Archived posting to the Leica Users Group, 2006/01/16
[Author Prev] [Author Next] [Thread Prev] [Thread Next] [Author Index] [Topic Index] [Home] [Search]Scott, The cheap stuff and good stuff argument in the past was all about design and engineering. The difference between a good speaker and a fabulous speaker was about using better materials, possibly a custom magnet, and paying more attention to the harmonics of the enclosure. The difference in the materials was small relative to the difference in price so that a smallish outfit could sell fewer units but still make good money as the margins were better. But when you come to semi-conductors you are looking at some very large fixed costs to build a fab plant. Start at nine figures. Yes, custom fabs are out there, but they get very large sums for small volumes as the number of rejects is quite high. To get really low costs you do the same as the memory chip folks, a dedicated plant with an engineering team that spends their whole life driving the sigma up, the return is the ability to stay in business as your product drops in price. So, for a specialty camera based on at least an APS-C sensor we will probably have to wait for some standardization probably in the generation of chips being designed now. Sort of the same analogy that Intel still makes 8088 chips in the same fabs but now gets a $1 or so a chip instead of $300 or so. But if an 8088 chip will do what you want, it can be used very effectively. Don don.dory@gmail.com On 1/16/06, Scott McLoughlin <scott@adrenaline.com> wrote: > > Comments below. > > Frank Dernie wrote: > > > If I have understood the issue properly - a gross assumption - the > > problem is that large imaging chips are a side-branch technology. > > From the outset R&D on chips has centred on speed and miniturisation > > with great success. The CCDs in P&S cameras, smaller than the > > smallest fingernail, are in step with this. APS and larger sensors > > are almost a new branch of the technology with complete new R&D > > requirements which not everybody either has or chooses to devote the > > resource to developing. On top of that the results available from > > current P&S cameras are completely satisfactory for 99% of their > > users. Development is likely to follow smaller with similar quality > > not bigger and better, simply because of market size. My son already > > prefers the pitiful camera in his phone to carrying an additional > > "box", for him and his friends the quality is good enough to outweigh > > the inconvenience. > > Sure, but there's always been cheap stuff alongside "not so cheap" > stuff. My dad still has (doesn't use) an old nasty Kodak Instamatic in > his desk drawer. I don't recall the existence of cheap mass market > cameras being a market hindrance to the universe of nicer cameras, so I > still don't feel like I have a grip on the "macro economics" of the > evolving camera market. > > > A bit like HiFi where size, capacity and battery life are more > > important in the portables market than sound quality, such for the > > minute proportion of the market where quality is more important there > > will soon be no products at all. > > Ugh, don't give me a heart attack! :-) I'm not a rich guy or an > audiophile, but I've got a modest Arcam amp/Paradigm speakers (from the > late 80's!) setup that sounds good for my smallish room where I listen > to music. Bought a cheap CD player, learned my lesson and have been > shopping around for replacement. > > My amateurish gear testing since I was in college (bought those speakers > back then down the road across from MIT up in Cambridge, MA) included > (still does) putting Joni Mitchell's "Court and Spark" on and listening > for the sound of her breath - inhales and exhales - and the sound of > Jose Feliciano's classical guitar (I'm and amateur guitarist) on the > title track. Makes all the difference between music and a "tone > generator." Unlike my own midrangey preferences, my ex wife used to be a > treble nut and used to always listen to Steve Winwood's "Back in the > Highlife" - she loved chimey highs. Kind of like good pot vs. good > coke. We never did merge our stereo systems, and probably should have > learned our lessons back then and never gotten married :-) > > > On top of this I understand that every silicon wafer on which the > > chips are made has a certain number of defects per area. Clearly this > > makes a situation where there is a chip area where 100% scrap is > > statistically inevitable and means also that reject rates will > > exponentially increase with size. I don't know how much effort has > > been put into reducing the number of defects per area recently, if > > the cost of super pure wafers is extremely high and the size of chips > > has continued to miniaturise it is entirely possible that there has > > been little recent research done on this area. > > > Excellent point. Circuit density has been the name of the game in > computer chips. A few years back, I was wondering if research into > decreased defects and increased yield-per-wafer would be increasing to > support making bigger sensors more cheaply. Skipping post-processing (no > new real detail there), I wonder how good the pixel density/noise is > going to get on small sensors. > > > It could well be that the projected market for large chips is so > > small that only small R&D budgets will ever be devoted to it and the > > chips themselves will always be special small production run items > > which are relatively very expensive. > > Frank > > > It's really pretty sad. You know, the heck with religion. We need a > "sensory jihad." Show the faithful a few 8x10 platinum prints, give > them some real Italian pizza and gelato, feel the cool caress of a well > worn Indian Madras shirt or the heft and density of a nice Austrian > loden cloth coat, let them listen to well recorded music (Chesky or > something) on a good stereo system. Let them have a peek at sensory > heaven, as it were. That'll make folks hopping mad about what they've > been missing and ready for a good fight :-) > > Scott > > > On 16 Jan, 2006, at 04:28, Scott McLoughlin wrote: > > > >> Makes sense to me. What doesn't make sense is why more companies > >> aren't then manufacturing more sensors. Business abhors a vacuum, as > >> it were. Certainly in other areas of semiconductor design and > >> fabrication > >> (and associated supporting chipsets, firmware and the like) there is > no > >> shortage of companies - and plenty of venture capital to start new > >> ones. > >> > >> What is the $$ volume of the camera industry (consumer, commercial, > >> industrial)? If it's relatively small, that might explain the > >> differences with > >> the rest of the (huge) chip industry. > >> > >> Scott > > > > > > > > _______________________________________________ > > Leica Users Group. > > See http://leica-users.org/mailman/listinfo/lug for more information > > > > _______________________________________________ > Leica Users Group. > See http://leica-users.org/mailman/listinfo/lug for more information >