

But this is not what audio/music people mean by the Dynamic Range of program material.  Their window of analysis is not 1 uS, say, the resolution of a decent digital scope.  Their resolution is the loudness in a 3 second time window which must be overlapping.  The spec is unclear to me, but in specifying "dBFS" it is clear to me they don't mean instantaneous levels but something like average or probably RMS levels...that is to say levels related to the equivalent sine wave measured with RMS.

And then the Dynamic Range is specified as the difference between the 10th percentile and the 95th percentile of these 3 second time windows.

Now it's still not clear to me how Roon uses this in level normalization.  There must be other parameters such as the maximum level, etc, because the R128 I've just described only relates to relative and not maximum levels.

My Simple Surround System

I started fooling around with Surround Sound in 2005, and that's about all it was, fooling around.  I had purchased 5 speakers from my brother-in-law, and I experimented with placing them on the back counter of my kitchen, where they quickly got in the way and had to be removed.

The current system really got going around 2018 or so, but it's still based on my 2005 Yamana HTR-5790 receiver, which has always worked very well for stereo and everything.  It's got very good amplifiers (for 7 channels!) and good performance everywhere.  It can decode common surround signals from coax and optical connections.  What it doesn't have is HDMI inputs and the ability to receive surround sound via HDMI.

That hasn't been an impediment, since I can either take the coax low res digital straight from my Oppo BDP-95 for movies, or the seven channel analog output from the Oppo for high resolution audio multichannel discs.  Normally I just use the analog for everything, so all my digital decoding is done by the Oppo.  The only way this is a "problem" is that I'm not getting high resolution (from high resolution audio discs, not movies) from the few high resolution discs I have in pure digital form for the fancy processing in the receiver.  But that doesn't matter much, since the Yamaha receiver doesn't do fancy processing on the seven channel analog inputs.  Instead, I do all the basic digital processing required in the Oppo, which has menus for setting the delays and levels for each channel.  So the only way that this is a "problem" is that I'm not getting all the fancy digital processing applications like Audessy, that may correct for EQ and phase.  And I wouldn't even want Audessy much, except the later versions which allow you to do the fine tuning on your phone.  And that set of Home Theater Processors are still very expensive even used, because they have that feature.

Anyway, I'm getting the seven channels and routing them to amplifiers in the Yamaha and thence to speakers.  With levels and delays set in the Oppo.  That's pretty much all there is to it.

Except, what about when there is 5.1 channel content?  I could play it back just as that, which is theoretically fine.   But because my side speakers are not optimally placed, I often find it better to play 5.1 channels on my 7.1 speakers by duplicating the side content in the back.  When I do that, I lower the level by 2dB so it still balances OK.  I use a rotary switchbox (made by dB systems no less) to switch from pure discrete 7.1 to 5.1 expanded to 7.1.  And then I have a preamp for the sides having a level control with two marked positions.  I use the higher position for discrete 7.1 and the lower one for fake 7.1.

My project to add optimally placed side speakers has gone nowhere in 5 years since I bought the required wall hangable small speakers, currently still in my bedroom to remind me to install them.

(I have tried many other solutions for the 5.1 to 7.1 conversion.  For some time I used a well known box from the  1980's.  After using it a year I found it distorted the sound intolerably.  Then I tried using various delays and EQ's.  Finally I decided that simply duplicating the sides in the backs, and lowering the level by 2dB, worked better than anything else.  I also tried a historic Integra processor, but found it did nothing useful, and was basically a pain in the neck because it only produced distortion if input or output levels exceeded 1 volt.  The Yamaha is good at least to 3 volts on inputs and outputs.)

I could solve this 5.1 to 7.1 conversion problem problem "better" with a fancy Home Theater Processor that might cost $5000 (do they still make those???) or more, but it's not been worth it to me when I could have afforded it (and now it's simply unaffordable).

Anyway, my simple idea is implemented with the units on top of my kitchen rack.  The bottom DEQ box is for the rear speakers and it only does level adjustment for the backs, plus level and spectrum displays (so you can see at a glance if the backs are doing anything).  The level adjustment feature that this box actually does could just as well be done by a preamp (like the upper two boxes) but I happened to have the DEQ and not another preamp.  (It inherently adds about 10mS of additional latency, which I have compensated for in the Oppo adjustments.  When used for the fake 7.1 the additional delay is a bonus that makes it sound a bit better--like an even larger room--but isn't that important either way)

The upper DEQ box is for the subs and currently does nothing more than level and spectrum display.  (I thought it was also doing some eq but it appears not.  At some point in the last couple years I bypassed it.  I'm not sure that was out of design or necessity--such as it might have been adding hum).  Currently the upper DEQ box is non-functional, it appears to have the usual power supply issue.  But since I'm not doing any processing there anyway, it doesn't affect the sound.  But it was very useful to have the spectrum display on the bass because I could measure ground loops exactly and work to eliminate them.  The spectrum display on the subwoofer signal enabled me to solve and fix about a dozen ground loops over the years.  The complicated kitchen electronics (my central computer, video,  and audio for the whole house, plus TV and radio) are prone to those.  Now the box that did the helpful bass spectrum displays needs fixing.

The upper two preamp boxes are for the sub and sides.  It allows convenient setting of the levels, which I need to do for the sides when changing from discrete 7.1 to fake 7.1.  Also historically I used to mess with the sub level a lot depending on recording.  But I've had it dialed in pretty well for everything recently and hardly mess with it at all anymore.  Still I like have "controls" that I can just reach up and control as opposed to complex apps which may not do anything until you take another measurement.

The stereo frequency response (including subs) is quite flat, though I think some of my low frequency optimizations that used to live in the subwoofer DEQ would have made it flatter there.  Still it sounds pretty good.

Rethinking the Oppo BDP-205 Filter Choice

My last post got 'updated' with a long discussion and testing of the Oppo BDP-205 filter choices.  I had been using the Linear Fast filter, thinking that was the best of the best.  But it seemed to induce about 5 clipping events beyond +4dB headroom I allowed for inter sample overs (ISOs), which shouldn't be happening at all (or at least Benchmark seemed to say it was only necessary to allow 3.1dB headroom for inter sample overs).  This is not necessarily the flaw of the Oppo or any of it's digital filters, but a deviance from my previous expectation and a seeming objective variance between the different filter choices.  (See update below.  I have subsequently determined that my counts of clipping events was wrong.  It looks like all the fast filters generated the same number of clipping events.  They just looked different.)

The other choice that looked best to me at that point (on technical considerations and examining Archimago's measurements) was the Brickwall.  I found that only induced 1 clipping event above +4dB, thereby seeming objectively better, plus having the best numerical specifications on noise, distortion, and loss at 20kHz.

I'm NOT going to choose any of the slow filters even if they eliminated such clipping events (in fact, I'd expect they might) because of their leakage effects.

But there's one other filter that might be the best of all, and it's actually Oppo's choice for the default, so it's apparently what they thought to be the best.  It's the Minimum Phase Fast filter.

This filter has the curious effect, like all other minimum phase filters, of moving the ultrasonic ringing past any transient, rather than being on both sides (acausal).  Intuitively this seems better to most, including me.  I was brushing it off last time as "unnatural" but in fact it is the natural thing you could get with real circuits rather than digital simulations, if you could make those circuits well enough (which in practice isn't possible).

We'll I'm long past thinking about ultrasonic ringing in the first place.  I'm more interested in low noise, extended response, and those being equal I'd consider the phase response.

The minimum phase fast HAS higher noise and distortion than the brickwall...but also it appears to have more ultimate bandwidth (a big plus) at least according to Archimago's spectrum graphs (which didn't look right, because the brickwall had the lowest loss at 20k, but on the graph it was cutting out well before 20k steeply).  So given the possibility that min phase fast has the widest bandwidth, and nearly as low distortion, that could make it the best.

Anyway it occurred to me it might not have these these above 4dB inter sample overs in the first place, and if so it would be an obvious choice.

But the test shown below shows it has the same single clipping event over +4dB as the brickwall.  So it's "equal" in that respect, and better than the linear phase which had 5 such events.

But...looking at that actual clipping event...it makes far more sense than with any of the other filter, to my analytical eye.  In fact it makes so much sense, I'm inclined not to "repair" it at all, as there's no repair that would preserve the underlying high frequency transient it is apparently trying to show (which previously I insisted had to be electronic...and it might be...but that hardly matters here) without smearing it.

It looks to me best just left very slightly clipped, for it's clear the clipping is right at the upper bound of where it's going to be anyway.  There's likely so little difference made in the single clipping event it's not worth lowering the recording level -0.5dB to avoid it (though a test might be warranted).  The clipping looks to be hardly making a difference...the peak would only reach a microscopic amount higher anyway, judging from the trailing ringing which is now concentrated on the "after" side making it easier to understand.  The other filters give very messy looking results that defy repair altogether, compared to how this looks.

Min Fast +4db clip (look for the ringing)

I'm now leaning towards leaving the Oppo at the factory default Min Fast setting.

Update:

My counts of clipping events may have been wrong.  I was just checking visually.  This can be misleading and depend on how much time is being displayed and also when exactly it starts.  I discovered this by cutting the first section of the Min Fast recording and discovering that there were no clipping events at all, instead of just one.  The one had disappeared because of the differing start time.  Then when I magnified and scrolled, I saw several events.  I probably made that exact same mistake with Brickwall.  Lets assume for now they all have the same number of clipping events.  (That also explains a recollection of another file I didn't report.  It was also Brickwall but had several clipping events, just like linear fast.)

So this number of clipping events was an entirely bogus analysis.  What still looks good is the concentrated and natural way the Min Fast makes each inter sample over clipping event look.  They just look right, whereas all the others just looked awful.

Right now I have no evidence that the factory default filter isn't the best, and one subjective guess that it looks best.  (I doubt I could hear the difference...especially in a double blind test.)

Update May 17

I listened to Min Fast filter last night and thought it sounded great.  Pure, harmonic, and no digital artifacted sound.

But now it appears that my bit about the look of the ISO clipping events (once again, the fact that there is clipping is not Oppo's fault, it's mine) and it's damned hard to tell even which one looks best.  The Linear Phase Fast filter, which I now see is Archimago's preference, does give very short ringing, shorter even than Brickwall, though it has that annoying pre-ringing also.  Archimago simply argued on merits, that linear fast is like previous Oppos and most players, and admitted that he's never heard a difference among filters nor was a difference found in earlier testing of similar filters.

The Audioholics tester (Gene Dellasala) also tried very hard and could not hear a difference among the filters.  His advice was leave it at the factory setting (Min Fast) and worry about more important things.

The difference in noise among the Brickwall, Min Fast, and Lin Fast filters is negligible.  The Brickwall noise level is slightly better with -119.0 instead of -118.8.  The THD numbers are identical at a barely measurable 0.0008%.  That indicates none of these three filters have significant leakage.  The Brickwall apparently cuts off a microscopic amount faster somewhere above 20kHz resulting in the tiniest bit of extra noise reduction but the benefit is so small, one might as well use the more "natural" looking Min Fast, and maybe that slight added bandwidth is a good thing (especially for someone like me, using supertweeters).  The Min Fast may have the widest bandwidth as it does have the lowest loss at 20kHz (-0.19 for the Min Fast vs -0.21 for the Linear Fast vs -0.20 for the Brickwall), and since that's the one and only superior spec for that filter, perhaps it's why Oppo chose it as the default.

Actually none of this matters to me  anyway because I only use the Oppo to play high resolution discs, with sampling above 44.1kHz, or SACD's.  Standard discs I simply rip to my computer and send the CD quality bits into my system and the DAC in the Oppo doesn't matter at all.  With high sampling rates on high resolution discs and SACD's, any of these filters is way more than good enough.  I could even go with the slow filters and not suffer significant alias leakage.

I think I'm going to follow Gene Dellasala's advice, which I was trending to anyway.

I also don't have any MQA discs (do they exist?) and when I stream the Oppo DAC isn't involved (and I no longer use a streaming service that supports MQA either).  Some apparently like the Apodizing filter with MQA.  But for other uses, the Apodizing filter is unappetizing to me in general because it adds a tad of high frequency ripple (almost certainly inaudible, but why have it anyway).

The Lavry AD10 Microscope

My 12 year old Lavry AD10, which had been left running 24/7 since I bought it (I won't be doing that any more) died when during some power surges when my home foundation was being repaired.

I sent it to Lavry, who repaired it for $600 by replacing the main board.  So basically I have a brand new Lavry AD10 (which appears to still be in production).  It might even be better than the original unit (purchased in 2010) ever was, but I never did complete measurements of the Lavry before.  I figured I didn't have anything good enough to measure it with.

The Lavry AD10 has a signal to noise specification that appears to be about 4dB better than the AD converters of the Tascam DA-3000 recorder (117dB vs 113dB).  Since the DA-3000 is a newer device, and noise specifications depend heavily on test protocols, I really didn't know which was better.  But I figured the Lavry was probably better, and I was right.  My measurements yesterday suggest the Lavry is indeed just about 4dB quieter than the DA-3000.

I've long said that digital converters are about the best thing made in audio because a lot of attention has been focused on them.   This may be less true of analog to digital converters nowadays (up until about 10 years ago, typical analog to digital converters were better than the digital to analog converters, somewhat counterintuitively, but now consumer DACs can have better than 130dB S/N which is as good as megabuck ADC's).

I've often said they were so good, digital converters are often and generally better than preamps.  Perhaps even my Emotiva XSP-1, which I envision as about the quietest digitally controlled preamp you can get under $10k (above which perhaps Mark Levinson makes even quieter digitally controlled preamps, though FWIW the Emotiva specs are better than all but the most expensive Levinson "Reference" models and about the same as those Reference models, so we're already getting about as good as it gets with the Emotiva XSP-1, in performance anyway).

Digital converters are probably better than nearly all non-digitally controlled preamps too, which in some cases (probably not many) might be still quieter.  But I like the digital controls for setting levels precisely and also ensuring perfect stereo tracking.  I got so frustrated by channel tracking I switched to digital preamps in 2000, my first being a Classe CP-35, not only because of the perfect level tracking but to my ear* it even sounded better than passive and non-digital preamps I had been using before.  I've found no need to go back to non-digital preamps since then.  (I found the XSP-1 to be far better sounding* than even the Classe).

Now I have studied this issue in some detail.  It appears that there is an unimportant flaw in the noise spectrum of the Emotiva which is made invisible by the noise of the DA-3000.  But like a microscope, the Lavry AD10 clearly shows this flaw, an ultra low frequency noise in the left channel only (both of my two Emotiva XSP-1's show this same identical flaw).

It's worth recounting how I came to use the Emotiva XSP-1 as my living room preamp for all "analog" sources (including the analog outputs of DVD-Audio players which preserve the full 24bit resolution while the digital outputs truncate it to 16 bits).  For many years I was using a passive switchbox to switch among such sources, and the level adjustment on the Lavry to set levels for digital encoding for my downstream digital processors.  But the Lavry level adjustment lever, which is great for long term settings, it a pain to reset on every disc.  A big pain.

Then I also discovered that the noise level of the big XSP-1, which I had originally purchased for my less high end bedroom system, was even lower than the tiny XPS-1 phono stage I was using in the Living Room, which itself was lower than my dB Systems high gain preamp (which stunned me, because the dB systems preamp and the XPS-1 and the XSP-1 all use the same low noise preamp chip for phono pre-amplification, but somehow it was about 10dB quieter in the XSP-1 as compared with everything else).

So I needed an XSP-1 in the living room just for the phono preamp alone.  And all my bench measurements of the XSP-1 suggested it was about as perfect as I was able to measure, with S/N better than 110dB and distortion below 0.005%.

So the combination of convenience and performance led me to use the XSP-1 not just for phono preamplification, but also "preamplification" (mostly downward level adjustment) for the special digital sources (DVD-Audio**, SACD, and HDCD) that cannot be output in their full resolution through SPDIF which I need for my digital crossovers and equalizers.

Now using the "Lavry Microscope" I can see a flaw in the XSP-1 more clearly.  This same flaw was invisible amidst the mere 4dB higher noise level of the DA-3000.

I obsessed over this Emotiva flaw for at least one day.  But although I toyed with the idea of using a passive switch rather than a preamp for the digital sources again, I have once again concluded the XSP-1 is "good enough" not to bother with that.  Slightly higher noise levels below 5 Hz (and still below -120dB) are just not that important.  That's probably what the designers of the XSP-1 thought too.

[Pictorial section being expanded.]

First I wanted to re-measure the setup I've been using, Oppo BDP-95 into Emotiva XSP-1 into Tascam DA-3000.  I also measured the Lavry with open inputs (not shorted, an oversight).  By itself, with open inputs, the Tascam had about 1 dB less noise than with the full chain.  That suggests that the Tascam was generating more than half of the noise, which would be about what I'd expect.  

The picture below shows about 5 minutes of recorded noise amplified digitally by Audacity by 94dB.  So we're taking a close up look at the noise, about as close as we can get as there is only about 0.3dB of headroom on the right side, which is the full chain of equipment, whereas just the Tascam by itself is on the right.

Things didn't look so good with the Lavry inserted in front of the Tascam to do the A-D conversion (using the Tascam wordclock signal for synchronizing):

The (top) left channel noise looks "noisier" somehow.  Notice that the right channel noise is significantly less than when recorded directly by the Tascam in the previous picture.  Almost 4dB lower in fact, just as the specs of the two AD converters suggest.  And if you carefully compare the right channel to either of the Tascam direct recordings, it's actually lower, but not as much as the right channel, because of some extra noise that's being added somehow to it.

My first concern, in fact the main reason I was doing these tests was to be sure that the Lavry, just back from an expensive full board replacement, was now working properly.  And this first measurement didn't look good.  And possibly because I go about things in a more round about way than necessary, I didn't fully prove that the Lavry was fine and good until about ten measurements later.  Well I was also concerned about my Emotiva and Oppo which might be just as expensive to repair.

And it turns out that the problem itself is no big deal.  But I've decided to tell this story mostly as I experienced it, so it becomes clear that way.

It seemed to me that the first and easiest thing to do would be to reverse the channels.  The right side of the picture shows the normal connections and the left side shows the reversed connections.

The extra noise in the left channel moves to the right channel after I reverse the cables.  So therefore, whatever it is that is causing the extra noise in the left channel with normal connections must precede those connections.  It possibly comes from the cables themselves, so I then tried putting the Lavry connections back to normal and reversing the output connections at the Emotiva:

After all the messing around with cables I had done just to get the audio XLR cables moved from the Tascam to the Lavry I worried that I might have damaged them.  But it was clear that it made no difference whether the cables were left-to-right reversed at the output or the input, the result was exactly the same.  So the XLR cables between Emotiva and Lavry were exonerated from causing the added noise.  

At this point I decided to "mute" the Emotiva preamp.  I'm not sure whether I dialed the volume all the way down, or turned it off.  (My recollection is that I was going to turn the volume down, but I made the recording without even turning the Emotiva on.  So I called that "muting the preamp" in my too-brief notes.)  The strange extra noise in the left channel went away completely.  From this point on, I was convinced that the problem was not in the Lavry, however I didn't fully verify that until a later test in which I put shorting plugs in the Lavry inputs.

It was at this time that it occurred to me that prior to having the Lavry repaired, I had taken great pains to connect the AC power to the same power strip as the Emotiva and all the front end components.  It looked impossible to make this change without disconnecting and moving heavy equipment like the almost 50 pound Denon DVD-9000 from the rack in order to get at the power strip behind it.  But somehow I managed to get it done anyway.  I even used the same 3 foot SJT power cord I had been using before.

The results seemed like a significant improvement when I was first examining them, but on checking them now, I'd suggest they were not any change at all.  However in either case the extra noise in the Left channel did not go away.  That part had not been fixed.

Then I removed extra cable after extra cable from the preamp in a whole series of tests I won't bother you with here because they were all the same.

Finally I went all the way with this sort of test by removing every single cable from the Emotiva except AC power and the output cables connecting to the Lavry.  I selected a balanced input (2) which was shorted with XLR shorting plugs.  The extra noise did not go away (note that the channels are still reversed because the output cables were reversed at the Emotiva).  The Emotiva left channel (bottom) noise looks different from previous images because now we are looking at just one minute instead of 5 minutes.  After spending hours doing these tests, I decided I could just as well get by with 1 minute recordings.  Now it's becoming clear that the extra noise is a very low frequency baseline shifting around -98dB in level (these pictures are amplified 94dB by Audacity, so a full scale noise would be -94dB down):

Now that I had the XLR shorting plugs out, I plugged them straight into the Lavry.  The result was the lowest noise of all with no extra noise in the left channel whatsoever.  To compute the peak unweighted noise level, I bring up the Amplify dialog one more time, and it shows me how much more amplification is possible.  In this case it is showing 5.78dB more amplification is possible  Since the starting amplification here is 94dB, the signal to peak noise level is 94+5.78  = 99.78dB.  Applying RMS adjustment and A weighting would probably improve this to almost 120dB or maybe better.

I concluded there was something "wrong" with the Emotiva XSP-1 preamp that was causing a small but measurable extra noise in the left channel.

So the next day, I moved the pile of mostly spare equipment away from the side of the rack so I could swap the living room and bedroom XSP-1's.  The living room XSP-1 measured above was purchased in 2018.  The bedroom XSP-1 was purchased in 2014 (it's also the second generation btw) but was repaired (basically refurbed) in 2019 by the factory.  As part of the repair they do a full AudioPrecision test which verifies it meets all specifications.  Since the repair, it has not been used very much (and I was careful to keep it turned off when not in use, something I should have been doing from the beginning because even if nothing else the display gets rather dim in about 5 years of on time).

As it turned out, my other XSP-1 was almost identical to the one measured above.  The added noise in the left channel looked almost (but not entirely) identical.  The noise in both both channels was just over a half dB lower, so I decided to keep this other XSP-1 in the living room from now on (and it won't get unwanted wear now that my home control system turns it the living room preamp on and off using the trigger signal).

It was only now I started to probe the noise itself.  You can see recurrent peaks in the range of 2-3 seconds (see the selected range in the above picture).  That means the noise is in the range of 0.5 Hz to 0.33 Hz.  We're talking about very low frequencies here.  And we're also talking about very low levels too, certainly 97 dB or so below peak level.  That's good, but not as good as the midrange noise floor which seems to reach down to -150dB or so (better than I should be able to measure, though it seems I can).  Here are the spectrum plots of the two channels, first the better one.  Remember to subtract 94dB from the levels shown in the spectrum.

Emotiva Right (better) channel

Emotiva Left (worse) channel

The main difference here is that the noise seems to rise below 4 Hz faster in the left channel.  In both cases the midrange and high frequency noise level is extremely low.  The signal to noise in the midrange is around 154dB (94+60).

This low level ultra low noise (ULF) noise in one channel would not even appear in an "A" weighted noise until you were many decimal places out.  Very low frequencies like this simply aren't audible and usually aren't measured either.  Listening to the noise (amplified by 94dB by Audacity) the left channel with the ULF noise if anything sounds softer and more pleasant than the right.*

Many preamplifiers are going to have input capacitors which completely filter such low frequencies out.  And sooner or later it is almost certain some component will filter it out.  And it's very low in level (-97dB or more) to start with.

A nearly identical very low frequency noise in just the left channel is a flaw in both my two units, including the second which was sent back to the factory for repair 2 years ago and hasn't been used much since.  It looks to me like a design flaw that Emotiva isn't concerned about, and truly it isn't of much importance either.

In fact, it's rather surprising that the Lavry AD 10 has frequency response which extends down to 0.33 Hz, but it appears like it must.  Whereas the Tascam DA-3000 does have about 4dB more noise, as the specs suggest, which partly covers up the Emotiva ULF noise, and partly it may have low frequency filtering which also hides it.

So at the end of the day, the difference seems to be that the Lavry captures the ULF noise of the Emotiva both because the Lavry is so quiet, and also because it has low frequency response that seems to extend to something like DC, whereas the Tascam does not.

But just because you can see something with some kind of measurement, doesn't mean it's important.  I can't imagine a single good reason why such a ULF noise would be important not to have.  At this point, I don't even have any data about whether other preamps have similar issues.  With most tube preamps, noise would overwhelm such small effects, and in my experience with tube preamps they flopped around their DC levels not just in microvolts but in millivolts if not volts.  So there you could say the XSP-1 has a bit of that kind of "tube character" in very attenuated form.

Though my theory is that it's the output servo loop of the left channel being closer to the power supply or computer or something like that.  An issue that could be resolved with yet another board layout revision.  But the XSP-1 is beyond revisions now, it's been discontinued (which I'm not happy about either, I think it is a very fine preamp and I don't believe Emotiva or anyone else has a close enough replacement for me now, though fortunately I don't need a replacement now, my two XSP-1's are working fine, good enough in my opinion--did I mention the midrange noise floor looks to be in the vicinity of -150dB, that's the kind of thing that actually counts).

The next day I decided to test an alternative theory, that the problem was being caused by a ground loop in the (coaxial) clock signal from the Tascam to the Lavry, possibly causing jitter.  (They are plugged into different AC power strips and that is almost unavoidable.)  I found that using the internal clock on the Lavry and the Sample Rate Converter on the DA-3000 made no difference.  (My notes are unclear if I also disconnected the clock, which wasn't entirely necessary for these tests but would rule out a ground loop on the input circuit.)  Also I shorted the left input on the Lavry but using the usual clock cable, which eliminated the extra noise in the left channel.  And remember the channel reversing experiments and shorting experiments above, which were also all done with the usual clock cable.  It seems to be disproved that the clock signal is in any way involved in this extra ULF noise.

Here the left channel is on top, but the Lavry left input is shorted.  The bottom right channel shows a little ULF noise, but unchanged from the right channel in previous measurements and far less than the left channel.

Unhappy that I had failed to note if I had removed the cable or not, I decided to do another followup test this time making sure I removed the cable.  By this time (and also in the previous set of tests) I had already figured out a small optimization.  I reduced attenuation on the Lavry by 3dB so that peak level is now -8 (around 5.5v peak) instead of -11 (3.8v peak).  Then I increased the gain on the Emotiva to +4.   The Emotiva can just as easily handle that balanced output voltage.  This optimization could in theory increase the S/N by 3dB.  (I set the Emotiva level to be just before the level it causes clipping on the J-Dunn test.  Since I reduced the Lavry sensitivity by 3dB I might have just raised the Emotiva to +3, but it now seems like +4 is the correct level to reach closest to 0dB on peak signals.)

The following tests showed that removing the clock cable (which requires setting the Lavry to internal oscillator and enabling the SRC on the Tascam) makes no difference at all to the ULF noise in the Left channel. 

Clock Cable being used, +4dB higher level

Clock Cable disconnected, 4dB higher level

It is looking like the level readjustments may also have reduced the ULF noise from the Emotiva.  The new noise levels (with clock connected as usual) are amazingly good, peak noise -99.2dB in right channel and -98dB in left channel.  A weighted RMS S/N would be in the vicinity of 120dB.  This is the full chain including Oppo, Emotiva, and Lavry.

Right Channel Noise

Left Channel Noise

Now I wanted to see how much the gain changes were making this better, so I went back to the old gain setting on the Lavry (-11dB reference level, ie +11dB gain).  In fact, the gain changes (previous set of measurements) were making a pretty big difference.  The new noise level (with Emotiva at +1 to optimize J-Dunn headroom as was done in last two measurements) is -95.5.  So there has in fact been about a 2.5dB improvement in lowering the ULF noise from the Emotiva by lowering the Lavry gain 3dB.  The other channel noise level has not changed as much, presumably because it's mostly being caused by the Lavry at this point.

Emotiva at +1dB, Lavry at -11

Most of the prior measurements were done with Emotiva gain at +0.  This was artificially making the S/N look "better" than it actually was.  Here is a replication of what I was doing before.  It "looks" like the peak noise level is -95.9 but that is misleading because full scale cannot be reached.

Emotiva at +0, Lavry at -11

Because I'm not boosting the level, I wanted to see how much effect this has on distortion caused by the Emotiva which may be putting out more than 2V balanced into the Lavry.  I used a CD on which I have recorded digitally generated (by Audacity) 880 Hz at maximum level (0dB) but no clipping.  I changed the Lavry gain to 0 for these tests.  I tested this "maximum output" CD at Emotiva gain levels of +0, +4, +9. and +10.  At Emotiva gain level +11 this signal was clipping the Lavry (even at 0dB gain on the Lavry so I could not even make the measurement).

At +10 (which is 6dB higher than my new standard level) there were no visible peaks above the -90dB bottom of the Audacity spectrum (but is this misleading?) and the bin value was -101dB but the peak value was -28.8dB.  That would mean about 4% distortion.  My guess is that the "peak" value is more representative of the harmonic distortion peak but it might be exaggerated.  All I can say for sure is that somewhere above +8 we turn a corner and distortion starts rising. 

Emotiva at +10 with max 880

The very best harmonic distortion measurements were with the Emotiva at +8.  At that point, the peak level of the second harmonic is is -110.6dB, corresponding to THD of 0.0003% or better.

Backing the Emotiva down to +4dB gain, the distortion rises slightly to -107.7, or about 0.0006%.

Someone lacking knowledge would just set the level to +8 (and the corresponding -4dB reference level on the Lavry--as it's clear the numbers have to add up to 12 to just avoid clipping the Lavry above 0dB).

However, I've long known about inter-sample-overs.  In between samples, the signal may rise above the 0dB level when rendered with oversampling--which must fill in the points between the points.   My sampler is likely to read these inter-sample-overs at least some of the time (or maybe nearly all of the time if I'm sampling at a higher rate than the original, which I need to do for decoding HDCD's for example).

I've long assumed that inter-sample-overs could reach as high as 6dB, though I've only ever actually measured about 4dB.  If in fact inter-sample-overs could get that high, I'd have to set the Emotiva gain no larger than +2 and the Lavry at -10 to avoid having them clip the Emotiva.  (But read on...)

Now, if you're not understanding inter-sample-overs and why I seem to be setting the MOL from the Emotiva at 6dB lower than the optimal distortion level (+8) for no good reason, read Benchmark's description of inter-sample-overs.

Benchmark sets their headroom for inter-sample-overs at 3.5dB.  Perhaps I'd measured 4dB because of rounding in the low resolution readouts on the I've been using.  Benchmark says the  maximum theoretical seems inter-sample-over is 3.01dB around 11kHz.

If only 3.5dB headroom for inter-sample-overs is needed, since I can't adjust in fractional units, I'd have to allow 4dB headroom below the optimal distortion at +8, thereby putting me right back to the adjustments I was assuming in my first "optimization," +4 gain on the Emotiva and -8 "reference level" (ie 8dB gain) on the Lavry.

So that now does look like the optimal adjustment, taking both headroom and noise considerations into account.  No only will no real signal cause rising distortion from the Emotiva, but no inter-sample-over will either (and many converters don't even handle those well...perhaps the true deficit in CD reproduction from the beginning).  In fact with the +8 setting I've allowed at least 0.9dB more headroom than necessary for inter-sample-overs, and possibly more (as I never bothered to measure the +9dB setting on the Emotiva).

(Note that the Oppo BDP-205, which was playing but paused during the noise measurements, and played the 880 Hz maximum level signal, has 2V XLR output.  Some say this is "all that's needed for any amplifier" but in fact my Krell FPB-300 required 2.8V and much higher to reach the true peak levels.  Can't always rely on inter-sample-overs to bring you there either--those depend on high frequencies usually.  I would have expected standard XLR level to be twice the RCA level, so 4V balanced. Anyway, with the Emotiva at 0dB it will also be putting out the 0dB level of 2V which is pitiful I think for a balanced output.  By boosting that 4dB, it's being boosted to 3.2V, and allowing another 3.5dB of headroom would put us around 4.7V.  Surely the Emotiva balanced outputs are still in their low distortion range at that point!  In fact, in the Secrets of Home Theatre and High Fidelity measured the lowest distortion at 5V !  (It bothers me they didn't probe the question "how high does it go", but as it turns out, 5V is all I need.  The 8dB gain setting on the Lavry officially corresponds to a voltage of 16dBu or 4.89V.)

Now I sort of remember that I'd come up with this setting (at least the -8 reference level) many years ago, driven by the Two Against Nature or Everything Must Go, both Steely Dan albums had incredible inter-sample-overs.  I figured out this same setting (and I think it was this one) empirically driven by the need to play the a Steely Dan DVD-Audio, because when I first played it, it clipped my sampler like hell.  This was how I personally discovered inter-sample-overs before I even read about them.  Benchmark refers to a particular Steely Dan track on Two Against Nature in their discussion and then analyzes a few others on that album and a few other albums notorious for the highest inter-sample-overs.

Then over time, I couldn't remember my results, and I drifted back to the obvious (-11).

Update May 12 2023

Oops, I wasn't thinking clearly.  While the "-8" reference level (ie 8dB gain) on the Lavry is still correct (that means the Lavry clips before the Emotive begins to distort...as things should be) the corresponding +4 gain is ONLY correct if there are no Intersample Overs (ISOs).  Only if I played a recording without ISOs, like my 880 Hz 0dB test disk, would that be OK.  Albums with a lot of ISOs would need to be played as low as +0.5 on the Emotiva, if Benchmark's claim that the maximum ISO is about 3.1dB.

I got plenty of ISO clipping when I tried to copy one of my all time favorite DVD audio discs, Pulse, by The New Music Consort.  It clipped with the Emotiva set to +4dB gain, clipped at +3dB, and so on.  Actually, it was still clipping at +0dB, which suggests maybe even Benchmark was wrong, ISOs may occur even greater than 4dB.  In this case, I was recording a 24/96 disc at 24/96.  I would think that would minimize ISO but maybe not.

The clipping at 0dB made me think that maybe my filter selection on the Oppo was in not optimal so I checked that.  I was thinking perhaps I was using one of those audiophile 'slow' settings.  But I found that I was using Linear Phase Fast.  That should be fast, in fact I'd think it was the fastest and best so that's why I chose it.

I looked over all the settings again.  I wouldn't want any of the 'slow' settings since those always leak aliases.  I looked over Archimago's investigation of the different filters.  IMO the only settings worth considering are Brickwall, Minimum Phase Fast, and Linear Phase Fast.  The Apodizing filter is weird having rippled HF response.  Minimum Phase Fast is Oppo's default, and it puts the ringing entirely after a pulse and the ringing is quite long, I consider that weird too.  The 'Corrected Minimum Phase Fast' is actually a slow filter having similar high frequency cutoff as the slow filters.  My subjective judgment of all the graphs and statistics is that the Brickwall filter is best.  It has the lowest distortion and noise, and second to the flattest response at 20kHz (according to the Rightmark evaluation) with the Minimum Phase Fast filter being 0.01dB flatter at 20khz, hardly worth sacrificing any noise or distortion for.

Audiophiles are inclined to dismiss brickwall filters I suspect mostly if not entirely by suspicion.  But this is not your grandfather's brickwall filter, this is a very precise brickwall filter.  (Archimago has no actual test of high frequency phase, which might show some way it could be inferior.  But the response is so flat and extended, phase errors must be pushed out in very high frequencies too.  And this is probably achieved without using any of the non causal FIR tricks that the linear phase filters use.)

Anyway, I tried it, and it might just be chance (caused by dithering or something) but it had fewer clipping events than the Linear Phase Fast filter.  With the Brickwall filter I simply had to repair one clipping ISO and there was almost 0.5dB of additional headroom left.  For the Linear Phase Fast there were 5 clipping events having to be repaired, including the weird wider than usual one shown below.

Brickwall (top) vs Linear Phase Fast

The Brickwall doesn't clip with this apparent digital error on the recording shown above, whereas the Linear Phase Fast does cause clipping (above +4dB headroom allowed for ISOs).  And the Brickwall ringing is slightly more compacted and seems to make more sense.  Generally, I liked the way the ringing on the ISO's looked with the Brickwall, in addition to having fewer of them clip.  I think I'm going back to Brickwall (it had been my option until a few months ago).  My feeling is that the Brickwall is less tricky and more honest than the other filters.  Among the filters, Brickwall may well be the mathematically least complicated.  Many of the others HAVE to be implemented with FIR digital filters because they are acausal.  They are approximations of things that are impossible to achieve with ordinary circuits.  Brickwall can in principle be implemented with IIR filters (but the Oppo may well use a FIR to approximate a more perfect Brickwall).

Indeed all the ISO clipping events on this recording are caused by ringing on very rare and unusual short impulses which look like digital errors.  Filtering out those pushes the peak level down to -1.8dB.  But even THAT peak level is still being caused by ISOs.  With no ISO's, the level should be -4dB, allowing my LF sinewave derived level setting of +4dB gain.  Here is what another of the ISOs looked like with the Linear Phase Fast filter.  Note that it appears to stem from a transient less than 4 samples wide at 96kHz (actually the impulse itself is less than 2 samples wide, or about 48 kHz, and therefore unlikely to be of acoustical origin).  Repairing some of these peaks may make them MORE audible as the inaudible high frequencies are replaced with audible lower frequencies which probably shouldn't be there either.  All the better to have fewer ISOs to repair, as with the Brickwall filter.

I made the last Pulse recording with the TV that I had used to set the filter choice still running.  To see if that made any difference, when the recording had completed one more pass (it keeps repeating endlessly on these Classic Audio DAD's) I stopped it, and kept recording for a minute, then turned off the TV, and tried to run for another minute (but it was shortened to 40 secs because the audio file had reached maximum length).  Amplifying the noise by 94dB, the point at which the TV was turned off is simply not visible.

TV turned off midway here makes no difference

That noise level, around -99dB peak unweighted*** (so approximately -120dB A weighted) is way lower than on the recording itself.  When I amplified the initial part of the recording, including before the Oppo started playing the disc, you can clearly see (at 36dB amplification) where the disc starts playing, the noise just rises out of nothing (the earlier -97dB peak unweighted noise from my chain of equipment is invisible at this level of amplification):

*** The actual value of noise measured may depend on how long you measure, when you measure peak noise.  There is sooner or later always a higher peak.  This is especially true given the ultra low frequency noise found in the Emotiva left channel...it is the primary driver for this noise dispersion, despite being at a tiny -103dB peak level itself.  In one second that low frequency level has barely changed, but in one minute it can do a lot of up and down dancing.

So the numbers I'm seeing now from the Emotiva at +0, the Lavry at -8, taken from the interval shown in earlier picture before where the TV was turned off, are these

1 second        -100.5dB

10 seconds     -98.7dB

1 minute         -97.7dB

2 minutes       -96.7dB

An advertising dept might see how small of an interval they could possibly go...  But nobody uses peak noise for specifications anyway, they use average noise weighted, but I can't do that as easily with Audacity.

I might establish a "10 second standard" for measuring this, though brag about the 1 second level.

(*In sighted listening tests.)

(** I know that DVD-Audio discs can be played through HDMI which preserves the full original 24 bit resolution, though it's still not possible for either HDCD or SACD on my system.  But it's still way better to have PCM computer files on my hard drive for automated playlist playback, and no files can be created from HDMI because it inhibits digital copying.  Unlike some, I believe 24/96 digital encoding and decoding is essentially perfect for any audio purposes, and the present results support this by showing digital conversion to be good enough to show flaws in an excellent preamplifier.  I see no reason to store DSD files as they would later need to be converted to PCM anyway for playback through my digital crossovers and equalizers.  So I might as well do the conversion (SACD, HDCD, or DVD-Audio to analog to digital 24/96) before storage, and it saves me a lot of hassle later too, when I'd be using the exact same components to produce 24/96 for my crossovers and equalizers anyway. )

The Button

It's been at least a month since I finally implemented The Button (the Listening Position EQ button).

I keep The Button next to the listening position chair.  When I sit down in the chair, I can press The Button and it enables the special Listening Position EQ which has much too much deep bass content to be used when I'm playing the system as background music (which is what I do most of the time).

So, when you press the button, you start getting the Full Effect at the listening position, which otherwise sounds very bass shy compared to the rest of the room (and the entire house).

Last year as I was doing some listening position EQ adjustments by ear and measurements, I decided that a Listening Position EQ was absolutely necessary to restore the missing "punch" in my system.  One problem is that there most of the notching out of room modes benefits the sound in the rest of the room.  At the listening position, these modes cause cancellation instead of augmentation.  So two EQ's are necessary if you use the system for both background and serious listening, especially for my problematic room set up.

(Why didn't I fix this by "fixing the listening position and speaker location(s)"?  Because my living room is decidedly a multipurpose room I also use for parties, and the only real solution to getting the bass right in a small lengthwise room like mine is to put the speakers in the middle of the room and listen at the back, as suggested by local late audiophile and engineer Alan--and what he did in his dedicated upstairs listening room, which had similar dimensions.  Putting the speakers in the middle of the room just wouldn't work for me.)

So last year I created two different EQ's.  "Listen" for the Listening Position and "Back" for Background.  I could fairly easily select them with the chairside EQ I was using for level and EQ adjustments, using stored memories.

But in December, the particular EQ unit that I was using as my chair side EQ died.  It is still on my bench awaiting repair.  The key stumbling block is that I have to deal with lead-free solder for the first time, and to do it right I also have a new powered solder sucker that may take some getting used to.  So while there has been one urgent thing to do after another here at the palace, otherwise known as Peterson Studios and Laboratories, that particular project of repairing the EQ has been procrastinated on time and again.  Which is not unlike many such things in the past.

Meanwhile I was forced to recall, guess, and scrape my blog for hints as how to recreate the "listening position EQ" using my Bass EQ unit, since sadly it was one photograph I never took before the chair side EQ died.  I did that in January.

But the truth is, I wanted to have The Button even when I also had the Chairside EQ working, simply as I believed it would be more convenient to use the button than fiddle with the memory menu on the EQ.

In one fantasy, I'd just sit down in the listening chair and The Listening EQ would automatically engage, then disengage when I got up.  That *would* be nice.  But even just having a simply "On/Off" button to press would be pretty close.

I finally got to it during a particular stretch of wintry weather.  The basic technology is stuff I'd used before, but had to refamiliarize myself with.  

I used an Midi Solutions Footswitch Controller (which I originally purchased many years ago for controlling a Behringer DCX crossover unit to change the absolute polarity, for absolute polarity blind testing.  It was controlled by the line level path of my ABX switcher).  I programmed the Footswitch Controller to send Program Change 2 when the circuit is made, and Program Change 1 when the circuit is broken.  The Midi Solutions app strangely doesn't code this from any menu option.  The only menu option available is to send a Program Change when contact is made, but nothing when contact is broken.

Meanwhile, the "circuit" is made and broken using an Insteon IOlink, same as I used for other automation tasks in my living room stereo.  (This is the 4th IOlink in use for the Living Room stereo).  The IOlink conveniently has connectors either for turning an internal voltage on or off, or making or breaking an external connection.

So then the IOlink is controlled by my Insteon system, which itself is controlled by my Universal Devices ISY994i home controller.  The Button is linked to control the IOlink.

It all works, but apparently (and strangely) the Insteon RF is a bit weak at the listening position, so sometimes I have to press the button more than once.  It's obvious when the Listening Position EQ is turned on, because a message temporarily pops up on the DEQ screen, the Memory light turns on, and also a light on the IOlink lights up (I could turn that light off, but I've decided to keep it as an additional indicator light), because if I were fiddling with the memories beforehand the Memory Light might not correctly indicate the Listening Position EQ, whereas the light on the IOlink is unambiguous.

I also have home control system automatically return the EQ to Background EQ when I press the Bedtime button at night, which also turns off the stereo.  So in the morning when I turn on the background music (the local FM classical station KPAC) from the bedroom it doesn't start blasting bass because I had been using Listening Position EQ the previous night but never turned it off.  This is important.  Automatic resets are important when an automated system is restarted.

[I plan to add pictures to this post.  But lack of pictures was what delayed me from posting about it for over a month, so I decided to proceed without them at first, which has been my usual approach anyway.\