denoir wrote:
I think that the X1 has a very weak AA filter.
And I also think that my claim about the X1 may have been wrong. I just looked at a couple of pictures and was surprised about the relatively poor per pixel image quality. I haven't used the X1 since I got the M9 and clearly my memory of the X1's image quality was exaggerated.
Here is a full rez image from the X1 - f/5.6 and infinity.
hmm, that doesn't look much sharper than my rokkor 24/2.8 on the NEX, but there are a number of stairstep artifacts so it's difficult to judge. i'd be very surprised if the zm 25/2.8 can't beat that (it certainly does at short distances from what i've seen.
Tariq Gibran wrote:
HUGE, huge news if Sony has solved the problem with wide angle rangefinder lenses that are mostly non-retrofocus in design.
Yep, if they could then just make the sensor a 24meg full frame and also come up with some sort of external optical VF linked with a RF mechanism to the focus cam on all the M mount lens and yet keep the price sub $3k I think it would be a runaway best seller and then some!
millsart wrote:
Yep, if they could then just make the sensor a 24meg full frame and also come up with some sort of external optical VF linked with a RF mechanism to the focus cam on all the M mount lens and yet keep the price sub $3k I think it would be a runaway best seller and then some!
The heavy lens on light body complaint is common, and certainly the lighter the lens the better, however in my mind it's a matter of technique.
With a heavy lens I never support the camera by holding the body, as you would with a DSLR. You hold by the lens barrel, and with MF glass I see no real problem. But it takes getting used to, and if you are used to big bodies it's going to feel weird.
I find it very hard to believe that cornerfix will not be needed with the CVs which have retro focus, but I hope I'm wrong.
s800 wrote:
Does the shutter actually stay open when the adapter is off -i.e. nothing at all is mounted?
GIven they have no viewfinder and are basically live view all the time the shutter needs to stay open, because when your using an alt lens, the camera doesn't even know its mounted.
If the shutter closed when the camera didn't detect the electrical contacts of the lens, you'd be stuck using native E mount only.
I suppose they could maybe do it where when you power off it closes the shutter over or soemthing.
Never was as big of issue with a DSLR because the mirror provides some protection but now with mirrorless it really is startling to see just how exposed the sensor is and how without a deep mirrorbox how right up front and center it is too
Best review quote ever! Clearly the primary target audience for this camera are not only the people that upgrade from a regular P&S but also stupid people.
I look forward to the continuation of this review, as I'm looking at the NEX 7 as my Alt shooter.
I wonder what they did to fix the corner problems, a coating on the sensor/filter, the sensor it self, or processing(looking for the signs of the issue when a shot is taken without a Sony lens connected?
The sensor is still covered with glass that is quite resilient, and I'm not too worried about damaging it. From a cleaning aspect, it's a lot easier to get in their with the mirrorless cameras.
LightShow wrote:
I wonder what they did to fix the corner problems, a coating on the sensor/filter, the sensor it self, or processing(looking for the signs of the issue when a shot is taken without a Sony lens connected?
I'm sure there will be confirmation soon.
It would be nice to know if this was fixed in hardware or software/firmware, if the latter I wonder if it could be applied to nex-3/5.
Regarding the color cast, which is obviously the big improvement with this camera, it has to be a feature of the new sensor and not software. The reason for that is simple - the camera has no idea which lens is mounted and different lenses give different color casts (otherwise it would have been a trivial thing to fix in the first place).
If a lens has a retrofocus design or not doesn't really matter. The key is the ray angle. A lens whose reare lement comes very close to the sensor and that has a small exit pupil will by necessity produce extreme angles to the sensor.
I present you with this award winning artistic impression of the process:
When a photosite is hit by a ray perpendicularly diodes separate the light into G-R-G-B (Green, Red, Green, Blue) components. When the photosite is hit at an angle the diodes don't function as they should but leak wavelenghts they should have filtered away and therefor we get the color cast towards the edges. What Sony engineers must have done (my hat off to them) is to design photosites that could take a hit at an angle without light leaking to neighbouring photosites.
Anyway, I've done some proper tests now, and I'm pretty happy with what I've found. I'll show you three different lenses, shot on the M9 without lens coding, shot the C3 and shot on the M9 with lens coding.
For those of you that don't know, the M9 uses a 6-bit 'bar code' on the lens in order for the camera to identify it. That way the camera can apply the correct EXIF and perform the necessary color corrections. These corrections have been imperfect until the last firmware update where they improved it a lot - to the extent that no CornerFix is needed for most lenses
The test I did was performed the same way I generate CornerFix profiles - I use a piece of styrofoam that I place against a window. I then press the front element of the lens against it and take the shot. The styrofoam diffueses the light so it evenly coveres the frame.
As you can see neither the corrected M9 nor the C3 are 100% perfect, but for me it's definitely good enough. Since the last firmware update on the M9 I have not felt the need to use CornerFix at all. And I very much doubt I'll have to need it for the C3. YMMV.
When a photosite is hit by a ray perpendicularly diodes separate the light into G-R-G-B (Green, Red, Green, Blue) components. When the photosite is hit at an angle the diodes don't function as they should but leak wavelenghts they should have filtered away and therefor we get the color cast towards the edges. What Sony engineers must have done (my hat off to them) is to design photosites that could take a hit at an angle without light leaking to neighbouring photosites.
Luka,
The photodiodes are monochromatic. The color filter array (RGBG) filters the light, and photons with corresponding wavelengths hit the pixels. The above images show what is called color cross talk. i.e. Once the visible spectrum is filtered through the CFA, that light should ideally go to that pixel, but some of it is going to the neighboring pixels.
What Sony must have done is to shift the CFA on top of the pixels (even the sensor in X100 does this - they mention it in one of their marketing promotional items). The idea is that instead of CFA lying dead center on the pixel, if the CFA array is shifted progressively with the pixels at the edges getting the most shift, then most of the light hitting at extreme angles can be converged into the pixels those light rays should go in to.. There are also other techniques like building "light guides" with metal stacks on top of the pixel which limit color cross talk, but the idea is the same - to prevent contamination.
Actually, looking at what you posted NEX-C3 might be a better sensor in terms of color cross talk than the M9 sensor (assuming NEX-C3 is not doing the correction in software).
"The sensor is still covered with glass that is quite resilient, and I'm not too worried about damaging it. From a cleaning aspect, it's a lot easier to get in their with the mirrorless cameras"
Douglass
Funny story: Mine got pretty cruddy by about 2 months. Had some spots that would not blow off. Bought all the sensor cleaner stuff, but it seemed to leave residue which almost looked like I had scratched the surface trying to clean it. Fiddled so much that at one point I coughed and WHAM two more spots, which defied hard scrubbing with pec pads and sensor cleaner.
Then one day, I pulled out the isopropyl. CLEAN!!!!
The photodiodes are monochromatic. The color filter array (RGBG) filters the light, and photons with corresponding wavelengths hit the pixels. The above images show what is called color cross talk. i.e. Once the visible spectrum is filtered through the CFA, that light should ideally go to that pixel, but some of it is going to the neighboring pixels.
Ajay, I was referring to the diodes in the CFA that filter the light by wavelength. I assumed that they are diodes tuned to a wavelength anyway but I suppose it could be any type of semiconductor filter where the band gap is tuned to the desired wavelength. I perhaps used the term "photosite" incorrectly - I was referring to the whole thing, CFA filter included - not just the diode that measures light intensity and that comes after the filter.
What Sony must have done is to shift the CFA on top of the pixels (even the sensor in X100 does this - they mention it in one of their marketing promotional items). The idea is that instead of CFA lying dead center on the pixel, if the CFA array is shifted progressively with the pixels at the edges getting the most shift, then most of the light hitting at extreme angles can be converged into the pixels those light rays should go in to.. There are also other techniques like building "light guides" with metal stacks on top of the pixel which limit color cross talk, but the idea is the same - to prevent contamination....Show more →
Shifting the CFA at the edges, would that really work? Wouldn't it mess up tele lenses that produce much smaller ray angles so that we'd end up with the reverse situation with tele lenses producing color shifts - or at least heavy vignetting? I know the M9's Kodak sensor uses that technique but it doesn't work all that well and I always assumed that they pushed it as far as they dared without sabotaging the output for normal and wide lenses.
Actually, looking at what you posted NEX-C3 might be a better sensor in terms of color cross talk than the M9 sensor (assuming NEX-C3 is not doing the correction in software).
They can't do it by software as different lenses produce different color casts and the C3 doesn't know which lens you've mounted when you are not using one of their lenses.
