I still can't define what that is though? Maybe it's too high contrast? Naw.. that doesn't sound right? Maybe it's the glass surface that bothers me. Maybe that only projection (either front OR rear) looks 'filmic' to me?? I don't know what it is.
Here I can wholeheartedly agree with you. I can't quite figure out exactly what it is either, but I suspect it to be a combination of factors. Things I kinda suspect are the (as you said) characteristics of being projected, whatever they are, plus I think in the case of DLP the color gamut is generally closer to film.
Neither CRT or DLP projection looks really like film, but to me they look quite different from each other and so their 'non-filmicness' seems to be from at least partly different sources. In the case of DLP the scaler artifacts, motion artifacts (scaler too maybe), lack of grain texture and 'discreteness' of the pixels among other things are the ones I can notice. For CRT projectors color gamut still looks like a tube glowing and some of above factors also apply. One nice thing about CRT is the lack of obvious pixelation.
Millions of moving mirrors will always cost a LOT to make.
Always is such a strong word. I guess it depends on how you define cost a lot. In fact the silicon surface micromachining process TI uses has great potential to become dirt cheap with adequate refinement. At one time not so long ago the cost per transistor on an IC was pretty high too. MEMS is young and is very promising in the general case for all sorts of sensors and actuators. Everything from navigational systems to network routers to print heads have the potential to benefit from this technology, and with this level of interest advances in capability and drops in production cost should be rapid. Of course drops in
consumer cost are not directly driven by production costs.
(Charlie can keep 2nd guessing this, but he needs to actually go SEE one).
I see one (LCD projector) all the time. I never took it apart though.
2D vs. 1D arrays, I appreciate the economies that could be enjoyed with a 1D system, but I tend to think, given the fact that MEMS is virtually certain to improve in quality and drop in price by orders of magnitude as time goes by, a 2D solution probably has better legs. I'm not sure how broad or how defensable the TI DMD patents are, but I hope they don't completely preclude competitors from creating MEMS based 2D solutions. Patents are slippery things (my name is on 10-14 so I have some first hand experience here) that can be very narrowly or broadly interpreted depending on many factors.
Previous Topic
Index
