200e

I am a fan of the Buchla 200e but like many people I can only dream of
being able to afford a system. 

In the meantime I want to learn more about how they work and to better
understand why they cost so much (and why they are worth it).

One thing I would really like to do is try modeling some of the core
modules in a Nord Modular 1. I know that someone did this before for
the Complex Waveform Generator module but unfortunately he lost the patch.

I have also been trying to find block diagrams of how some of the
Buchla modules work but the information seems to be scarce. I was also
told that users who receive the User Manual are told not to distribute
it. This seems odd to me. I would imagine that any prospective buyers
of a Buchla system would definitely want to read the manual ahead of
time to gain a far better understanding of the systems capabilities
and to help select modules to purchase.

Is there anywhere that I could get some of these details?

thanks!

the cost of the 200e is due mostly to the ultra-small scale of their
production; basically B&A is still only don buchla, hanging in there
since 1964 (this is not Moog... we don't have t-shirts...)

but there are other factors. small-scale production is relatively
cost-effective for analog stompboxes, compared to digital stuff, which
requires expensive components, super expensive development tools, and
exotic development skills. and boutique pedals still often cost 4 or
5x their mass-produced counterparts. scale that up to the kind of
labor, materials, and r+d that a massive hybrid modular requires, and
factor in a very very limited production capacity... the math is not
good. it's kind of amazing that the thing exists at all.

anyway, the topic of cost has been discussed a great deal on this
forum and elsewhere. (check out the front page of the yahoo group...)

so, moving on: the only real reason don doesn't distribute the manual
digitally is that it undergoes revisions as the firmware changes and
he doesn't want obsolete information floating around. 200e, eternally
beta... i have been wanting to make a public guide to the system,
since this comes up about once a month, but i'm pretty damn busy as it
is... i would love for someone else to do it, and a couple people have
offered but nothing has materialized as yet.

sometime, though... i promise...

let's see: we don't have block diagrams. they would look really weird
and probably not that useful, definitely more trouble than they'd be
worth for me and don and anyone else who's actually working on the
system. the only useful level of abstraction is the schematic and
firmware spec, which we don't distribute for obvious reasons.

(the block diagram for the 261e would be like: [knob input] ->
[compute frequency]->[sine wave]->[some super weird shit happens with
vactrols and opamps]->[audio out] ... hm, not so helpful...)

as for modelling the 259e, the basic idea of crossfaded wavetable
synthesis (or wave-terrain synthesis, in a fancier representation) is
bread-and-butter in the digital world, should be pretty easy with yr
Nord. the 259e weirdness comes largely from what's actually in the
tables, though... google chebyshev polynomials... chop those up and
use yr imagination... try stuff out... best i can offer.

the 261e would be very hard to model digitally because the waveshaping
is pure analog, and i would be hard pressed to provide even a
quasi-useful analysis of the circuit. people have a damn hard time
modelling a stupid vacuum tube... i'd guess that the best approach to
modelling a 261e waveshaper or a vactrol gate would be perceptual
("add some harmonics to a sine wave..."); a deterministic physical
model would get hairy real quick.

same deal with, e.g. amp modelling: a block diagram of an amp provides
only deep and well-coded clues to the construction of an analogous
nonlinear transfer function (the computationally viable digital
equivalent).

hope that helps...

/eb

On Wed, Aug 6, 2008 at 3:29 PM, intellijel <danjel@intellijel.com> wrote:
> I am a fan of the Buchla 200e but like many people I can only dream of
> being able to afford a system.
>
> In the meantime I want to learn more about how they work and to better
> understand why they cost so much (and why they are worth it).
>
> One thing I would really like to do is try modeling some of the core
> modules in a Nord Modular 1. I know that someone did this before for
> the Complex Waveform Generator module but unfortunately he lost the patch.
>
> I have also been trying to find block diagrams of how some of the
> Buchla modules work but the information seems to be scarce. I was also
> told that users who receive the User Manual are told not to distribute
> it. This seems odd to me. I would imagine that any prospective buyers
> of a Buchla system would definitely want to read the manual ahead of
> time to gain a far better understanding of the systems capabilities
> and to help select modules to purchase.
>
> Is there anywhere that I could get some of these details?
>
> thanks!
>
>

Hi Ezra,

I was definitely aware of the tabou topics for this board but I guess
I didn't make clear that I was not challenging or questioning the
price. I do hope to be able to buy one in the future!

I am actually electrical engineer and have done some design work for
music technology companies. The wizardy done by Buchla is obviously at
a much higher level than what I do but I am still very curious :)
As a result I can easily estimate the time that would go into building
a developing something as complex as most of the Buchla modules and it
is easy to see where the high costs come in, especially when compared
to mass produced products like something by Moog, Korg, Yamaha etc.
I am also aware of how much more work and tools are required for
creating any type of digital system. One of the companies I worked for
had the following departments who were divided between many projects
(matrix company topology)
1. R&D for developing algorithms (most of whom were PhD's
2. DSP department for creating OS's and efficient code from the algorithms
3. uC dept. for creating UI and interfacing to the DSPs,
4. hardware dept. to design the system circuits and debug hardware,
5.  a PCB layout specialist
6. mechanical engineer to measure and model all physical components,
7. industrial engineer to design the casings/look/ergonomics
8. technical writer + graphic designer to create the manuals
9. techs to perform safety and EMI compliance testing
10. manufacturing dept to manage component stock and help assemble
prototoypes
11. Test dept. for methodically testing all functionality

I am probably forgetting stuff! So when I work on my own projects at
home I am fully sympathetic with anyone creating complex systems with
limited resources - both in terms of tools and manpower.

I can understand from some of your explanations why the block diagrams
may not help. 

There are some schematics floating around on the web where people have
tried to emulate certain Buchla modules, maybe I will have better luck
there.


btw, any chance that Buchla will make some smaller one panel synths
like Serge has done with the Animal and Creature? It would be great to
be able to get a small taste of the Buchla!

sorry, i realize this whole thing was pretty snarky. not my intention.

i have some more useful thoughts for you (i think) but it'll have to
wait til after work... and maybe a stiff drink.

prosper,

\eb

On Wed, Aug 6, 2008 at 4:11 PM, ezra buchla <ezra.buchla@gmail.com> wrote:
> the cost of the 200e is due mostly to the ultra-small scale of their
> production; basically B&A is still only don buchla, hanging in there
> since 1964 (this is not Moog... we don't have t-shirts...)
>
> but there are other factors. small-scale production is relatively
> cost-effective for analog stompboxes, compared to digital stuff, which
> requires expensive components, super expensive development tools, and
> exotic development skills. and boutique pedals still often cost 4 or
> 5x their mass-produced counterparts. scale that up to the kind of
> labor, materials, and r+d that a massive hybrid modular requires, and
> factor in a very very limited production capacity... the math is not
> good. it's kind of amazing that the thing exists at all.
>
> anyway, the topic of cost has been discussed a great deal on this
> forum and elsewhere. (check out the front page of the yahoo group...)
>
> so, moving on: the only real reason don doesn't distribute the manual
> digitally is that it undergoes revisions as the firmware changes and
> he doesn't want obsolete information floating around. 200e, eternally
> beta... i have been wanting to make a public guide to the system,
> since this comes up about once a month, but i'm pretty damn busy as it
> is... i would love for someone else to do it, and a couple people have
> offered but nothing has materialized as yet.
>
> sometime, though... i promise...
>
> let's see: we don't have block diagrams. they would look really weird
> and probably not that useful, definitely more trouble than they'd be
> worth for me and don and anyone else who's actually working on the
> system. the only useful level of abstraction is the schematic and
> firmware spec, which we don't distribute for obvious reasons.
>
> (the block diagram for the 261e would be like: [knob input] ->
> [compute frequency]->[sine wave]->[some super weird shit happens with
> vactrols and opamps]->[audio out] ... hm, not so helpful...)
>
> as for modelling the 259e, the basic idea of crossfaded wavetable
> synthesis (or wave-terrain synthesis, in a fancier representation) is
> bread-and-butter in the digital world, should be pretty easy with yr
> Nord. the 259e weirdness comes largely from what's actually in the
> tables, though... google chebyshev polynomials... chop those up and
> use yr imagination... try stuff out... best i can offer.
>
> the 261e would be very hard to model digitally because the waveshaping
> is pure analog, and i would be hard pressed to provide even a
> quasi-useful analysis of the circuit. people have a damn hard time
> modelling a stupid vacuum tube... i'd guess that the best approach to
> modelling a 261e waveshaper or a vactrol gate would be perceptual
> ("add some harmonics to a sine wave..."); a deterministic physical
> model would get hairy real quick.
>
> same deal with, e.g. amp modelling: a block diagram of an amp provides
> only deep and well-coded clues to the construction of an analogous
> nonlinear transfer function (the computationally viable digital
> equivalent).
>
> hope that helps...
>
> /eb
>
>
> On Wed, Aug 6, 2008 at 3:29 PM, intellijel <danjel@intellijel.com> wrote:
>> I am a fan of the Buchla 200e but like many people I can only dream of
>> being able to afford a system.
>>
>> In the meantime I want to learn more about how they work and to better
>> understand why they cost so much (and why they are worth it).
>>
>> One thing I would really like to do is try modeling some of the core
>> modules in a Nord Modular 1. I know that someone did this before for
>> the Complex Waveform Generator module but unfortunately he lost the patch.
>>
>> I have also been trying to find block diagrams of how some of the
>> Buchla modules work but the information seems to be scarce. I was also
>> told that users who receive the User Manual are told not to distribute
>> it. This seems odd to me. I would imagine that any prospective buyers
>> of a Buchla system would definitely want to read the manual ahead of
>> time to gain a far better understanding of the systems capabilities
>> and to help select modules to purchase.
>>
>> Is there anywhere that I could get some of these details?
>>
>> thanks!
>>
>> 
>

ezra buchla wrote:
> sorry, i realize this whole thing was pretty snarky. not my intention.
> 


On the contrary, it seemed pretty right on to me. Additionally, it 
illustrates perfectly how the minds of the Buchlas work:

"(the block diagram for the 261e would be like: [knob input] ->
[compute frequency]->[sine wave]->[some super weird shit happens with
vactrols and opamps]->[audio out] ... hm, not so helpful...)"

This is great. It says that y'all are capable of making extremely 
complex circuitry, but aren't too full of yourselves to call it "super 
weird shit."

And this line:
"people have a damn hard time
modelling a stupid vacuum tube... "

spells out exactly why we still need to have so much analog realm 
hardware for making musics, even if it's partly digital.

-- 
./luther