On Mon, Aug 23, 1999 at 11:23:09AM -0000, Mark Walsh wrote:
>
> I don't see this as stealing at all.  No one is forced to buy these
> adaptors.  The time to create my own adaptor for programming chips would be
> much more expensive than the cost of the Emulation Tech. parts.  The price
> isn't a function of their costs so much as a function of the value to me.
> While they are pricey, I don't see them as being unreasonably so.
>
> The great thing about our system is that you can create your own adaptor,
> undercut their prices, dominate the market, and make lots of money.
>
> Or maybe not.
>
> Mark Walsh

This is exactly right. One thing I do notice about places
like Emulation Tech is the size of their catalogs. Even
just a simple listing of all the part numbers runs for pages.
How many IPC-standard package outlines are there? How many
can reasonably be adapted into a DIP or PGA format?
So if any of us were to go into the business of manufacturing
adapters, which packages would we do? Probably SOIC for
sure, because that's where the biggest loss is to the
low-volume developer. SOIC is also reasonably easy because
you can have just a few sizes that will fit a wide range
of parts, just by underutilizing the parts. But what about
SSOP, TSSOP, QFP, TQFP, BGA, etc? How many designs would need
to be done, how many parts would have to be inventoried?
If you just pick off the easy designs, would that be enough
to generate a steady demand, or would most people go to the
Emulation Techs of the world because there they don't have
to wonder if the part they need will be available?
What are the costs of creating, documenting, validating and
maintaining the specs? How much would it cost to offer a
waranty? What are the costs associated with holding inventory?
There is of course much more to a business like this than
simply having one batch of boards made.

One reason that we can individually have such parts made
for ourselves at such a low cost is that so many of the
*real* costs are hidden or nominally allocated to other
tasks. If you work in a small contract shop, you might
design an adapter as part of your effort on a single
contract. Would you bill this to overhead or would you
charge the design time it back directly to the contract?
If you need ten of an adapter and get 40 made because
that's where the price break is, do you think of your
per-adapter cost as one tenth of the bill or as one
fortieth of the bill? If you are a hobbiest, then the
calcuation of costs can be very messed up, because one's
own time would normally carry no direct value in the
calculation, at least in the sense that there wouldn't be
any identifiable, out-of-pocket monetary cost associated
with the hours of labor you spend on your hobby.

This is why, personally, I think that it makes more sense
to collect together designs and techniques for making
adapters, and have them available as a sort of public
domain knowledge base. If people would want to paste
one into the Gerbers for some board they are having
made, then great. If somebody does this and then tells
the list that they've got maybe a couple dozen of these
available in case anyone needs one, and that they'd like
maybe $3 to cover their associated costs, then even better.
If someone can think of some low-overhead way to act as
a clearing-house for the boards that individuals make this
way, then fantastic, especially if it can be kept mostly
on a barter basis.

But personally, I think that setting up shop as a
cut-rate adapter vendor is likely to be a more difficult
endevor than it might initially appear.

--Bob

--
============================================================
Bob Drzyzgula                             It's not a problem
bob@drzyzgula.org                until something bad happens
============================================================