Yahoo Groups archive

SynthModules

Index last updated: 2026-04-28 23:14 UTC

Message

Spec benchmark myths

2005-01-25 by Gary Chang

I hope that this doesn't seem too far OT, but I have been biting my
tongue through this discussion about specs, and thought that I would
toss a few observations out....

1).  Although the mathematical dynamic range (not signal-to-noise) of
CD players is 96db (16 bits x 6db = 96db), no CD player actually plays
at that spec, due to error correction.  In fact, among pro 16 bit
products, only older technology, such as the NED Synclavier, which did
not employ error correction and was linear 16 bit, actually came
closer to the mathematically perfect 96db.  Most CD players actually
playback with around 14 bit resolution (84 db).

2).  Signal to noise in audio components has slowly edged its way to
quieter numbers, mostly due to functional necessity.  When people
recorded to 3 track analog tape and consoles had 8 channels, a
Telefunken Tube mic preamp with 50-something db s/n was fine;  it was
too noisy when 24 tracks became the state of the art, so discrete
component transitor technology emmerged (NEVE, Trident, API), which
had s/n in the 60-something db range.  Fast forward - Signal to noise
of today's PREMIERE audio components (Massenberg mic pre's, Daking
Limiters, Crane Song Compressors, etc., each costing $2,000 and up) is
around the 100-something db, designed for use in 48+ track 24 bit
recording (most of which muster about 20bits of dynamic range - about
120db).

What is significant to note is when you talk to designers of this
stuff (I have discussed this with Jeff Daking and George Massenberg),
they all talk about POWER SUPPLIES!  "What is the PSU range?"  (Most
of this high end stuff is +-24v, so discrete component transistors are
only used).  In the Class A stuff (single ended), only a +24v PSU is
used with $100 transformers converting to and from the balanced world.
 The basic core of their design world is their own personal design of
the discrete transistor op-amp.  Serge created such an op-amp in the
70's, based on the then-state-of-the-art DBX opamp.  It is what is
found in the UPAP, and all of the VCAs in the Serge.  But it only
+-12v, not +-24v.  There is a huge difference in performance, due to
the PSU.

The PSU is the key in isolating your synth from noise on the ground
pin of your studio.  It's implementation also directly effects how
quiet your system is.  

When I built my Wiard system, I had several discussions about this
with Grant, and decided to have two PSUs - one for the oscillators and
mixers, and one for the rest.  The result is a rock stable and quieter
system than when I used one PSU to power everything.  The single most
dramatic upgrade that you can do for improving your synthesizer system
is to upgrade your power supplies.  Module location (and therefore PSU
association became very important - if I put a noisy Woggle Bug on the
same supply as the Oscillators, the draw from the led display of the
constantly changing state of the Bug would effect the audio performance.

Sorry for the rant....

Gary

Attachments

Move to quarantaine

This moves the raw source file on disk only. The archive index is not changed automatically, so you still need to run a manual refresh afterward.