Here is an e-mail response from Victor to a question I asked him regarding a few things:
"We voice changes to Joseph's components whenever we can or feel we need to over three speaker systems--Joseph's Sophias, Cedar's Coincidence, and my ESS Transars.
When Joseph and I A-B'ed the new caps vs the Hovlands on the 2.1 through Joseph's Wilson Sophias, we felt the Hovlands had a narrower bandwidth (less top and bottom and a bit more forward in the midrange), were definitely less open (that is, more compressed), and less transparent (more veiled). Because they captured musical texture and harmonics less well and the attacks were less clear, the Hovlands were smoother sounding, but to our ears at a very great cost. Joseph visited Cedar and they conducted an A-B of the new changes to the 2.1 (minus the later power supply changes) vs the equivalent 2.0. Cedar and Joseph both felt the new caps etc. made a dramatic improvement over the old. Had the changes been felt to be subtler, we would have voiced over my speakers to confirm, but the feeling of all of us over the two speaker systems was that the improvement was dramatic and not subtle.
Joseph's Sophias and Cedar's Consonance speakers are three ways. We prefer to voice with three ways because of the inherent limitations of two ways. Designing a two way speaker is very difficult because the crossover point is usually at or near the point at which the ear is most sensitive--1000 Hz-2500 Hz. This means phase anomalies play a critical role, and also this is the point at which woofers sometimes become a little more ragged and tweeters are operating at the limits of their low frequency range. To minimize the possiblity of nasal elements occuring when the human voice is played, two way speaker designers often design their crossovers to slightly de-emphasize these frequencies--not greatly but slightly, maybe a couple of dB. The old Advent speakers which were such popular best sellers, had a very pronounced but narrow hole in the midrange at the crossover point of a few dB. But the designers counted on the human ear's ability to fill in gaps to compensate for this. (I enclose an essay I wrote for Joseph on Audio Listening which touches on this phenomenon.) Nevertheless if one compared an Advent with a quality three way speaker, one could distinctly hear the hole in the midrange at about 1500 Hz.
You refer to the "breath of life" in your posts. Much goes into the "breath of life" but certainly midrange "presence" is a significant component of it. If you have a two way speaker, and if it, as do so many, has a slight dip in the midrange (not a great dip, just a modest carefully contoured depression along those critical frequencies to prevent any hint of honkiness or a nasal quality to vocals), you might welcome the Hovlands slight midrange emphasis--more than welcome it, need it to give you that "presence" so much a part of the breath of life.
Having committed irrevocably to the new caps, we now needed to listen very, very carefully to what we had now created.
This led to our feeling that the extended frequency response of the new caps was exposing high output breakup at crescendoes. To counteract this, we made the power supply changes that increased the voltage. The result of these changes was to increase dramatic contrasts, to supply extra reserves of power when crescendoes occurred, and thus to end the high output break up which we had found objectionable. It also had the effect of "softening" or "subduing" the high end and with it the objectionable edginess, and to all our ears it did this without rolling off the high end. What I'm trying to convey is that it is common for even careful listeners to confuse the added brightness caused by various forms of ringing and loss of high frequency control with extended bandwidth. We did not alter the bandwidth, or the frequency response, only the break up and edginess, but subjectively to listeners who are noticing this subliminally, this comes through as a softer or subdued top end.
The "warmth" or "darkness" you hear is not in the caps you mention, but inherent in the power supply changes. To get back to the old TP 2.0n, you would need to sacrifice all the gains of the power supply mod, which gains are extensive in our opinion. Changing the value of those caps will not do it.
I will ask Joseph if there is a simple way to add a little emphasis to your top end. Meanwhile, please give us a chance to think through these issues before you make any changes to the unit. Despite our asking you not to make changes, it seems you made changes and could not wait until we told you how to make them in a way that did not affect the traces. Sherod, we try to respond as rapidly as we can, and I think we do a pretty good job. I don't think from the time you put your last question re the Hovlands to us until we responded was even five days. That shouldn't be too long to wait.
We will try to respond to this new request as soon as we can, but what you want to accomplish is not easy because so many elements went into the changes you hear. Isolating one from the other is not easy. We will try to have an answer for you within a few days.
If you want to post any part of this answer on the Discussion Forum, feel free to do so."
