The D2Xs/D300 sensors have a density of 3.3MP per square cm.
The D3/D700 share a density of 1.4MP per square cm. and much bigger pixels.
Clarity of detail is dependent upon how many of the sensors photosites (pixels) that particular area of detail falls upon.
An individual photosite (sensor pixel) does not record any variation in either chroma or luma that may fall across it, all light of whatever variety of colour or luminosity that falls across an individual photosite is scrambled to render a single average value.
This value then triggers the release of electrons which are gathered up and whisked away to the image processor at the end of the exposure.
That's a VERY simple version of how it works anyway
The more pixels/photosites a particular area of detail falls upon when it is projected onto the sensor by the lens, then the greater the amount of recorded detail will be.
And vice versa...!
I never expected the D3 to perform anywhere near as well as the D2Xs.
I was thinking though that perhaps the D3 might give me slightly more sharpness than the D300 - it should, but I'm not really convinced that it does.
It's a really nice camera and I'd love to own it, but I'm not sure if it's going to give me 3 grands worth of advantage over the D300.
Apologies again for the technical glitch - especially to Big PhilT who put in so much effort with his responses - sorry matey, don't hit me
Posted by Paul Gibbs on August 11, 2008 at 5:35 AM
The easy thing Andy is to get a D700 then, esp as it fits all the D300 goodies and offeres most of the D3's SB900 related goodness too.
Going back to the discussion with PhilT, assuming you have the same number of pixels on a subject, IE the D3/700 is closer to the subject than the D2x/D300, then the educated guess is that pixel per pixel the D3 should be sharper than the D2x, with much easier design for the microlenses and larger photosites.
However my gut feeling is that the D2X is sharper per pixel with the D300 some where in between. Though i think the D40 is still the sharpest perpixel, though you got alot of moire due to the cheaper AA filter used.
So even from Phillip's point of view, with both cameras in raw resolution the D3 is very close to the D2X, and out resolves it in the vertical axis.
Though these tests were run in jpeg which will cause issues with the different processing in both cameras not being equal.
Posted by Andy Astbury on August 11, 2008 at 6:58 AM
To be honest Paul I just go by the results - not the numbers
Posted by Paul Gibbs on August 11, 2008 at 7:12 AM
Same here, which is why i think the D2X is sharper.
The D3/D700 does offer alot of flexability though, with the ability to shoot in conditions i never could with the D2X without flash, or hand held.
Colour wise i'm noticing a slight shift from the D2x images, but in general the images are easily correctable in Capture one, though Phase one have told me they wont be releasing an update until autumn. The dust shaker is a big update to the D700 though, which the D3 sorely needs.
Posted by Mike Smith on August 11, 2008 at 7:50 AM
Isn't is a bit more complex than simply the number of pixels per inch, as I understand it (and correct me if I am wrong) but each Pixel/photosite is calibrated to record either Red, Green or Blue. The exception is the Foveon sensors which have a more film emulsion like structure of vertically stacked photosites. So the sequencing and the positioning of the photosites can also affect the image output quality
Each sensor has a sequence of RGB photosites that are staggered to (usually) give a hexagonal patterning, but the manufacturers have recently been playing with this sequencing and layout of the RGB to improve apparent colour tone/sharpnss of images, one manufacturer (and I cannot remeber which) has alternating lines of alternating green photosites ie the patterning is not an even RGB each line.
So unless the photosite sequencing is identical between each type of sensor you cannot make a straight deduction that the pixel density is affecting image quality
Posted by Major Gilbert on August 11, 2008 at 9:25 AM
Andy what are you like as they say in Norfolk.
Stop trying to justify keeping your D2X's and go and get a D3. You know it makes sense(Del boy)
Oh forgot to say you should have gone to Spec Savers.(Even bigger grin).
Discussion
5 replieshttp://wildlifeinpixels.com/nikon.html
with the following addendum.
The D2Xs/D300 sensors have a density of 3.3MP per square cm.
The D3/D700 share a density of 1.4MP per square cm. and much bigger pixels.
Clarity of detail is dependent upon how many of the sensors photosites (pixels) that particular area of detail falls upon.
An individual photosite (sensor pixel) does not record any variation in either chroma or luma that may fall across it, all light of whatever variety of colour or luminosity that falls across an individual photosite is scrambled to render a single average value.
This value then triggers the release of electrons which are gathered up and whisked away to the image processor at the end of the exposure.
That's a VERY simple version of how it works anyway
The more pixels/photosites a particular area of detail falls upon when it is projected onto the sensor by the lens, then the greater the amount of recorded detail will be.
And vice versa...!
I never expected the D3 to perform anywhere near as well as the D2Xs.
I was thinking though that perhaps the D3 might give me slightly more sharpness than the D300 - it should, but I'm not really convinced that it does.
It's a really nice camera and I'd love to own it, but I'm not sure if it's going to give me 3 grands worth of advantage over the D300.
Apologies again for the technical glitch - especially to Big PhilT who put in so much effort with his responses - sorry matey, don't hit me
Going back to the discussion with PhilT, assuming you have the same number of pixels on a subject, IE the D3/700 is closer to the subject than the D2x/D300, then the educated guess is that pixel per pixel the D3 should be sharper than the D2x, with much easier design for the microlenses and larger photosites.
However my gut feeling is that the D2X is sharper per pixel with the D300 some where in between. Though i think the D40 is still the sharpest perpixel, though you got alot of moire due to the cheaper AA filter used.
Interestingly:
Nikon D2X Absolute resolution Horiz LPH 2400 Vert LPH 2000
Nikon D3 Absolute resolution Horiz LPH 2200 Vert LPH 2200
http://www.dpreview.com/reviews/nikond2x/page27.asp
http://www.dpreview.com/reviews/NikonD3/page33.asp
So even from Phillip's point of view, with both cameras in raw resolution the D3 is very close to the D2X, and out resolves it in the vertical axis.
Though these tests were run in jpeg which will cause issues with the different processing in both cameras not being equal.
The D3/D700 does offer alot of flexability though, with the ability to shoot in conditions i never could with the D2X without flash, or hand held.
Colour wise i'm noticing a slight shift from the D2x images, but in general the images are easily correctable in Capture one, though Phase one have told me they wont be releasing an update until autumn. The dust shaker is a big update to the D700 though, which the D3 sorely needs.
Each sensor has a sequence of RGB photosites that are staggered to (usually) give a hexagonal patterning, but the manufacturers have recently been playing with this sequencing and layout of the RGB to improve apparent colour tone/sharpnss of images, one manufacturer (and I cannot remeber which) has alternating lines of alternating green photosites ie the patterning is not an even RGB each line.
So unless the photosite sequencing is identical between each type of sensor you cannot make a straight deduction that the pixel density is affecting image quality
Stop trying to justify keeping your D2X's and go and get a D3. You know it makes sense(Del boy)
Oh forgot to say you should have gone to Spec Savers.(Even bigger grin).
Major.