It's a confusing and complicated world: SD vs HD
Last June, I said "
PS: I will save the discussion on SD vs HD on the 90.1 MIT speaker cables for a later time."; re-reading posts from around then, PeterA has been saying that he prefers the SD setting, finds phase issues with HD, but then liked the audition back in June, and as it turned out, I had hidden the SD/HD setting from the attendees, only to reveal in the end that we were all listening exclusively to HD all along. So again, it's a confusing and complicated world... and apologies that, despite best efforts, the following are still very thick.
So let me go back to what this MIT Articulation Pole technology is all about, and what the differences between SD and HD are, and what this Fractional Articulation Technology (FAT) is all about; the following is a direct quote from MIT as can be found at
http://media.theabsolutesound.com/buyers_guides/BG_Cables_2013.pdf
Robert Harley: Give us an overview of the Oracle Technology.
Bruce Brisson: The Oracle technology started with me designing an optimizing network for each of the octaves. We started at A440, because that’s the frequency most musical information surround and we worked on a network for each subsequent octave, up and down, from there. That was the original Oracle technology. We built on that by coming out with what we call HD, for high definition. What that meant was that we extended the Oracle technology to ten octaves and optimized for seven harmonics. The harmonics extend linearly from those octaves. Linearly means any harmonic has to be a whole number no fractions. If there is a fraction involved, and it deviates much, it becomes a spurious overtone. Instruments and voices produce overtones, but we don’t want to produce them in an audio signal-carrying conductor or in any amplifier. That covers the first two evolutions of the Oracle. From there we created Fractional Articulation Technology, which is where our research led us next; to make optimizing networks for the musical information between the octaves. Not extending forward from the octave by something greater than one, but the inter-octave spectra, which we call notes. The distance between one note and another note can be expressed in cents: 100 cents is a half step, 200 cents is a full step. With Fractional Articulation Technology we can optimize the spectra between the octaves.
RH: How specifically does the cable design do this?
BB: By forming what we call poles of articulation. This is simply how much energy a cable or a network stores and releases. Associated with that release of energy is a rise time and a fall time, which is accompanied by a settling time. We install networks in the cables which are formed from inductors and capacitors and resistors. We begin our optimization at A440, and then move out. So first we do the octaves then we do the six harmonics of the fundamental which is your 7th harmonic. Then we go between the octaves and we add networks to form the poles of articulation. But they don’t have the magnitude that an octave would they don’t store as much energy. We make sure that the rise time and fall times are within certain parameters that we have learned, empirically, doing this over 30 years, that the audiophile’s ear responds to.
Takeaways:
- Pole of Articulation = optimized energy stored and released at certain frequencies and their harmonics, and the key here is the rise and settling times (what we commonly call 'speed')
- HD = ten octaves and optimization for seven harmonics
- SD = unknown
- FAT = optimization between octaves
Next, regarding each PoA and what this energy stored/release is all about, the patents (e.g.
http://www.freepatentsonline.com/5956410.pdf) claim
power losses varying by frequency:
"[for a cable with total distributed capacitance of 90pF] using the impedance meter it was shown that at a frequency of 80Hz the impedance phase angle was found to be -85.46 degrees, and at a frequency of 431Hz the phase angle was found to be -87.37 degrees". He then mathematically calculates power losses at 80Hz as being 7.92% and at 431Hz 4.41%. MIT basically claim that there is exactly one frequency in any cable where the
power transfer is optimal (cable articulates perfectly at that frequency), and everywhere else there are losses (less articulation), small or large: "
The parallel capacitance of the cable is predisposed to store a greater magnitude of electrical energy than at other frequencies. Furthermore, the series distributed inductance of the cable is also predisposed to store a greater magnitude of electrical energy at certain frequencies than others."
These claims then lead to those articulation poles and another patent
http://www.freepatentsonline.com/6658119.pdf, wherein it is also claimed that the networks are chosen such that Voltage/Current phase angles (aka power transfer) are optimal at the chosen Pole frequencies, to compensate for the cable's natural Voltage/Current phase angle deviation (=power losses) at those frequencies. The end goal, they claim, is to produce a "
more uniform audio output". On the other hand, they also do admit that "
in other embodiments the network can be designed to emphasize certain frequencies". MIT have a a long and winding white paper on this Power Factor Correction in audio cables, that solves power transfer issues, at
http://www.mitcables.com/pdf/Transportable_Power_101.pdf if anyone cares to read it. BTW, Power Factor and PF Correction are not MIT terms, they are AC terms -
https://en.wikipedia.org/wiki/Power_factor - and PFC is widely used.
So, another takeaway:
- Pole of Articulation = attempt to optimize power transfer at that frequency (and per the above, also harmonics)
As if this is not thick and/or confusing enough, here's what MIT also says about the Oracle Matrix HD 90.1 speaker cables (and now 90.2):
By selecting HD and engaging F.A.T., one hears lifelike transients and improved detail within the octaves of complex music.
Where am I going with all this... back to the music and the differences between the SD & HD settings on my speaker cables, of course.
With the advent of the 30SV preamp (about which I will write up some other day) joining the 400RS amps, the differences are quite pronounced to me... Last weekend, I had PeterA and Al M over, right before the Magico S7 demo at Goodwin's. I left the setting at HD, and wanted to see their reactions (we never talked about the cables). It was obvious to me something bothered Peter, and I was hearing it too ever since the 30SV arrived, but didn't say a word: somewhat unnatural treble is how I would describe it; too emphasized at times, perhaps etched; 'glassy midrange' is also a term I have used in past threads that I googled...
For the past few weeks I have experimented switching back and forth between SD and HD, and here are my conclusions: I have been listening to HD before the 30SV because the brain was struggling to hear extra treble detail; at the same time, I was willing to ignore some midrange glassiness; with some material, HD was not offensive, it seemed to help, others it destroyed the treble. MIT's claim of "lifelike transients" and "improved detail" in HD w/ FAT are, in fact, evidence of unnatural edginess, not additional 'speed' (though to their credit, they do claim the setting one will choose will be system dependent). With a very high resolution system like mine at this point, the "HD" setting appears to operate like a contrast switch, which can be pleasing in some cases, but is just not natural. I played strings on digital for Peter and Al, and I could hear even from the sidelines that some violin notes just exaggerated and unnatural (and mind you, my panels measure virtually flat in their entire operating range, as shown in graphs a page or two ago). I repeated the same experiments with analog and the articulation switch on my MA-X phono cable, and turning it up (HD) creates an edgy, unnatural and perhaps tipped up treble experience.
At the S7 demo last weekend, and numerous product demos and private auditions at Goodwin's I have attended, they used the SD setting. Despite the fact 'HD' can be pleasing with certain recordings, it is just not natural, and possibly flat out wrong or not well executed. At this point, 'HD' is history for me. With SD, the entire midrange and treble is just more natural and realistic, period. Apologies to my friends for using them as guinea pigs!