stevens37y: "solid block of aluminum"They should use titanium. Looks and feels much better.
A camera made from same material as *bicycles*? That's so pedestrian!(badabum)
cgarrard: I wonder if Leica will consider make matching toilet paper for this camera?
Matching? As in, machined from solid aluminum? Yikes!
groucher: Should be a good caver's camera - a bit of grit, mud, water, bashed around in a tackle bag. Don't like the silly touch screen though - will it work with muddy hands?
From the manual:"If you intended to take a shot and discovered that your hands were muddy, ask the enclosed butler to wipe them for you with dry, clean cloth."
Daniel Lee Taylor: I don't get it. This is an EOS M with an optional EVF but without IS, the 11-22 UWA, or EF lens compatibility...at 12-20x the price depending on which component you're looking at.
Who in their right mind would drop thousands of dollars to buy this thing? For the money you could have a Sony A7, and I dare anyone to suggest that IQ would be better with the Leica T.
Leica has made some great cameras and lenses in the past, but lately it seems like the red badge means "sucker."
Hey, my $36 Casio tells better time than a Rolex, but for some occupations it comes with a territory:http://media.mensxp.com/media/photogallery/2013/May/1369377624_75825.jpg
The *real* men of status would choose one machined from solid enriched uranium. Gives that extra unique touch.
It would have to be carried in a lead container, of course, but that's not a problem - they have people for this sort of thing.
BorisK1: Funny how their sample image's corners are completely blurred.
If you're showing off a great new feature, wouldn't it be nice to, you know, actually show it off?
On a more serious note - why add a whole new stage in sensor manufacturing (read: very expensive), just to make the corners slightly sharper? *And*, like the article mentions, to produce sensors unsuitable for zooms and for interchangeable lens cameras?
Makes no sense to me.
Unless, like others mentioned earlier (http://www.dpreview.com/articles/2279255612/sony-s-curved-sensors-may-allow-for-simpler-lenses-and-better-images?comment=0577067993), there's a disconnect between marketing and engineering, and the real benefits have nothing to do with the curvature of the focal plane and everything to do with mechanically stressing the silicone (which enhances the sensor's performance - meaning you get less noise and more DR).
Likely, in final production the sensors won't actually be curved, just stressed.
Aside from what I care about (which very definitely isn't a $10,000+ camera with a fixed-everything normal lens), there's a matter of feasibility.
Commercially, a specialized sensor that could only work in cameras with non-interchangeable/non-zooming/non-WA/non-macro lenses, just wouldn't sell.
Now that I think about it, $10,000 estimate is probably on the low end. The lens-type limitation would reduce the number of sales, which would makes it even *more* expensive, which would reduce the number of sales even further, and so on.
On the other hand, a new sensor technology that would deliver an extra stop of light sensitivity to *all* cameras, would fly off the shelves. Just like BSI does.
"get the Sagittal and Tangential planes to coincide on the Petzval surface" - if I understand this correctly, a comparatively simple symmetrical lens system can correct for astigmatism, while still having a curved image plane (see p.42 here):http://www.iap.uni-jena.de/iapmedia/de/Lecture/Imaging+and+aberration+theory1396134000/IAT13_Imaging+and+aberration+Theory+Lecture+8+Astigmatism+and+field+curvature.pdf
Still, that's a highly specialized solution. Aside from ruling out zooms and interchangeable lenses, I suspect ultrawide and macro lenses would be tough, too. And probably not pancakes either (at least not ones that, like RX100(3), rely on exotic aspherical elements).
So we come back to marketing - such cameras would have a limited market, would require a separate stage in manufacturing process, and be very, very expensive.
On the other hand, what if mechanically stressing a sensor gives as much benefit as, say, Sony's own BSI? *That* would be headline news.
Sorry, this still doesn't sound convincing. Improved corner sharpness (at some apertures, at some distance settings) just does not sound important enough to warrant a separate *type* of sensor, only suitable for fixed-focal-distance, non-interchangeable lenses.
The benefits are not that huge - a slightly simpler and smaller lens (it would still need to correct for things like coma, astigmatism, geometric distortions, etc.)
The drawbacks are enormous - no zoom, no interchangeable lenses. A sensor that only works for a single type of camera. This means low sales volume, coupled with a separate manufacturing process. This, in turn, increases the cost significantly.
While RX1(R) are very nice, they are already quite expensive. An RX1 with a bent sensor would easily cost twice as much, probably more.
How many people buy a slightly smaller RX1 for $5,000 - $10,000?
Funny how their sample image's corners are completely blurred.
Macx: 1) When considering equivalence (i.e. DoF, diffraction and total light/shot noise) sensor plays second fiddle to the lens: Given the same FoV and the same physical (virtual) aperture, you'll get the same image characteristics, no matter if you're shooting an image on your phone or on your medium format digital back. The thing that is holding your phone back in performance is for most intents and purposes the tiny lens and not the tiny sensor.
2) The difference in DoF and shot noise between different formats only come to play at the extreme ends of the exposure gamut: Bluntly, on a bright day, shooting for a wide depth of field your phone will give you roughly the same performance as your dslr. Only when the phone is out of its comfort zone the difference becomes apparent. Cameras using 1", 4:3 or APS-format sensors are a few stops shy of the comfort zone of current FF cameras. If you're not actually using those stops there is little to distinguish it.
Macx: Color depth refers to the number of bits per color. In terms of sensor characteristics, it's basically DR of individual color channels. MF sensors have significantly narrower DR than FF sensors. Which is why the highest usable ISO settings are so much higher at FF.
Canon shooter: The thing is, *base ISO* is not equivalent for different sensor sizes. It's calibrated to get the same brightness for identical exposures.At the same exposure parameters (ISO, shutter duration, and f-number), a 4/3 sensor gets a quarter of photons that a FF sensor does. To get the same brightness, 4/3 camera has to pay with 2 extra stops of noise.
mgrum: The bottom line that everyone always misses when the whole equivalence thing comes up is this:
*** It all depends on what lenses are actually available ***
A 24mm f/1.4 full frame lens is equivalent to a 12mm f/0.7 lens for micro four thirds. But no-one makes one so there is no equivalent to the 24mm in the real world. Likewise Hasselblad's claims of medium format offering shallower depth of field due to the larger sensor is nonsense, as the required lenses simply don't exist.
"A 24mm f/1.4 full frame lens is equivalent to a 12mm f/0.7 lens for micro four thirds. But no-one makes one so there is no equivalent to the 24mm in the real world"- They make a focal reducer for micro four thirds that converts a 24mm f:1.4 into a 12mm f:0.7 with a m4/3 mount. Look up "Speed Booster", by Metabones.
The problem with medium format cameras is the tiny sales volume. This means there's not a lot of money in sensor development, so their sensors are always generations behind the mainstream, and usually developed for different applications than photography, and aren't really optimized for picture taking.As far as I know, the only real advantage of MF is in resolution, and even that is dwindling. They lag behind full frame, and even some APS/C sensors, in DR and noise.Also, MF sensors' dimensions are smaller the original sizes of MF film. The "cheapest" ones are something like 1.25 of an FF sensor. To get twice the area of FF, you need to be in $40000 territory.Finally, MF lenses are slow. On FF, f:1.4 is mainstream, f:1.2 is normal, f:1.0 is rare, but available. On MF, f:2.8 is rare, f:2.0 is nonexistent.I'd say, aside from the resolution, there's very little advantage in MF, aside from mystique, prestige, and tradition.
GeraldW: A nicely done article!!! Kudos!
Something that occurred to me is the fact that all four cameras had very closely, the same pixel count. Meaning that the surface area of each photo site was smaller as sensor size decreased. Since the light intensity is the same for all four lenses when at f/1.2, that means the larger sensors have many more photons striking each photo site - so the "total light" concept works pretty well. However, if the pixel count is not similar, then shouldn't we see differences based on the relative sizes of each photo site? To me, it argues for fewer and fewer MP as the sensor size decreases, in order to maintain a larger surface area on each photo site. That should improve the signal to noise ratio at each photo site, shouldn't it?
A "signal to noise ratio at each photo site" only matters if you are examining the sensor output at 100%. On the other hand, if you're comparing final images of a fixed size (for example, 13x19 prints), the rules change. In that case, higher pixel pitch improves resolution, DR, *and* noise characteristics of the resulting image.The best results come from oversampling - when the sensor's pixel count is several times larger than the resolution of the final image.
Camera's "comfort zone" is not just about light levels. Large sensors generally have more DR, which makes a big difference in high-contrast scenes.
sderdiarian: And the value winner is...the Pentax WG3. Same lens and other specs as the re-badged WG4, but sells for $249 on Amazon. Look no further.
Agreed. The faster lens makes a huge difference in usability.
jeffharris: With such small sensors, If manufacturers would keep the megapixels LOWER the image quality could be much better! 10MP or below is perfectly good for cameras like theses.
I have Olympus TG1 and an earlier Tough camera (TC130?)
Increasing megapixels actually *increases* the IQ of the final image. A large-MP image delivers high resolution, though with noise and shallow DR.Then, if you like, you can use a combination of NR and downsiampling to obtain a smaller image with less noise and deeper DR - the same IQ you would've gotten from a lower-MP sensor, but with more resolution.
Mike, IIRC, Sony tough cams have touchscreens, which don't work when wet.
PedroMZ: I cannot understand why manufacturers do not use the larger sensor from the G series Canon or the Lumix LX 7 for example. The cameras would still be very pocketable (the zooms are only very modest anyway) and the IQ would be very substantially improved. These cameras are often the ones you have with you in wonderful unspoilt countryside with great landscape opportunities yet you are rewarded with images of such mediocre quality (even at ISO 100) that anything bigger than A4 is not contemplatable.
Downsizing images with good software on a powerful PC will improve iQ by about the same margin.
to Mike FL:Mike, nothing stops you from reducing the resolution in postprocessing. Or setting the camera to shoot in lower resolution (I often do so).
duplicate post - deleted