Ok, and here then is my fatal question on the TVS-D in relation to discussion on art and technology as well as photography plain, re: Marc, Lynn, Ken Lo, Rico, et al. Of course Shu-Hsien, and possibly his connection to Finney, if this situation is still as was. Ken Lo's notes and pictures with Marc's give the references which Yeo Yin Khoon brought me to this forum in the first place to find, and he himself seems very interested in notes, mentioning also the useful comparison on
http://imageevent.com/otto/5mpcompact.
I've tried to understand some distinctions between the Canon S50 and the TVS-D - of course considering whether either can be a good choice within a personal bound of not carrying a D-or-otherwise SLR or prosumer (G3) size, while appreciating their sure benefits, and wishing for a particular level of ability for working into a variety of available light.
At first glance, the two pocket cameras have many similarities as we know. Hard to agree with design where they don't, in fact, such as leaving out full manual control and RAW on the Contax, but this could be acceptable with some flavours of an opened mind about 'photographic expression balanced into contemporary facets of life', and with some ideas about peculiarities and limitations of digital anyway so far.
My backstop on all this, not to be mentioned again, could be an inherited Leica M2-Epson 3200 scan combination.
So, what does one see with a Canon S50? At first, apparent clarity of resolution to limit of sensor, even reasonably clean corners, at first seeming quite nicely done. Successful attenuation of most issues with CCD-lens interaction blooming (blue-violet fringe on contrast). Nice features, full manual settings, RAW. Well-controlled noise, agreeable color.
What about TVS-D, then, by comparison? Quite evident brilliance and tonal range on many types of pictures, supported into difficult conditions by very impressive lack of flare, surely that T* lens. Apparently much greater preservation of detail into shadow gradations. Maybe a color issue or two, probably manageable. Consistently visible noise, of a kind which in fact does not feel to detract so much from pictures. Less control, but perhaps adequate to very many purposes and actual constraints, perhaps fit in fact to some view of digital-photography practice.
And, with the TVS-D also, full 5MP images which appear as consistently to show a difficult to describe but definitely lower and grittier resolution. Which prominent fault seems to disappear into almost velvety smoothness and dynamic range, when the image is res&led into 'next lower size; - 3 MP or so.
Beside those pictures from the TVS-D, the at-first seemingly detailed pictures of the Canon then appear quite 'flat', generally 'milky' in all cases, even when there is no flare likely to suspect. At full 5 MP, the strict resolution superiority does not seem to deliver the sense of presence in image details that the unusual look ot the TVS-D seems to. The Canon shadow detail appears definitely less, and doesn't come up by increasing post-camera contrast, even after apparently standard raw processing (that could be adjusted if the tools allow it). At downs&le size or normal viewing distance, the TVS-D seems to have more perception of detail even though it does not strictly resolve. These are my impressions.
There are very attractive pictures posted here, and they are all down-s&led - perhaps because in the art Marc mentions of his own photography and I am sure others see for themselves on own, the reduced size just does look good, so that is posted. Contrariwise, all reviews so far criticize the 5 MP TVS-D pictures as being just 'average' in resolution - I think walking around this qualitative difference that is not so simple to describe.
On the
http://imageevent.com/otto/5mpcompact site, the full-size images are present, if one drills down far enough - I was fooled the first time. And that's where I learned first this distinction - because the TVS-D in relative wide-angle shots with detail (the full boat canal, the full Danish platz with statue, for ex&les) looked considerably better than the S50, but the roles were a bit reversed every time one drilled to the full-resolution basis.
Again, this is simply what I can find to see. The gritty full resolution is less well assured by Steves-Digicam and DCresource tests, due to their flippant attitude and apparent lack of care, but it is there, and I believe it by now. I asked Jeff of DCresource if he'd missed what Marc discovered of the default lower-res Jpeg default on the TVS-D, but he said he wasn't interested in looking into it, and I actually don't think that's the problem now.
So what can we ask? First, I have some feeling it would be very interesting to discover through connections such as Finney's, whether there is as seems likely from reading between lines so far that Kyocera-Contax decided to have the TVS-D work just in the way I seem to find, given it is accurate without having both cameras to try.
If they did, given the ex&le of Canon etc. for the chip, I think they did it with purpose. I would not be surprised to find that this purpose is the quality of a 'final print' photo. Finney might like to argue that with a clean &lifier chain, not suppressing the CCD's own noise actually can help preserve shadow detail between the bits of A/D resolution - a form of interpolation, and known technique.
As some substantiation for this detective chase, another camera I've looked a bit into is the Leica D-Lux (latest TVS-size Bauhaus-case release). It is different, but does apparently also possess considerable image noise, again not unpleasant, and with seeming its own non-Canon take on improved picture tonal range. There is at least one comment that they designed for print quality, not substantiated.
If these speculations are true, then there really is the result of a distinct take on image quality from Contax. I'm pretty interested that it should be so, as the impression is that this goes some real distance towards relieving the constant dynamic range and 'tonal snap' holdbacks in less than EOS-1d or Sigma digital photography. I would think that if this were substantiated, others in the field would be interested. It's very healthy for we who take pictures, and it's very healthy also for the industry to have individually valid distinctions between cameras.
Of course, it could be as the reviews like to suggest, that it's just a matter of results in an electronics company trying to do photography, as others besides Kyocera do. Maybe I am really missing something in this speculation, and it's just true that the TVS has some very nice qualities, but should be treated in them as a lower-pixel camera. That would be good to know, and decide on, too.
If we really understood this, then it would be clearer whether the right kind of raw processing could bring the S50 into the apparent TVS regime of contrast, snap, and tonal range, or whether the T* lens is a completely key factor.
I'm going to hold onto my developing imaginations, that camera companies ought to be a lot more interested in opening up their in-camera firmware as well as raw formats to outside and definitely multiple developments. In a sense, those building specialized RAW converters for high-end DSLR's are already doing it. The result would be equivalent of the differences in film, with attendant ability to fit to styles and desires. The key would be that you could reconfigure your own camera - after buying one on lenses, form factors etc. which become the more important. I am sure others will have their own ideas on how much this might do for both expressive possibilities and equally for real-economic possibilities of markets.
But first I think we need to get a really good idea, how that set of straw hat and cup pictures so carefully arranged by Marc are really different - because they surely are.
Hope I haven't asked too much of the forum's intents, and the advantage is that we will not know less if there's not more to understand. Off before I lose nerve to send it
.
regards to all, Clive (Steward)
