DNA synthesizer

Took a look at some DNA synthesizer models today. I was particularly intrigued with the Dynatech ML1000 DNA synthesizer , since it's actually available on the ebay for measly $799 dollars, the cost of a crappy new laptop or a decent new desktop. The way DNA synthesizers work is simple (in theory, everything is simple). Hook it up to a computer terminal with requisite platform, enter the DNA sequence you want synthesized, pour in sugar in one end and the DNA comes out the other end. The problem is, the technology isn't really all that mature yet. Most machines today seem to get around 90% accurate synthesis with simple DNA fragments, say about 300 base pairs. But once we get to the more complex level of synthesis, the kind of sequence we'd actually need to do something that matters the accuracy drops significantly.  Human genome is about 3 billion base pairs long (with 23 chromosomes, but we all knew that), so synthetic human-like lifeform is unrealistic given the present state of biotechnology where we might or might not manage to stumble over 1000 base pairs.

Even so, this is quite interesting. With the Moore law seemingly applicable to the biotechnology sector as well as the traditional computing sector, seeing a DNA synthesizer capable of synthesizing DNAs of around a billion base pairs wouldn't be out of question during our life time. If the registry of standard parts , or at least something close to it can come of age soon, we might even be ale to expect certain synergistic relationship between the two, one field accelerating the other through financial support and academic endeavor.