[sdiy] SSM2040

[ASSI]

On Freitag 02 November 2007, Barry Klein wrote:
> take a read of this discussion:
> http://www.edn.com/blog/1700000170/post/1850011785.html
>
> Read the last two paragraphs.  How is it going to be inexpensive to
> make your own IC?  How would any of us, even if we had 10K$ even
> proceed? I suppose there is some university teaching this stuff but
> it is not common knowledge otherwise.  Are they hinting it will be?

On IC design and having prototypes made.  If you know the right people 
you can even get a few mm² on a testchip mask set for free, as there's 
always some slack that can't be filled otherwise.  Dicing and packaging 
isn't all that expensive either.  Most universities have taken part in 
prototyping programs (MOSIS, Europractice, etc.) and have produced 
impressive results for comparatively small money - a handful of 
packaged chips for around $3k was one example I witnessed while still 
at the university.  What Don is apparently hinting at is that for 
smallish analog circuits you don't really need all the expensive CAD 
tools that are normally used to design much more complex chips - you 
can go from idea to GDSII data just using free software with a little 
bit of extra effort compared to the industry-grade tools.

Things get much steeper when you want to go from prototyping to 
production - the NRE cost kills you unless you have something that 
needs to be produced in millions of pieces and can be sold quickly.  
The only viable (and proven) route to small-scale production I've seen 
so far is http://www.arraydesign.com (brainchild of the venerable Hans 
Camenzind of '555 fame).  One problem of the 700 series is the 
relatively bad pnp, but hey the 3080 didn't have a good pnp either.  
The CEM chips were based on an earlier incarnation of that same idea 
(wafers probably run by GEC Plessey) - if somebody could send me a 
non-working (but not burnt out) CEM chip, I'd open it to make a die 
photo...

Anyway, running a single lot would probably cost around $10k and yields 
between 15k...250k untested dies.  Factor in capital, yield, test and 
packaging cost and some margin and an IC done that way will cost around 
$10-$15 to the end user (perhaps not surprisingly, that's the price 
that Doug Curtis used to call).  Not at all bad, but of course not 
affordable to the average single person - but quite certainly within 
reach for a small company. 

Just found this, BTW: http://curtiselectromusic.com/


Achim.
-- 
+<[Q+ Matrix-12 WAVE#46+305 Neuron microQkb Andromeda XTk]>+

SD adaptation for Waldorf rackAttack V1.04R1:
http://Synth.Stromeko.net/Downloads.html#WaldorfSDada