strangeandbouncy said:
Hi John,
I in no way meant any disrespect.
none taken...
Your knowledge and understanding far far far excedes my wildest dreams. And, on reflection, I have not tried any 2520 in a mix-bus. Nor 47910 for that matter. Personally, I am lucky enough to use three different mixbuses daily. 2 discrete, and one chippy, so I am well aware of the differences. I stlll think that most of the *sound* of a console is from the mixbus. It must follow that this sound is governed not only by the devices used, but also the circuit topology, or all those chippy 80s and 90s consoles would sound the same - 5534 n all.
Like i said I wrote on this subject of console design back in 1980
My judgement then was that hard parts were mic preamps and sum bus, with sundry other opportunities to make mistakes.
Since then IMO digital consoles have made summation theoretically perfect, as long as you are in the digital domain.
I know of an API that now has Forsell 993 on the mix amps instead of 2520 . . . .
Whatever,,, I recall Paul Wolff proudly showing off his first SMT opamps at an AES show, a long ass time ago...
I do like your chilli analogy, and I totally agree with the idea of being able to bypass . .
NOw I wonder what your opinion of this might be . . . .
http://www.weiss-highend.ch/oem.html
. . . . and Abe, dont even think about this one . . . . .
OK, I am not a fan of DOAs. Even if well done they are just not practical, and if poorly done they are POORLY DONE. I repeat that I doubt I could personally make a DOA better the SOTA we can now buy off the shelf with distortion down some 150 dB or so.
Glancing at the Weiss white paper I see plenty that I like....
He borrowed the inductor in the input LTP from Deane Jensen (RIP) who came up with it decades ago... I see lots of classic discrete design techniques (cascode buffered input devices, current source loads, etc). Of course there is nothing wrong with standing on the shoulders of others, I do it all the time. The only thing in there you can't do better inside an IC is the LTP inductor and perhaps the dielectric used in compensation/bootstrap caps (C1, C2)
I suspect that Weiss opamp sounds neutral (i.e. it sounds like the feedback network is telling it to sound). Whether it sounds audibly different than a SOTA uber-opamp, my speculation is not likely. In theory the flat open loop gain of the Weiss means error terms do not cause phase shift, an arguable advantage when operating at huge closed loop gain, but one shouldn't operate opamps at high closed loop gains, they are far better used for general purpose modest gain applications. For high closed loop gain we have other topologies better suited for the high noise gain to exploit.
Sorry about the veer... but you asked...
JR
Note: I am apprehensive about THD specs with that many zeros after the decimal point, since most are not measurable, with that much resolution at low/modest gains as we want to use them. Applying large closed loop gain, and then extrapolating to calculate the silly low distortion figures, makes me a little suspicious that the design is actually optimized for the high closed loop gain... Not the worst sin in the world, but a question that make me wonder.
While I won't lose sleep over the angel's foot steps as they tip toe around on the pin head in their slippers. I'm pretty sure I can hear -100 dB down let alone -150dB...
