Professional Web Applications Themes

RAW vs JPEG comparison for D70 - Photography

I did this study to convince myself to switch over to RAW: <http://www.edgehill.net/1/?SC=go.php&DIR=Misc/photography/raw-vs-jpg&PG=1&PIC=3> It's just a wierd ugly picture in dim yellow light but shows the differences pretty clearly....

  1. #1

    Default RAW vs JPEG comparison for D70

    I did this study to convince myself to switch over to RAW:
    <http://www.edgehill.net/1/?SC=go.php&DIR=Misc/photography/raw-vs-jpg&PG=1&PIC=3>
    It's just a wierd ugly picture in dim yellow light but shows the
    differences pretty clearly.

    paul Guest

  2. #2

    Default Re: RAW vs JPEG comparison for D70

    On Sun, 16 Jan 2005 11:09:53 -0800, paul <net> wrote:
     

    Given the big difference in file size, I see no reason to always shoot in RAW
    mode... jpg is good enough for most things.

    I think you need a reason to shoot in RAW... most pictures ultimately don't
    benefit...

    If you have to zoom in to see a difference - why bother? Unless you need a small
    crop zoomed!

    Bob Guest

  3. #3

    Default Re: RAW vs JPEG comparison for D70

    The differences between shooting raw/NEF and shooting jpeg with the D70 are
    enormous and you are fooling yourself if you think there are no differences
    or else you do not understand the differences.
    If you want high quality snapshots why burden yourself with a camera as
    complex, big and heavy as the D70? There are far better alternatives for
    that purpose.
    The difference in color fidelity, artifacts and every other technical and
    aesthetic quality of the image is so enormous that the only reason to use
    jpeg with the D70 is if you do not have enough room on your storage card for
    anything but jpeg.


    bmoag Guest

  4. #4

    Default Re: RAW vs JPEG comparison for D70

    I have not used RAW until today because JPEG is pretty good & it's going
    to be a pain to change. I have been dissapointed with the sharpness of
    the images I'm getting though so I did this little test. I was suprised
    to see much more noise in the RAW at the same settings (no sharpening or
    adjustments on the JPEG (except saturation) and interestingly the RAW is
    more saturated.

    We'll see how I like the real results. One thing is the parallel port CF
    reader in the side of my laptop slows down the system and I nearly
    filled my CF card today so it's taking 22 minutes to download to an
    external drive (no room on my laptop HD) and the computer is dragging
    like a snail as I type.

    You will notice on the test that there is shadow detail revealed after
    applying a strong curve which is nearly absent in the jpeg. I think it's
    fair to zoom in to 400% for the side by side comparison because you can
    see the real difference. I actually did the comparison at 800% & it was
    real easy to see exactly what was different... nothing subjective, you
    can count the number of picels it takes to cross a sharp edge & see
    highlights that were completely absent in the JPEG. Bare eyes are just
    not good at picking that out but the added shadow detail is really
    significant. I may want to crop macros and may want to make large prints
    where it would be noticeable. Even small prints appear to have more
    information like the missing shadow detail.
    paul Guest

  5. #5

    Default Re: RAW vs JPEG comparison for D70

    On Mon, 17 Jan 2005 00:42:22 GMT, "bmoag" <net> wrote:
     

    Oh I know about the differences... but I have taken great shots in both.
     

    Because it takes better pics then any of my other cameras... including my Dimage
    which cost the same new... a crappy snapshot is crappy regardless of the
    camera... I found I needed the dynamic range of the D70, which can come out on
    the jpeg if you're lucky!
     

    ummmm no, there aren't, actually... I've tried lots of 'snapshot' cameras,
    Kodak, Olympus, Minolta... give me a big sensor!
     

    Or if you want good pictures all the time, regardless of whether or not you are
    taking snapshots, product shots, or ( level crossing safety reference shots for
    the railway, which I do ), and don't have a 'hard' shot to take, which requires
    raw. Most important shots I take end up cut down for my web site, and I only use
    raw if I have a very wide light range in the photo. But I imagine some folks
    would need raw all the time.

    But I agree, the raw is much better for important photos (as I said in my last
    post). But for everyday use, the D70 takes awesome jpegs.


    Bob Guest

  6. #6

    Default Re: RAW vs JPEG comparison for D70

    On Sun, 16 Jan 2005 20:58:12 -0800, paul <net> wrote:
     

    What your are seeing I don't really call raw as the NEF does use in
    camera processing such as color balance, sharpening, and even dark
    frame subtraction.

    It's really a tiff that has been compressed using Nikon's lossless
    compression into an NEF. I just don't see the NEF as a truly RAW file.

    In addition there isn't that much difference in size. NEFs run
    between 5 and 6 megs with the majority running in the mid 5 range.
    JPGs run just about half that. Considering the cost of the camera,
    after about 400 shots a 1 gig CF card is cheaper than film.
     

    I have an old USB 1 card reader. It downloads a 1 gig card in about 10
    to 12 minutes and doesn't even put a load on the computer.

    I can and often do find I'm running word, Firefox, Thunderbird, Agent,
    Photoshop CS, and downloading a large file all while transferring
    images from the CF card.

    Now it gets more complicated as this computer serves as a gateway for
    my other computers so my wife may be surfing the Internet, sending and
    receiving e-mail, doing searches and transferring files through this
    machine in addition to what I'm doing at the same time.

    The system does not slow down until running Photoshop CS AND my film
    scanner. Then it's not only scanning, but processing up to 5 60
    megabyte files. Then it starts page file swapping, but until that
    happens the computer shows no sign of being sluggish.

    It has one gig of DDR RAM and three HDs with a total capacity of about
    half a terabyte as do two of the other three machines.
     
    In its next incarnation the computer is going to have 1.5 or 2 Gigs of
    ram and 3, or 4 serial 250 Gig HDs in a RAID.

    <snip>

    Roger Halstead (K8RI & ARRL life member)
    (N833R, S# CD-2 Worlds oldest Debonair)
    www.rogerhalstead.com
    Roger Guest

  7. #7

    Default Re: RAW vs JPEG comparison for D70

    On Mon, 17 Jan 2005 03:50:26 -0500, Roger
    <net> wrote:
     
    >
    >What your are seeing I don't really call raw as the NEF does use in
    >camera processing such as color balance, sharpening, and even dark
    >frame subtraction.[/ref]

    From what I've read (e.g. Thom Hogan's eBook and other places), the NEF
    itself _doesn't_ have these affects applied (expect possible the dark-frame
    subtraction[**]). It _records_ the camera's settings for them at the time
    the shot was taken, but you're free to use or discard these when you
    process the file.
     

    A D70's NEF stores the raw-ish [see below] sensor data in a TIFF wrapper,
    but it isn't _really_ a TIFF image file in the sense most graphics packages
    would mean by "TIFF file".

    [*raw-ish] The D70's NEF storage _does_ involve some loss compared to the
    ultra-raw sensor readings, though according to Nikon, none that you would
    see. Roughly (according to Thom Hogan), all 12 bits are saved for shadow
    and low mid-tones; high mid-tones and highlight values are split into a
    number of different sized groups (i.e. a certain amount of rounding takes
    places within a number of intensity bands). This is (supposedly) done in a
    non-linear way that mimics the way our eyes work.

    [**] and you can stop this happening (apparently) by turning the camera off
    after it has taken the "real" photo but before it's taken (and subtracted)
    the dark-frame.

    Regards,
    Graham Holden (g-holden AT dircon DOT co DOT uk)
    --
    There are 10 types of people in the world;
    those that understand binary and those that don't.
    Graham Guest

  8. #8

    Default Re: RAW vs JPEG comparison for D70

    On Mon, 17 Jan 2005 03:50:26 -0500, Roger
    <net> wrote:
     
    >
    >What your are seeing I don't really call raw as the NEF does use in
    >camera processing such as color balance, sharpening, and even dark
    >frame subtraction.[/ref]

    All of which can be switched off during import, so it can't be using
    in-camera processing, can it?. These settings are merely recorded into
    the NEF so the importer (that truly does the processing) can apply
    them if wanted. Depending on your workflow, NEF is RAW.

    No dark frame subtraction is done unless you specifically request the
    noise reduction feature - which slows things down considerably.
     

    This isn't true at all. A NEF is a RAW that's been compressed. There
    is nothing TIFF like about it. The compression isn't actually
    lossless, but the differences are minor and definitely worth the
    double capacity you get on the CF card. You can't say that for JPEG,
    albeit half the size again, the differences are no longer minor and
    permanent color balance, sharpening, exposure damage have been
    built-in to these 8-bit quantized images.
     

    I get 150 NEF to a 1 gig card. Solution - a second $69 1Gig CF card.
    (oh, and I had to get a $200 portable 40Gb hard-disk gizmo with CF
    reader built in for use when I go on vacation - or a laptop is an
    alternative).
     
    >
    >I have an old USB 1 card reader. It downloads a 1 gig card in about 10
    >to 12 minutes and doesn't even put a load on the computer.[/ref]

    You should upgrade to USB 2.0, Just go buy a $20 card and whack it in
    your PC. Delays are frustrating, add stress, and stress is what kills
    people.

    --
    Owamanga!
    Owamanga Guest

  9. #9

    Default Re: RAW vs JPEG comparison for D70

    On Mon, 17 Jan 2005 14:49:18 +0000, Graham Holden
    <of.post> wrote:
     
    >>
    >>What your are seeing I don't really call raw as the NEF does use in
    >>camera processing such as color balance, sharpening, and even dark
    >>frame subtraction.[/ref]
    >
    >From what I've read (e.g. Thom Hogan's eBook and other places), the NEF
    >itself _doesn't_ have these affects applied (expect possible the dark-frame
    >subtraction[**]). It _records_ the camera's settings for them at the time
    >the shot was taken, but you're free to use or discard these when you
    >process the file.

    >
    >A D70's NEF stores the raw-ish [see below] sensor data in a TIFF wrapper,
    >but it isn't _really_ a TIFF image file in the sense most graphics packages
    >would mean by "TIFF file".
    >
    >[*raw-ish] The D70's NEF storage _does_ involve some loss compared to the
    >ultra-raw sensor readings, though according to Nikon, none that you would
    >see. Roughly (according to Thom Hogan), all 12 bits are saved for shadow
    >and low mid-tones; high mid-tones and highlight values are split into a
    >number of different sized groups (i.e. a certain amount of rounding takes
    >places within a number of intensity bands). This is (supposedly) done in a
    >non-linear way that mimics the way our eyes work.
    >
    >[**] and you can stop this happening (apparently) by turning the camera off
    >after it has taken the "real" photo but before it's taken (and subtracted)
    >the dark-frame.[/ref]

    Can anyone else throw some light on this dark-frame issue?

    <g>

    From the info I can find, D70 dark frame noise reduction isn't done
    unless you switch on the painfully slow noise-reduction option
    (usually reserved for long-exposures).

    From the sketchy info I can find, the trick above [**] is used to get
    the D70 to do a slow-sensor read (as part of the noise reduction
    option) but interrupt it from doing the dark-frame reduction
    (something astronomers don't want it to do).

    --
    Owamanga!
    Owamanga Guest

  10. #10

    Default Re: RAW vs JPEG comparison for D70

    In message <com>,
    Bob <Amps> wrote:
     
     [/ref]
     
     
     

    That is really a rather weak and incomplete demonstration of the
    benefits of RAW, IMO. It is really just comparing in-camera conversion
    to conversion somewhere else, and with noise-reduction.

    The biggest benefits of RAW are the increased dynamic range, and the
    one-step ajustment of color, contrast, brightness, etc.

    With the camera that I currently use, the Canon 20D, a in-camera JPEG
    taken in daylight with normal contrast literally throws away 1 stop of
    green highlights, about 1.5 stops of blue highlights, and almost 2 stops
    of red highlights when making the JPEG! That means that under daylight,
    you could increase the exposure a stop and get twice the number of
    levels representing the subject, and half the noise, for superior image
    quality. Or, you could expose as normal and capture the details in
    specular highlights better. When shooting red flowers, they often get
    destroyed in JPEG mode because they clip almost two stops lower than
    they would in a RAW file.
    --

    <>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
    John P Sheehy <komm> 
    JPS@no.komm Guest

  11. #11

    Default Re: RAW vs JPEG comparison for D70

    Owamanga wrote:
     
    >
    >
    > Can anyone else throw some light on this dark-frame issue?
    >
    > <g>
    >
    > From the info I can find, D70 dark frame noise reduction isn't done
    > unless you switch on the painfully slow noise-reduction option
    > (usually reserved for long-exposures).
    >
    > From the sketchy info I can find, the trick above [**] is used to get
    > the D70 to do a slow-sensor read (as part of the noise reduction
    > option) but interrupt it from doing the dark-frame reduction
    > (something astronomers don't want it to do).[/ref]

    I'm also real unclear on the dark frame noise reduction thing. Does it
    actually take another picture? I don't hear the shutter go off again.
    You do have to wait for the same time exposure again for bulb shots. I
    have mine set for auto noise reduction for 'long exposures' and this was
    close to a second but at ISO 200 so maybe the jpeg got noise reduced.

    One thing I noticed in this test and yesterday's shoot with RAW is that
    RAW has a LOT MORE NOISE! I worked with the basic jpegs mostly but if I
    was having trouble stretching the exposure with curves or wanted extra
    detail I'd go back & open the RAW then paste it ontop of the jpeg but
    below the curves and the noise was really bad in some cases though yes
    it was sharper. With both in there, I could click the layers on & off &
    see obvious differences. Adding sharpening to the noisy RAW images
    tended to emphasize that noise even more.

    I think I tend to prefer the noise reduction the jpeg processing creates
    unless I'm dying for sharpness or there is some subtle & critical
    dynamic range requirement. Nikon Capture noise reduction destroyed
    detail, maybe other programs could clean up RAW better?

    the test again:
    <http://www.edgehill.net/1/?SC=go.php&DIR=Misc/photography/raw-vs-jpg&PG=1&PIC=4>
    The jpeg looks natural and smooth, the RAW looks splotchy, noise reduced
    is too simplified though it does have more dynamic range than jpeg.
    Notice the brush strokes on the wall at the left and the reddish band at
    the right.

    In this one, notice the bright spot center-right that is almost gone in
    the jpeg:
    <http://www.edgehill.net/1/?SC=go.php&DIR=Misc/photography/raw-vs-jpg&PG=1&PIC=3>
    and the reddish dark background detail that is lost in the jpeg.
    paul Guest

  12. #12

    Default Re: RAW vs JPEG comparison for D70

    In message <net>,
    paul <net> wrote:
     

    The JPEG from the camera *COMES* from the same data that is in the RAW
    file. If the RAW conversions you are doing outside the camera have more
    noise, then the JPEG converter in the camera is likely smoothing away
    detail. RAW image noise reduction of any quality takes too long to do
    in-camera, so you can be very certain that in-camera loss of noise is
    also loss of detail.
    --

    <>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
    John P Sheehy <komm> 
    JPS@no.komm Guest

  13. #13

    Default Re: RAW vs JPEG comparison for D70

    Another thing: BreezeBrowser is super fast loading & browsing RAW files.
    Maybe faster than fine jpegs in irfanview. It is still a pain going
    through another step though.
    paul Guest

  14. #14

    Default Re: RAW vs JPEG comparison for D70

    In message <net>,
    paul <net> wrote:
     

    Any clean-up performed in the camera during JPEG conversion happens too
    quickly to not also "clean up" image details as if they were noise,
    also. Almost any noise-reduction program that takes a little bit of
    time to clean up an image will preserve detail as best it can while
    cleaning up noise.

    Of course, if the lens is so soft or out-of-focus, or the camera-shake
    blur is strong enough to obscure fine image detail, it isn't there to
    begin with.
    --

    <>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
    John P Sheehy <komm> 
    JPS@no.komm Guest

  15. #15

    Default Re: RAW vs JPEG comparison for D70

    On Mon, 17 Jan 2005 12:22:41 -0800, paul <net> wrote:
     
    >>
    >>
    >> Can anyone else throw some light on this dark-frame issue?
    >>
    >> <g>
    >>
    >> From the info I can find, D70 dark frame noise reduction isn't done
    >> unless you switch on the painfully slow noise-reduction option
    >> (usually reserved for long-exposures).
    >>
    >> From the sketchy info I can find, the trick above [**] is used to get
    >> the D70 to do a slow-sensor read (as part of the noise reduction
    >> option) but interrupt it from doing the dark-frame reduction
    >> (something astronomers don't want it to do).[/ref]
    >
    >I'm also real unclear on the dark frame noise reduction thing. Does it
    >actually take another picture? I don't hear the shutter go off again.
    >You do have to wait for the same time exposure again for bulb shots. I
    >have mine set for auto noise reduction for 'long exposures' and this was
    >close to a second but at ISO 200 so maybe the jpeg got noise reduced.[/ref]

    Of course, for the noise reference picture, it wouldn't need to open
    the shutter again (in fact, it wants the sensor to be dark to do the
    scan).
     

    Yes, the jpeg is masking this in it's murky algorithm.
     

    Beware when working with RAW to visit every preference of the importer
    and make sure it's not doing something you don't want it to, such as
    (this is the Photoshop RAW importer BTW):

    * Sharpening.
    * Introducing a 5% shadows (clipping the dark end of the histogram)

    ...and typically you'll want to do a bit of:

    * Anti vignetting.
    * Choosing the right color temp.
    * Exposure modification.
    * Saturation boost.
    * Color noise reduction.
    * Chromatic Aberration if using Sigma lenses. :-p
     

    Well, you can always make a JPEG from the RAW, with the right quality
    settings, you'd loose that noise too.

    Without a doubt, external algorithms (and I'm talking Photoshop here)
    are going to be superior to any in-camera ones. They don't have speed,
    memory or processor limitations that the camera firmware does.

    But realistically, so what if you can see it at 400%? The important
    thing is which one prints better.
     

    I'd prefer it if you had made a 100% version, for a more natural
    comparison. At 400% they all look to me.
     

    Yep.

    --
    Owamanga!
    Owamanga Guest

  16. #16

    Default Re: RAW vs JPEG comparison for D70

    In message <com>,
    Owamanga <com> wrote:
     

    Photoshop doesn't really have any intelligent noise-reduction, though.
    It basically only filters out noise-like things, which may actually be
    detail.
    --

    <>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
    John P Sheehy <komm> 
    JPS@no.komm Guest

  17. #17

    Default Re: RAW vs JPEG comparison for D70

    On Mon, 17 Jan 2005 21:20:58 GMT, komm wrote:
     
    >
    >Photoshop doesn't really have any intelligent noise-reduction, though.
    >It basically only filters out noise-like things, which may actually be
    >detail.[/ref]

    True, there are plenty of third-party photoshop plugins that
    (apparently) do a good job in this area. DSLR noise hasn't been an
    issue for me - I usally light stuff up before photographing it.

    <g>

    --
    Owamanga!
    Owamanga Guest

  18. #18

    Default Re: RAW vs JPEG comparison for D70

    In article <com>,
    Owamanga <com> wrote: 

    [ ... ]
     
    >
    >This isn't true at all. A NEF is a RAW that's been compressed. There
    >is nothing TIFF like about it. The compression isn't actually
    >lossless, but the differences are minor and definitely worth the
    >double capacity you get on the CF card. You can't say that for JPEG,
    >albeit half the size again, the differences are no longer minor and
    >permanent color balance, sharpening, exposure damage have been
    >built-in to these 8-bit quantized images.[/ref]

    O.K. I'm not really *sure* what is the case in the Nikon NEF,
    but it appears to be similar to the images saved by the Kodak/Nikon
    NC2000e/c which I used prior to my D70. It was built on a Nikon N90s
    body (whose autofocus seems faster than that of the D70), with a back
    which includes a heavy and large sub-base to contain the PCMCIA disk
    drive, the RAM, the CPU and the batteries.

    I spent some time trying to yze the format, so I could
    (perhaps) avoid having to use the Kodak plugin in PhotoShop Elements,
    and be free to work with my unix machines instead.

    Based on what I could see, the image format (which had a ".TIF"
    extension) in that camera had:

    1) A TIFF wrapper around both the raw image and a raw thumbnail.
    The TIFF wrapper allowed the structure to save the camera
    information and the two images in a known standard format.

    2) The raw images were truly that. The individual pixel
    information from the sensor. Note that one pixel will be
    viewing through a particular combination of filters, and the
    adjacent pixels will be viewing through other filter
    combinations. The filters seem to be two bar patterns at right
    angles, so some pixels have one filter, some have the other,
    some have both, and some have none. In any case, no pixel has
    full color information.

    Each pixel was saved as though it were a B&W Pixel (8 bits on
    the NC2000e/c, 10 bits on the D70 apparently), and it is up to
    post-processor programs to convert that into three-color pixels.
    As a result, there is a growth from a single byte of information
    (8 or 10 bits per pixel) in the raw, to three or four bytes per
    pixel (8 or 10 bits per pixel per color). Thus processing the
    RAW image into an uncompressed TIFF file results in a growth
    in size of at least a factor of three -- before the final TIFF
    trims off the extra two bits per pixel.

    So -- the plugin from Kodak, (and the software from Nikon for
    the D70) knows the pattern of the filters, and how to combine the
    information from adjacent pixels to make a proper three-color pixel in
    each location. (I suspect that there is something a bit marginal in the
    processing of the edge pixels, but they seem to work around it pretty
    well.) Of course, the in-camera JPEG processing does the same in the
    D70, but it was not available in the NC2000e/c

    Now -- I have finally found that both the D70 NEF images and the
    NC2000e/c images can be processed by the dcraw program, which will
    compile nicely on any of a mix of unix machines, much to my pleasure.

    Included in the doentation with that program is a detailed
    examination of the pixel color format on one of the other cameras which
    it handles, which is similar to the NC2000e/c and the D70.

    I don't happen to have a NEF file in my computer at the moment,
    so I can't easily repeat an earlier test, but IIRC, the unix "file"
    command decided that the ".NEF" file was truly a variant of a ".TIF"
    based on internal evidence. (The file command is quite useful for
    yzing strange files.)

    Enjoy,
    DoN.
    --
    Email: <com> | Voice (all times): (703) 938-4564
    (too) near Washington D.C. | http://www.d-and-d.com/dnichols/DoN.html
    --- Black Holes are where God is dividing by zero ---
    DoN. Guest

  19. #19

    Default Re: RAW vs JPEG comparison for D70

    On 17 Jan 2005 20:57:48 -0500, com (DoN. Nichols)
    wrote:
     
    >
    > [ ... ]

    >>
    >>This isn't true at all. A NEF is a RAW that's been compressed. There
    >>is nothing TIFF like about it. The compression isn't actually
    >>lossless, but the differences are minor and definitely worth the
    >>double capacity you get on the CF card. You can't say that for JPEG,
    >>albeit half the size again, the differences are no longer minor and
    >>permanent color balance, sharpening, exposure damage have been
    >>built-in to these 8-bit quantized images.[/ref]
    >
    > O.K. I'm not really *sure* what is the case in the Nikon NEF,
    >but it appears to be similar to the images saved by the Kodak/Nikon
    >NC2000e/c which I used prior to my D70. It was built on a Nikon N90s
    >body (whose autofocus seems faster than that of the D70), with a back
    >which includes a heavy and large sub-base to contain the PCMCIA disk
    >drive, the RAM, the CPU and the batteries.
    >
    > I spent some time trying to yze the format, so I could
    >(perhaps) avoid having to use the Kodak plugin in PhotoShop Elements,
    >and be free to work with my unix machines instead.
    >
    > Based on what I could see, the image format (which had a ".TIF"
    >extension) in that camera had:
    >
    >1) A TIFF wrapper around both the raw image and a raw thumbnail.
    > The TIFF wrapper allowed the structure to save the camera
    > information and the two images in a known standard format.
    >
    >2) The raw images were truly that. The individual pixel
    > information from the sensor. Note that one pixel will be
    > viewing through a particular combination of filters, and the
    > adjacent pixels will be viewing through other filter
    > combinations. The filters seem to be two bar patterns at right
    > angles, so some pixels have one filter, some have the other,
    > some have both, and some have none. In any case, no pixel has
    > full color information.
    >
    > Each pixel was saved as though it were a B&W Pixel (8 bits on
    > the NC2000e/c, 10 bits on the D70 apparently), and it is up to
    > post-processor programs to convert that into three-color pixels.
    > As a result, there is a growth from a single byte of information
    > (8 or 10 bits per pixel) in the raw, to three or four bytes per
    > pixel (8 or 10 bits per pixel per color). Thus processing the
    > RAW image into an uncompressed TIFF file results in a growth
    > in size of at least a factor of three -- before the final TIFF
    > trims off the extra two bits per pixel.
    >
    > So -- the plugin from Kodak, (and the software from Nikon for
    >the D70) knows the pattern of the filters, and how to combine the
    >information from adjacent pixels to make a proper three-color pixel in
    >each location. (I suspect that there is something a bit marginal in the
    >processing of the edge pixels, but they seem to work around it pretty
    >well.) Of course, the in-camera JPEG processing does the same in the
    >D70, but it was not available in the NC2000e/c
    >
    > Now -- I have finally found that both the D70 NEF images and the
    >NC2000e/c images can be processed by the dcraw program, which will
    >compile nicely on any of a mix of unix machines, much to my pleasure.
    >
    > Included in the doentation with that program is a detailed
    >examination of the pixel color format on one of the other cameras which
    >it handles, which is similar to the NC2000e/c and the D70.
    >
    > I don't happen to have a NEF file in my computer at the moment,
    >so I can't easily repeat an earlier test, but IIRC, the unix "file"
    >command decided that the ".NEF" file was truly a variant of a ".TIF"
    >based on internal evidence. (The file command is quite useful for
    >yzing strange files.)
    >[/ref]

    Okay, I stand corrected. Approximately 0.01% of the file is TIFF like,
    in that the NEF's compressed RAW data and thumbnail is wrappered in a
    TIFF structure.

    Beyond that, all similarities end.

    The D70 NEF is basically RAW as you saw, because of the fact that you
    need to de-bayerize the data before you can see the picture. I am not
    aware of the NC2000e, but if it's like the D100, the difference would
    be that the RAW data is not compressed. (You should be able to tell
    based on file-size).

    --
    Owamanga!
    Owamanga Guest

  20. #20

    Default Re: RAW vs JPEG comparison for D70

    In article <com>,
    Owamanga <com> wrote: 

    [ ... ]
     [/ref]

    [ ... ]
     

    Agreed. I spent some time trying to find something for unix
    which would display the image without processing through Kodak's plugin.
    (I now have that, thanks to "dcraw"). But the best that I could do with
    "the GIMP" (without dcraw) was to view the thumbnail in B&W, and one
    other program allowed me to view the full sized image in B&W -- and in
    either case, the image looked *way* too dark.
     

    The size of what comes out of the NC2000e is approximately 1/3 of
    the size predicted on the number of pixels and 3 byte/pixel. Slightly
    larger, to allow for the camera data and the thumbnail. Kodak opted to
    keep it in that format in the camera and convert it only when it reached
    the computer where it was to be processed to benefit from the smaller
    file size, as it was using PCMCIA disks, not compact flash cards. The
    biggest that I have found for it is a 340 MB drive, IIRC.

    That camera was originally produced in 1996 for the AP, so it
    had a couple of other features of use to a reporter:

    1) A built-in microphone for voice annotation.

    2) The ability to connect it to a cell phone to send the
    images and sounds back to the office.

    When first introduced, the price was around $14.4k, which cured
    me of wanting it at the time. :-)

    However, when I found one at $700 at a hamfest, the thought of
    all of my old Nikon glass waiting caused me to buy it then. (only 1.3
    MP, but lots more flexible than the CP 950 which I used for a while.)
    I've gotten a lot of use from it -- but I do find the D70 better for
    most of what I do -- and especially like the ability to review the shots
    on the LCD display, instead of having to wait until I get to the
    computer. And the D70 is a *lot* lighter (and quieter, since it doesn't
    have the N90s body winding the non-existent film with each shot. :-)

    Enjoy,
    DoN.

    --
    Email: <com> | Voice (all times): (703) 938-4564
    (too) near Washington D.C. | http://www.d-and-d.com/dnichols/DoN.html
    --- Black Holes are where God is dividing by zero ---
    DoN. Guest

Page 1 of 2 12 LastLast

Similar Threads

  1. Resultset Comparison
    By alagukannan in forum Coldfusion Database Access
    Replies: 1
    Last Post: December 8th, 07:59 PM
  2. ASP comparison question...
    By K in forum ASP
    Replies: 3
    Last Post: August 1st, 03:39 PM
  3. Mac - PC Comparison
    By Ed A. Ortiz in forum Adobe Photoshop 7, CS, CS2 & CS3
    Replies: 6
    Last Post: July 16th, 07:49 PM
  4. raw vs. ufs comparison
    By Jan Andres in forum Sun Solaris
    Replies: 5
    Last Post: July 16th, 10:54 AM

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139