> -----Original Message-----
> From: Paul Schreiber [mailto:synth1@...]
> Sent: Thursday, October 10, 2002 9:49 PM
> To: MOTM listserv
> Subject: [motm] Can there be a 'best'? [long rant]
>
>
> Hmmm...it seems that my use of the word 'best' gets me into
> more hot water than anything (except the time I said the Moog
> modular VCA had the same sonics as AM radio).
>
> I have even heard comments to the effect that there is like
> an unwritten 'rule' that musicians avoid using 'best' at all,
> because it implies a sort of arrogant snooty air (like
> "...so-and-so is the best bass player...") so the prevailing
> attitude is a neutral-at-best "hey, everything's cool!"
>
> This is 180 degrees out-of-phase with engineering. The whole
> ∗point∗ of engineering is to show/prove that A is a better
> solution than B, because blah blah blah. There are always
> multiple criteria about what constitutes "good, better, best".
>
> And certainly I have learned that the clashes between
> 'unemotional' engineering and 'emotional' music can go all
> over the place, and in many cases I think musicians don't
> WANT to know specifics as it somehow removes a 'layer of
> magic' between the musician and the audience.
>
> There are cases that seem to be a 'no brainer' to the
> engineering side and no amount of explaining will change the
> attitude of the user side. My favorite example:
>
> "Does MOTM come with banana jacks?"
> "No, because they are unshielded."
> "I like the colors, and the fact you can stack them."
> "Err...they are unshielded. Do you see any other pro audio
> gear using them?" "Well, synths that use them sound pretty
> good to me!" "That may be true, but what about a case when
> you DO get hum or noise induced?" "Why do you slag other
> synth manufacturers?"
>
> For me, any audio gear (not just synths) has no business
> using bananas because of this one fact. It doesn't matter if,
> to your ears, Synth A with bananas sounds as pure and clean
> as the driven snow. It's a bad ∗engineering∗ decision. The
> reality is, the use of banana jacks is a leftover from the
> early '70s based on cost. In 1974 I bough 50 Switchcraft
> jacks for like $110! 50 banana jacks would have been about
> $35. That is a significant difference. That gap is still
> there today: $1 versus about 30 cents.
>
> There are a 100 ways to shave pennies, that add up to
> dollars. Back in the '70s you could lower your pcb price by
> not having a solder mask and silkscreen. This made board
> stuffing a royal pain, and you can get copper foil
> delamination if you aren't careful. PC boards were relatively
> expensive as there were no CAD tools, no DRC (design rule
> checking: the schematic net list is checked by the computer
> against the routing). However, there is absolutely NO REASON
> that today's electronic products not have a solder
> mask/silkscreen. In fact, the pcb house I use charges
> ∗more∗ to leave it off because of yield issues. Yet, I still
> see these types of boards being made. If there is a
> difference, it can't be over $1 or so.
>
> Besides the obvious electro-mechanical scrimping, there is
> the design and the corresponding parts selection. In
> electro-mechanical intensive designs like a modular synth,
> the actual parts content can be as low as 5% of the overall
> cost (things like EGs and simple LFOs). If you serve a low
> cost market, in many cases you are self-limited in the parts
> selection. However, in many cases the difference between a
> nominal part and a REALLY GOOD part can be 50 cents. The
> mind-game you start playing is you start wanting that 50
> cents for yourself, and so you say "Heck, no ones gonna know
> the difference".
>
> I played that game for many years at Tandy and elsewhere. The
> 'moment of truth' for me came when I was designing a compact
> AM/FM receiver. Sanyo makes all the radio chips in the world,
> and they have 3 different FM demod chips, priced like 28
> cents, 40 cents and $1. I had over 50 schematics from every
> stereo receiver on the planet, all brands. About 60% used the
> 28 cent one, 40% used the 40 cent one and ZERO used the $1
> one. Even the most expensive, stand-alone FM tuner (Marantz)
> used the 40 cent one. So, I get demo boards for it and the $1
> one to measure and listen. Also, it turned out the $1 one
> used a 33 cent Toko tuned trap coil for the 19.2 demode
> filter (any hams out there?).
You had to ask: KC6UDI. :) Of course with the internet and really cheap
phone service, my ham days are nearly over.
> The difference between the 2 was STUNNING. The $1 + 33 cent
> coil blew the doors off the 40 cent one. Local classical FM
> radio sounded almost CD quality. I was dragging everyone into
> the lab for A/B tests and they all agreed the more expensive
> one was the way to go. Except my boss. "Can't afford it!" I
> said I could scrimp elsewhere (power supply was a favorite
> scrimp, ie the OB-8), but I realized that the ∗idea∗ of using
> "the most expensive" brought horror and shame to the manager.
> Picture the staff meeting:
>
> "Johnson, that new FM radio sounds great! Who did the
> design?" "Schreiber. Did I mention he used the most expensive
> Sanyo FM chip?" Stunned silence, followed by nervous paper
> shuffling and coughs. "He....did.....what??! Oh....my....GOD!!"
>
> This is why I am producing MOTM. I want it to represent what
> it means to be the best. I don't go out and hunt expensive
> parts for the sake of a "gold plated toothpick" as MOTM has
> been called. I just refuse to use inferior parts when there
> are better ones out there. The R&D in the 500 and 600 reflect
> this even more (there are rotary encoders that I could get
> for $7, but I'm using a Greyhill avionics-grade optical
> encoder with 1 MILLION full cycle rotations guaranteed with
> stainless steel housing. This way, 20 years from now, it's
> still working perfectly. My cost is about $26).
There are no toothpicks in MOTM, gold plated or otherwise! ;)
I agree with you in just about all aspects of this email, but OTOH there
is no reason to put a part in that is twice as much if one 1/2 the price
meets the intended design goals. What I mean is if you put in a dip
switch (only for an example) to set some parameters, there is no reason
to put in a super high usage part good for 250,000 cycles, when the
switch at worse case will get moved 1000 times. At that point it's just
wasting everyone's money except the people that manufactured that great
part.
> Even if my use of 'best' make you squirm, I'm not gloating,
> bragging or slagging. This is just the engineer in me talking
> facts. I wanted MOTM to raise the bar, to show the 'unclean'
> that there is an alternative way to design audio gear (like
> looking inside a Mark Levinson or Krell audio amplifier). I
> think I have succeeded, and it's not just me. I have other
> people designing HW and SW that have more ability than me in
> many areas. I am grateful that they feel that MOTM is where
> they themselves have decided to contribute.
Most of that high end stereo stuff is quite excellent, but you gotta pay
$$$ for it. I listen to stereos now that just kill my ears, but I can't
tell the owners of those products that. They are happy with their
purchases, and that's great. Just as some people are happy with Doepfer.
I had a Paia 8700 years ago and I thought it was great until I learn
what else what out there. (plus those 1/8" jacks crackled after a while)
I'm certainly not worried that my motm will fall apart when I'm 60. :)
> I admit that when I hear the end result, the music, I want to
> shout from the rooftops. I just make paint and brushes,
> others use the modules to ∗create∗. I listen to every audio
> synth demo and CD I can find. If I was in the market for a
> modular, it is a 'no-brainer'. I have yet to hear ANY other
> synth demo that made me want to chunk MOTM in a rat hole and
> say "That kicks my butt! It's time to close shop." Rather,
> the opposite.
>
> OK, the therapy session is over :) Please don't start in over
> bananas (Les!). I know several of you have replaced the 1/4
> with them: don't tell me (my poor babies! Sob....)
>
> Paul S.
Keep up the great work Paul!!
Tony