From: PENQUAKR74@aol.com
Date: Wed Sep 17 2003 - 10:43:56 EEST
Bathsheba:
If you consider the advancements in the RP technology and the progression of
how the products have been delivered to the public, then you will see that in
1987 the first machines were large and required, and still do, very expensive
lasers. Because of the cost of these machines, just as mainframes were costly
and required sharing (anyone recall GE's timesharing business?), service
bureaus were the best means for small and medium sized businesses and for large
businesses with the occasional need for models, to have access to these
expensive machines. RP machines cost approached the $100,000 mark and less, and
Z-corp's 50-60K machine forced many to respond to the market which is heavily
driven by form and fit. But as Z-Corps processes became more useful through
postprocessing developments, such as metallization and other infiltration techniques
created by its users enabling molding and casting, it began to compete for
more of the business normally reserved for the larger SLAs. Stratasys has beco
me a force because it has developed lower and lower cost machines, selling into
markets where service bureaus were the main option for designers and
manufacturing engineers. The materials developed have become well-known and their
shrinkage properties tabulated to enable dimensional and quality control beyond
what was possible a mere 5 years ago. From the equivalent of a mainframe to a
desktop RP machine capable being shared over a network has occurred in less 15
years (1987-2002). The drop in 3D market value is not a surprise given its
entrenchment for so long in the large machine market. Companies like Z-Corp
and Stratasys now dominate the desktop market, with maybe Objet and potentially
Envision Technologies which also uses DMD technology in its machines, getting
ready to make a big splash. So, while you may have a certain insight or
perspective as a user, the trends are easily seen that paraphrasing Andy Grove's
observation, Grove's Law, as it's popularly known: RP Technology will double
it's capability to produce products less expensively every 18 months. Along with
that is the fact that materials technology, the "ink" if you will, will also
advance to match the machines technical progression. Step back, create a
timeline and you will see this clearly. The only thing holding these advancements
back has been the endless lawsuits and bickering in the courts over
infringing patent rights. 3D's solution was to acquire its competitors which stretched
it too thin and got it involved in too many battles, decreasing the
investment it should have been making in smaller footprint, user friendly machines. As
for a cleaner part and a smoother part or introduction of color, those are
simple problems to solve based on variable layering, faster curing, and
in-machine post processing. Remember the first xerographic processes were extremely
dirty and it took almost 15 years for Xerox's predecessor, Haloid Corp, to
clean it up in the 50's.
Finally, in rebuttal to the durability and aesthetic look of a replacement
part for a consumer product, the parts that were made by SRI's process were
silicon nitride ceramic that were tested by Allied Signal at high temperatures,
approximately 1000 C. That is durable. And it could be done in plastic also.
If one wanted to have all the knobs match on a stove range, make a new set.
But, in my opinion, just having a replacement knob that slightly different,
instead of pliers, is far better than having none at all. It won't be too long
before you will find a machine for the consumer market. The first ones will be
at Kinko's.
Scott Taper
Technology Commercialization Consulting
TCC has joined with Andreé Driskell Associates
(www.andreedriskellassociates.com) to add proposal and business plan preparation services for responses to
commercial and Government RFPs and other funding opportunities. TCC accepts
selected innovations for technical and market assessment and licensing.
In a message dated 9/16/2003 9:50:34 PM, sheba@bathsheba.com writes:
<< On Tue, 16 Sep 2003 PENQUAKR74@aol.com wrote:
> But this could have been a ceramic knic knac, jewelry model, a toy, an iron
> temperature dial, some other personal item or discontinued replacement part.
>
> RP machines will follow the same market progression that personal computers
> and computer printers have followed with apparently greater speed of new
> product development. From 1955 to 1980, computing systems moved from
mainframe to
> mini to micro, a period of 25 years to go from millions of dollars to
> thousands. It took a mere additional 10 years to get to under 1,000
dollars. If the
> analogy continues to hold up, we can expect a 1,000 dollar machine in 2005,
> about half the time it took to get to the same point for personal computers
> because the prior advancements in computing power translates to the RP
market.
>
> I welcome any comments on the foregoing.
I guess I'm pessimistic about that analogy. Going from a $20k
computer to a $1k computer was mostly a matter of incremental
improvements to existing technology, and making the marketing decision
to put it in a cute plastic box. But for RP the technology doesn't
exist: no machine is near, in ease of operation or usefulness of
output, to the level that would open a consumer market. Cute plastic
boxes abound, but most people aren't fooled.
I tend to think the biggest hurdle is the material science. Your
stove knob is an excellent example: it's not useful unless it's as
tough as my old injection-molded knob (preferably tougher, since the
old one broke!), a good color match to my other knobs, including fine
detail for the calibrations, and without visible layering. It also
must not require any postprocessing.
No process I'm aware of comes near these requirements. Only Sanders
has fine enough layers, almost. Only SLA has enough material
strength, maybe. Only ZCorp has color, if you like pastels. Fill in
your favorite here. All are wildly deficient in those areas where
they don't excel, and all require cumbersome postprocessing.
So my bet is that big breakthroughs are needed before that consumer
machine is foreseeable. Of course it's possible that they've already
been made, under a bushel somewhere...I feel like it would be an awful
lot of progress to be hiding.
OTOH, I wouldn't be surprised if a sub-$5000 concept modeler, based on
a combination of existing technologies, did appear in the next several
years. Probably looking more like SLA than anything else. There's
nothing about what these machines do that is inherently expensive,
they just haven't been exposed to economies of scale. HP, or a
handful of other companies, could do it.
But I don't expect a household appliance that would appeal to Joe
Sixpack, even though he has a digital camera these days. My crystal
ball is showing a tool for well-heeled 3D shops, with a price point
and market similar to the Microscribe arm.
It would still be a giant step forward -- if it can build the model at
the top of this page, http://www.bathsheba.com/misc/preview.html, I'm
standing in line to write that check. Knock wood....
-Sheba
-- Bathsheba Grossman phone (831)429-8224, fax (831)460-1242 Sculpting geometry bathsheba.com Solidscape prototyping protoshape.com Protein crystals crystalprotein.com
This archive was generated by hypermail 2.1.7 : Sat Jan 17 2004 - 15:18:06 EET