In-body stabilization with a wide range of movement that allows movements (tilt / swing / shift) for architectural use, rapid focal plane forward/back movement for focus stacking, astro-tracking, video frame-rates for multi-parameter stacking (e.g. high and low exposure for each of three focus shots for post-processing depth of field control and HDR , 6 shots in 1/10th sec.).
Advanced photon-counting sensor with flip-chip, thinned BSI, TSV, bonded to analog amp and digital preprocessor chip with an amp, memory buffer and preprocessor for each few thousand pixels, giving extremely fast readouts (and likely the ability to do other tricks.) Free-space (~1 -2 cm) optical sensor output and inductive power to sensor for internally wire-free sensor operation.
Mirrorless or semi -mirrorless design, (mostly transparent but retractable mirror design). Totally open, all-function remote control system. GPS / intertial tracker/ heading / elevation / auto time. Scr. tilt, HP jack, wifi, mono ver.
Valiant Thor: A nice value for $300 however I would have liked Canon to upgrade the mounting plate to a more substantial plate of pre-famulated amulite surmounted by a malleable logarithmic casing in such a way that the gold spurving contacts ran in a direct line with the panametric fan. Also, the hydrocoptic marzul vanes should have been fitted to the ambaphascient wain shaft so that side fumbling is effectively prevented. The main focus ring should have been of the normal lotus-odeltoid type placed in panendurmic semi-bulloid slots of the stator with the threads being connected by a non-reversible tremmy pipe to the differential girdle spring on the up-end of the grammeters. Oh well, maybe the next version . . .
There is much to what you say, but I'd like to add that a chalcogenide-based frammis knurl volution on the forward rotical assembly would also have made it easily 20% more cromulent.
So at a bit less than double the price of the Pentax 645z, you get the same size sensor, no camera, half the frames per second, and about 8 stops lower maximum ISO. Oh, and less than half the pixels on the screen.
Or for the same price as the Hasselblad, you can get a Pentax with better engineering, better metering, better autofocus, better low-light performance, beter proceesor, better ergonomics, everything integrated SLR-style, and two or three new lenses designed for digital.
Wait, let me think about this....
Gary Yelland: the lens is not that fast, no gps, no wifi zoom range is ok but not stella, so to sell this one the images would have to be amazing and superfast startup and shoot time.
Gripe, whine, moan. WTF is with the commenters here?
This is an incredible set of features for the money. The zoom range is perfect, it's at least as fast a lens as the competition even in some much more expensive cameras, it doesn't skimp on the wide end, focuses close, and it has an optically stabilized telephoto long enough for any common use. AA batteries are great, available everywhere and cheap enough to have spare sets. It's light and pocketable. The back-illuminated sensor will be sensitive for its size, and downsized to 1-4Mp it should have good image quality even in marginal light. Bigger sensors are always nice, but for $250 with a decent lens this is amazing. It has all the most important features in a total package of usability that has no real competition at this price. I hope the shutter lag is low and the focus is reasonably fast, but even if the camera is mediocre aside from the reported features, it's still going to sell very well.
wkay: the majority of these pictures basically stink, why do I need this camera? maybe half the problem is the poster playing around with camera raw with no idea what he's doing
ProfHankD: This is actually a huge deal. Circuitry under a micro/nano-fab device is not a standard technology and this is "to increase the production capacity" -- I wasn't aware anybody was doing this in large-scale commercial application. Very cool, Sony!
Placing circuitry under sensors allows for significantly smarter circuitry per sensing element, and I've been working on smarter control to go under sensors for about a decade. Basically, the idea is to build a parallel supercomputer under a chip covered in sensing elements. Here's a 1-page summary:
The reason this is more important for phone sensors is that the circuitry is a larger fraction of the sensing element size with tiny pixels. You also need higher amplification which makes noise more of an issue and having good processing under pixels means less routing of touchy analog signals. Oh yeah... with smaller die sizes, they also can better tolerate yield problems during development.
Here's Sony's own technical backgrounder on the subject:"Sony's Stacked CMOS Image Sensor Solves All Existing Problems in One Stroke"http://www.sony.net/Products/SC-HP/cx_news/vol68/pdf/sideview_vol68.pdfKey points - with fragile thinned back-illuminated sensors already needing a bonded silicon support substrate for mechanical reasons, it's a good idea to use that silicon area for something useful. *65nm logic coming out this month, 45nm in the future*Benefits: higher frame-rates, higher sensitivity, lower noise, more pixels, higher dynamic range, built-in DSP functions, lower costs, easier semiconductor processing customized for both sensor and logic without the usual compromises, quicker design-to-production cycles*first bonded-logic chip coming out this month will be a 1/3" 13Mpixel sensor with RGBW pixels and "HDR movie" mode which improves color in brightly-lit scenes.
Jogger: the resolving power will be limited by the lens methinks, even in the samples provided, the 100% are useless except for surveillance
gasdive - no, the resolution is limited by the sizes of the individual lenses, not the size of the whole board, so your resolution would be more on the order of arcminutes than arcseconds. They are using 1 big lens rather than the conventional camera-array setup for just this reason. (Over long distances the fluctuating distortions in the atmosphere are the limiting factor, anyway. With sufficient framerates and processing this can actually be turned into an advantage.) The real problem with high resolution cameras is the vast amounts of data that have to be stored and processed. 1 gigapixel x 8 bits x 30frames/s (which is just an 11x11 array of quad-HD (QFHD) cellphone cameras) = about 30 gigabytes per second raw data, 3.7TB/hr - which can be reduced by a factor of 10 or more for storage, but each byte needs hundreds of operations done on it, so that's on the order of teraflops. Even very power-efficient processors still will need hundreds of watts.
K_Photo_Teach: What I see here is new technology that will improve low light performance regardless of whether it is put into a camera or a DSLR.
But it shows that the camera phone market is driving the innovation. Between this and the Nokia 808 who knows what the future will bring? Imagine the Nokia 808 with great low light performance?? It already rivals Full Frame DSLRs in resolution. Take a look here:
limlh - This could benefit large sensors eventually. This gives the possibility of having many more amplifiers and even ADCs and low-level digital processors per pixel. With current technology the number of amplifiers scales with the length of the longer edge of the sensor. With a bonded chip, it can scale with the area of the sensor and also use much shorter and more uniform length connections. This could lead to lower readout noise with faster framerates.
Rubenski: DPR wrote: 'Good photo quality, if you don't look too closely'.
Come on guys, what kind of standard is this?!
And so long as you never need anything over ISO 800. After that it's just crummy.
Ultan: Compare the studio test shots with the Canon G1 X, for instance on the globe. Judging by which size type is visible, the G1 X does as well at ISO 3200 as the Sony Cyber-shot DSC-HX200V does at ISO 200! At ISO 3200 on the Sony, even the "INDIAN OCEAN" label on the globe is illegible - the second "I" has been smeared away, the "OC" is nearly gone!
It's irritating that the ratings on the conclusion page can't be compared between the two - even though they are competing cameras - somehow the G1X has been classified as a "Mid Range Interchangeable Lens Camera / DSLR", even though it has a fixed lens, is smaller in every dimension and weighs less than the Sony, and there is an "enthusiast large sensor compact" category. (Another error in the G1 X review is the date - it says Jan 2011; it should be 2012). In fact, none of the cameras with a decent sensor size can be compared with the phone-cam-size sensor cameras like this Sony, even when the big sensor cameras may be smaller and lighter.
(Continued)The G1 X has over 9.3x the sensor area of the Sony, pixels 11.8x bigger, but costs only 1.67 times as much as the Sony. Lenses are just about as fast - both f/2.8 wide, f/5.6 vs. f/5.8 tele. Both have image stabilization. The wide angle equivalent is about the same 27mm vs. 28mm. The Sony goes to 810mm, which, while impressive, is useless without either a tripod or full sunlight. The Canon only goes to 112mm, which is good enough for most purposes. The Sony is over 5x faster in taking bursts of pictures, but its JPEGs are filled with artifacts and it has no RAW capability. Unless you never take pictures indoors or at night and do daytime sports photography on a tight budget, the Sony is the wrong camera.
Compare the studio test shots with the Canon G1 X, for instance on the globe. Judging by which size type is visible, the G1 X does as well at ISO 3200 as the Sony Cyber-shot DSC-HX200V does at ISO 200! At ISO 3200 on the Sony, even the "INDIAN OCEAN" label on the globe is illegible - the second "I" has been smeared away, the "OC" is nearly gone!