1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

Nikon denies Nikon 1 system is dead but considers higher-end mirrorless camera

Discussion in 'News - Discussion' started by CSBC, Sep 22, 2016.

  1. Andrew Flannigan

    Andrew Flannigan Well-Known Member

    Definitely true for film but I don't see how it applies to digital sensors. If you have 16 million photosites on a 1" wafer and 16 million photosites on a 24x36 wafer then given equivalent fields of view you're going to get the same information on both, aren't you? The difference will lie in the amount of amplification you require and therefor the noise content but you'll still have split the image into the same 16 million mosaic squares.
     
  2. El_Sid

    El_Sid Well-Known Member

    True but, and it's a big but, the actual physical size of the sensor is smaller meaning that any given size of print the 1" sensor image has been enlarged more than one from a full frame sensor by a factor of 2.7. The equivalence in megapixels does go someway to offset this but any actual optical aberrations such as chroma, spherical or even just camera shake will be more obvious because of the greater degree of physical enlargement.
     
  3. Andrew Flannigan

    Andrew Flannigan Well-Known Member

    I can see how that might seem to be the case but isn't it true that abberations are reduced in line with the diameter of the lens because the angle that the light bends through is smaller due to the smaller target area? Camera shake will be less of a problem too because of the shorter focal lengths involved (as in the old rule of thumb that minimum shutter speed should be the reciprocal of the focal length). I also wonder about the magnification claim because an array of 4920×3264 pixels is the same whether it came from a tiny sensor or a big one, that's basic electronics.
     
  4. PeteRob

    PeteRob Well-Known Member

    A photograph starts with the scene focussed on the sensor by the lens. When that image is reproduced to viewing size it is enlarged and the smaller the original focussed image size (sensor size) the bigger the degree of enlargement. The lens has to be good enough to produce a quality image on the sensor that will stand that enlargement. So you cannot really build poor quality lenses into small sensor cameras and hope for A3 prints. Thus you need good quality lenses.

    The digital discussion is how many pixels do you need to resolve the image focussed on the sensor hence the pixel density is also important in enlarging the scene. Generally pixel density increases as the sensors get smaller sufficiently for the enlargement process to work.

    Another aspect is that most people view on a monitor which has far fewer pixels than the image does, so results are never properly observed.

    The old rule of thumb for minimum exposure time to avoid camera shake should count for sensor size i.e. reduce time by x1.5 for APS-C if not accounting for image stabilisation.

    If I put my mind back I am fairly sure there was a debate when the 20+ MP sensors first came out as to whether they "needed" better lenses becuase they would out-resolve the images that current lenses were capable of producing on the sensor. It was a misguided discussion I think.
     
  5. Benchista

    Benchista Which Tyler

    .

    No, because the rule of thumb only holds good (insofar as it does at all) for 35mm film; you would need to consider the equivalent focal length taking into account the crop factor. It's really an issue of, er, magnification, rather than focal length per se, which makes sense if you think about it.

    Yes, but they're smaller pixels. In good conditions, there aren't really any performance issues, but as the ISO goes up, the signal to noise ratio worsens more quickly than with larger sensors. As you say, basic electronics.

    I have a 1" sensor Canon compact, and it falls between my phone and my CSC in terms of performance. I certainly wouldn't buy a compact with a smaller sensor than that, but 1" is not a bad compromise in a very small body. Makes much less sense in a CSC IMHO.
     
  6. Andrew Flannigan

    Andrew Flannigan Well-Known Member


    I found this useful description at http://www.photocourse.com/itext/pixels/pixels1.pdf...

    In other words, the number of pixels and not the original sensor size determine the print size.
     
    exspmr likes this.
  7. PeteRob

    PeteRob Well-Known Member

    Nearly. When printing you make a choice as to what size you want the print and take into account how many pixels you have available. You can choose to interpolate between actual pixel values (upsize) or remove pixel values (downsize) according to the planned print size, viewing distance and image content. So the number of pixels is not determining but influencing. The sensor size doesn't come into it as far as the printing goes but the sensor, processing and lens used all affect what is being printed.
     
  8. 0lybacker

    0lybacker In the Stop Bath

    Japanese camera companies do seem to get in a muddle over corporate decisions, these days. Is it really pride and 'loss of face' or have they been caught out by their own rapid march into digital?
     
  9. 0lybacker

    0lybacker In the Stop Bath

    Hmmn, that doesn't sound right. Where's our sensor expert gone?
     
  10. PeteRob

    PeteRob Well-Known Member

    It's a bit simplistic. If you view at [much] more than 100% on a monitor or print at a small number of pixels per inch without upsizing you will see solid blocks ( pixillation). That is all it is saying.
     
  11. Andrew Flannigan

    Andrew Flannigan Well-Known Member

    That's the point I'm making. You'll see the same image from a FF sensor or a 1/2.3 sensor if pixel count, electronic noise, angle of view and relative lens quality are the same. The question is whether these conditions can be met in practice.
     
  12. P_Stoddart

    P_Stoddart Well-Known Member

    Well no, because although there are the same number of pixels. As pointed the image has to be generated in the 1st place by a lens. It is fair more tricky for a lens to get a clean image on a 1/2.3" area than 24x36mm area. :)

    That's why a 1" sensor IMHO strikes the balance in terms of optic and sensor. Also 1/2.3" sensor cameras have no DOF to play with. :)

    Both Sony & Panasonic now have 1" sensor bridge camera. Both perform very well.

    I think like all new formats the CX is going to take time to grow. The plus for the CX is that it has a serious lens maker supporting it ie Nikon.

    FF (24x36mm sensor) is now clearly becoming the medium format or taking it place.

    Here a piece of the puzzle as to where we are going Canon have a APS-H sensor of 250MP!

    http://www.canon.co.uk/about_us/pre...meras_accessories/aps_h_size_cmos_sensor.aspx

    That means a 366MP FF sensor could be made LOL

    It also means you could have a 50MP 1" sensor. But the CX format is aimed I believe at consumers or serious amateur not pro. Therefore 30MP should be the top for that.

    It comes down to this if you need 12x from a lens (assuming 50mm on FF as standard lens) you need to carry 600mm for FF sensor design. That is a serious piece of glassware

    http://imaging.nikon.com/lineup/lens/singlefocal/Telephoto/af-s_600mmf_4g_vr/


    Now if as a serious amateur if you could get the same reach but with less glassware which is lighter make sense

    http://www.europe-nikon.com/en_GB/p...enses/fx/af-s-nikkor-70-200mm-f-2-8e-fl-ed-vr

    Where Nikon CX cameras have a edge over the Sony & Panasonic bridge designs is that you can put fast primes on the CX bodies.

    I am expecting down the line the noise ceiling for 1" sensors to hit the magic 1000ISO mark. That put cameras using these sensors into the right all round envelope for consumer usage.

    The Nikon system can be made into travel camera with a 8.8 - 220mm lens added matches the Sony RX10 II :)
     
  13. 0lybacker

    0lybacker In the Stop Bath

    Yes, Pete, but that is what the monitor is showing you at whatever magnification level you can go to. But each individual pixel, if I have been led to understand by Fuji, Leica, Nikon, et al, and remembered correctly sits in its own space, may be shaped and/or coloured and/or have an individual filter over it and/or a group of adjacent pixels. Hence maker's claims for 'superior this' and 'better that' as well as photographers' concerns over lens diffraction and loss of fine detail when stopping down for depth of field - my brain has temporarily mislaid the technical term as I'm currently doing about three things at once not counting this post! - so yes, pixels would have a physical size and shape. Who was our sensor expert? Bawbee? One of the astro-photographers on here, IIRC. Cheers, Oly
     
  14. PeteRob

    PeteRob Well-Known Member

    Photosite not pixel. The pixel is properly the pointwise output from processing the raw data. For conventional sensors the pixel count and the photosite count are the same.
     
  15. P_Stoddart

    P_Stoddart Well-Known Member

    What we are debating is not if the Nikon 1 system can match pro kit. Its if it has a place in the CONSUMER market. :)

    I think it does, as one who has got good comments from judges over work created on a bridge camera with a sensor smaller than 1" it stand to reason good art can be created with a 1" sensor system.

    The biggest draw back to going below 1" in the sensor department is noise ceiling is going to be difficult to improve and getting DOF effect is another issue.

    So the system I think strikes the balance with having a sensor big enough for noise ceiling and DOF but small enough for small lens to give the reach consumer want at a reasonable price hit.

    Ok consumers could by the Sony RX10 II or the FZ1000 as a solution to flexible shooting with low weight. But if you want to have that extra bit of flexibility then the Nikon 1 system give you that.

    MFT does offer a similar set up but as sensor improve abit more so that noise ceiling on 1" sensor get into 4 digits then MFT might get squeezed out. :)
     
  16. RogerMac

    RogerMac Well-Known Member

    Please excuse my ignorance but could somebody point me to a simple definition of "Noise ceiling"?
     
  17. PeteRob

    PeteRob Well-Known Member

    I would understand this to mean a measure of noise below which results would be considered acceptable and above which not. The camera manufacturers impose one with their "max ISO" setting.
     
  18. RogerMac

    RogerMac Well-Known Member

    It seems to be being used as if it were a very precise scientific measurement so I would expect there to be a strict definition somewhere but all my googling only produces vague references. I can understand "Max ISO" but not "Noise ceiling"
     
  19. PeteRob

    PeteRob Well-Known Member

    There will be defined measure of signal/noise ratio. I expect there are different crtieria as to what is acceptable or not.
     
  20. Andrew Flannigan

    Andrew Flannigan Well-Known Member

    Like a lot of terminology used in internet chats, "noise ceiling" sounds terribly scientific and signifies nothing.
     
    El_Sid and Benchista like this.

Share This Page