Resizing Images - How to get best quality outcome

Discussion in 'NZ Computing' started by Alan, Nov 8, 2008.

  1. Alan

    Alan Guest

    Hi All,

    I am resizing images and I would like to understand how to get the
    best outcome possible when printing (ignoring the issue of the paper
    type).

    I am using IrfanView v4 but my question is generic I think.

    For example, I have an image that is:

    4000 x 2670 pixels at 300dpi = 338.7mm x 226.1mm

    I would like to rezise the image to be a native 290mm x 194mm (more or
    less - the aspect ratio being unchanged).

    I could do this by at least one of the following approaches (or
    another that you suggest!):


    1) Reducing the pixels to be 3425 x 2286 still at 300dpi

    What does this actually do though? Does the computer go through and
    average out the pixels (when reducing the number) hence creating
    fuzziness?


    2) Maintaining the pixels at 4000 x 2670 but increasing the resolution
    to 350dpi

    On the face of it, it seems non-sensical to increase the resolution of
    an existing image, but the number of pixels remain the same, so there
    is no averaging of pixels to worry about.


    3) Should I be taking into account the maximum resolution of my
    printer? It is an HP Laserjet with maximum resolution setting of
    ImageREt 2400 - I guess that means 2400dpi?

    If I aim for 2400dpi, then I would change the pixels to:

    27402 x 18291 at 2400dpi = 290mm x 194mm

    Perhaps that is the best approach, since adding more pixels appears
    less 'lossy' than taking them out?



    Is there one best way to do this?

    Thanks,

    --

    Alan.

    The views expressed are my own, and not those of my employer or anyone
    else associated with me.

    My current valid email address is:



    This is valid as is. It is not munged, or altered at all.

    It will be valid for AT LEAST one month from the date of this post.

    If you are trying to contact me after that time,
    it MAY still be valid, but may also have been
    deactivated due to spam. If so, and you want
    to contact me by email, try searching for a
    more recent post by me to find my current
    email address.

    The following is a (probably!) totally unique
    and meaningless string of characters that you
    can use to find posts by me in a search engine:

    ewygchvboocno43vb674b6nq46tvb
    Alan, Nov 8, 2008
    #1
    1. Advertising

  2. Alan

    Your Name Guest

    "Alan" <> wrote in message news:gf54ck$snl$...
    > Hi All,
    >
    > I am resizing images and I would like to understand how to get the
    > best outcome possible when printing (ignoring the issue of the paper
    > type).
    >
    > I am using IrfanView v4 but my question is generic I think.
    >
    > For example, I have an image that is:
    >
    > 4000 x 2670 pixels at 300dpi = 338.7mm x 226.1mm
    >
    > I would like to rezise the image to be a native 290mm x 194mm (more or
    > less - the aspect ratio being unchanged).
    >
    > I could do this by at least one of the following approaches (or
    > another that you suggest!):
    >
    >
    > 1) Reducing the pixels to be 3425 x 2286 still at 300dpi
    >
    > What does this actually do though? Does the computer go through and
    > average out the pixels (when reducing the number) hence creating
    > fuzziness?
    >
    >
    > 2) Maintaining the pixels at 4000 x 2670 but increasing the resolution
    > to 350dpi
    >
    > On the face of it, it seems non-sensical to increase the resolution of
    > an existing image, but the number of pixels remain the same, so there
    > is no averaging of pixels to worry about.
    >
    >
    > 3) Should I be taking into account the maximum resolution of my
    > printer? It is an HP Laserjet with maximum resolution setting of
    > ImageREt 2400 - I guess that means 2400dpi?
    >
    > If I aim for 2400dpi, then I would change the pixels to:
    >
    > 27402 x 18291 at 2400dpi = 290mm x 194mm
    >
    > Perhaps that is the best approach, since adding more pixels appears
    > less 'lossy' than taking them out?
    >
    >
    >
    > Is there one best way to do this?
    >
    > Thanks,


    Printing to a laser printer or inkjet printer with a resolution of more than
    150dpi is pretty pointless (assuming you are printing the image at 100% of
    it's X-Y mm size). It simply wastes time sending data that the printer
    doesn't print anyway.The printer may say it's resolution is 600dpi, 1200dpi,
    2400dpi etc., but that's when printing in only one-bit depth pages (ie. pure
    black or white). A colour / grayscale image is made of of multiple dots
    printed in a pattern to fool the eye into seeing a certain colour, as can be
    seen in old newspapers and comic books when the dots were bigger.

    You are also wasting time when you scale a large image down after inserting
    it into a Word document (or InDesign, etc.) since the application is still
    sending data the printer doesn't use. Images should always be printed at
    100% of their X-Y mm size. Images on websites should also be displayed at
    100% of their X-Y size, and at a resolution of 72pdi - 96dpi.

    Reduce the resolution of the image to 150dpi. Then scale the image to the
    290mm x 194mm in Photoshop or whatever application you're using. The image
    may go slightly blurry / lose small detail since the application has to
    average the pixels (ie. if you were scaling by 50%, then every four pixels,
    two X and two Y, has to become one pixel). In Photoshop or similar you can
    carefully use a sharpen function to clean it up a bit if it's really
    noticeable once printed ... to a large extent ignore what's on-screen since
    a screen only has a resolution of 72dpi - 96dpi and will display the images
    as blurry anyway.

    And ignore what some idiots will try tell you about resolution being
    irrelevant ... it's not, which is why a professional print company or
    newspaper will ALWAYS say to send them images at particular resolutions for
    the process they are using.
    Your Name, Nov 8, 2008
    #2
    1. Advertising

  3. Alan

    Alan Guest

    "Your Name" <> wrote in message
    news:gf56j8$hui$...
    >
    >
    > Printing to a laser printer or inkjet printer with a resolution of
    > more than
    > 150dpi is pretty pointless (assuming you are printing the image at
    > 100% of
    > it's X-Y mm size). It simply wastes time sending data that the
    > printer
    > doesn't print anyway.The printer may say it's resolution is 600dpi,
    > 1200dpi,
    > 2400dpi etc., but that's when printing in only one-bit depth pages
    > (ie. pure
    > black or white). A colour / grayscale image is made of of multiple
    > dots
    > printed in a pattern to fool the eye into seeing a certain colour,
    > as can be
    > seen in old newspapers and comic books when the dots were bigger.
    >


    Hi,

    Whilst it may be a waste of time, will I get a worse outcome if I send
    a 2400dpi image to a printer that actually only does 150dpi?

    Is it different if I send a 2400dpi image to a 250dpi printer?

    I chose those numbers to ask the generic question as to whether I will
    get a better outcome by the dpi of the image being an exact integer
    multiple of the dpi of the printer, as opposed to a non-integer
    multiple?

    >
    > You are also wasting time when you scale a large image down after
    > inserting
    > it into a Word document (or InDesign, etc.) since the application is
    > still
    > sending data the printer doesn't use. Images should always be
    > printed at
    > 100% of their X-Y mm size. Images on websites should also be
    > displayed at
    > 100% of their X-Y size, and at a resolution of 72pdi - 96dpi.
    >


    Okay - I'm not doing any of that, but noted.

    >
    > Reduce the resolution of the image to 150dpi. Then scale the image
    > to the
    > 290mm x 194mm in Photoshop or whatever application you're using. The
    > image
    > may go slightly blurry / lose small detail since the application has
    > to
    > average the pixels (ie. if you were scaling by 50%, then every four
    > pixels,
    > two X and two Y, has to become one pixel). In Photoshop or similar
    > you can
    > carefully use a sharpen function to clean it up a bit if it's really
    > noticeable once printed ... to a large extent ignore what's
    > on-screen since
    > a screen only has a resolution of 72dpi - 96dpi and will display the
    > images
    > as blurry anyway.
    >
    > And ignore what some idiots will try tell you about resolution being
    > irrelevant ... it's not, which is why a professional print company
    > or
    > newspaper will ALWAYS say to send them images at particular
    > resolutions for
    > the process they are using.
    >
    >


    Thanks,

    --

    Alan.

    The views expressed are my own, and not those of my employer or anyone
    else associated with me.

    My current valid email address is:



    This is valid as is. It is not munged, or altered at all.

    It will be valid for AT LEAST one month from the date of this post.

    If you are trying to contact me after that time,
    it MAY still be valid, but may also have been
    deactivated due to spam. If so, and you want
    to contact me by email, try searching for a
    more recent post by me to find my current
    email address.

    The following is a (probably!) totally unique
    and meaningless string of characters that you
    can use to find posts by me in a search engine:

    ewygchvboocno43vb674b6nq46tvb
    Alan, Nov 9, 2008
    #3
  4. Alan

    Alan Guest

    "ChristianKnight" <> wrote in message
    news:...


    Try opening a picture with the standard paint program and resizing
    there.
    You should have a clearer picture with the smaller
    Christ's love


    Hi,

    Thanks for that.

    Which of the three options will give me the *best* printed results (in
    the general case)?

    Thanks,

    --

    Alan.

    The views expressed are my own, and not those of my employer or anyone
    else associated with me.

    My current valid email address is:



    This is valid as is. It is not munged, or altered at all.

    It will be valid for AT LEAST one month from the date of this post.

    If you are trying to contact me after that time,
    it MAY still be valid, but may also have been
    deactivated due to spam. If so, and you want
    to contact me by email, try searching for a
    more recent post by me to find my current
    email address.

    The following is a (probably!) totally unique
    and meaningless string of characters that you
    can use to find posts by me in a search engine:

    ewygchvboocno43vb674b6nq46tvb
    Alan, Nov 9, 2008
    #4
  5. Alan

    Alan Guest

    "OZZY Boy" <> wrote in message
    news:...
    > On Sun, 9 Nov 2008 11:36:33 +1300, "Alan" <> wrote:
    >
    >>Hi All,
    >>
    >>I am resizing images and I would like to understand how to get the
    >>best outcome possible when printing (ignoring the issue of the paper
    >>type).
    >>
    >>

    >
    >
    > Go post this on the Many Photo news groups, then you should get
    > facts not just guess here..
    >
    >


    Hi,

    Thanks for replying.

    I may do that, but since I posted here, I'll engage the answers I get
    first, then see how it goes.

    Thanks,

    --

    Alan.

    The views expressed are my own, and not those of my employer or anyone
    else associated with me.

    My current valid email address is:



    This is valid as is. It is not munged, or altered at all.

    It will be valid for AT LEAST one month from the date of this post.

    If you are trying to contact me after that time,
    it MAY still be valid, but may also have been
    deactivated due to spam. If so, and you want
    to contact me by email, try searching for a
    more recent post by me to find my current
    email address.

    The following is a (probably!) totally unique
    and meaningless string of characters that you
    can use to find posts by me in a search engine:

    ewygchvboocno43vb674b6nq46tvb
    Alan, Nov 9, 2008
    #5
  6. In message <gf6arq$8fq$>, Alan wrote:

    > Whilst it may be a waste of time, will I get a worse outcome if I send
    > a 2400dpi image to a printer that actually only does 150dpi?


    Honestly, once you get up to numbers that large, I wouldn't worry about it.

    > I chose those numbers to ask the generic question as to whether I will
    > get a better outcome by the dpi of the image being an exact integer
    > multiple of the dpi of the printer, as opposed to a non-integer
    > multiple?


    There are different kinds of resampling algorithm, giving various tradeoffs
    between image quality versus computational simplicity. Last time I looked
    at Photoshop, for example, it gave you a choice between simple bilinear
    interpolation versus a more complex bicubic (slower, but better-quality)
    interpolation.

    With the better algorithms, and with typical photographic images, it makes
    very little difference whether the scaling ratio is an integer or not. And
    also, if the pixel density remains above 150dpi, it's going to be rather
    hard to tell the difference anyway.
    Lawrence D'Oliveiro, Nov 9, 2008
    #6
  7. Alan

    Enkidu Guest

    Alan wrote:
    >
    > Thanks for replying.
    >
    > I may do that, but since I posted here, I'll engage the answers I get
    > first, then see how it goes.
    >

    Hmm, Roger's replies are, by rule of thumb, less than correct or
    helpful. I'll just say that same model printers *can* produce vastly
    different results. It might be best to experiment.

    Cheers,

    Cliff

    --

    Tax is not theft.
    Enkidu, Nov 9, 2008
    #7
  8. Alan

    Alan Guest

    "Lawrence D'Oliveiro" <_zealand> wrote in
    message news:gf6b5j$7k7$...
    > In message <gf6arq$8fq$>, Alan wrote:
    >
    >> Whilst it may be a waste of time, will I get a worse outcome if I
    >> send
    >> a 2400dpi image to a printer that actually only does 150dpi?

    >
    > Honestly, once you get up to numbers that large, I wouldn't worry
    > about it.
    >
    >> I chose those numbers to ask the generic question as to whether I
    >> will
    >> get a better outcome by the dpi of the image being an exact integer
    >> multiple of the dpi of the printer, as opposed to a non-integer
    >> multiple?

    >
    > There are different kinds of resampling algorithm, giving various
    > tradeoffs
    > between image quality versus computational simplicity. Last time I
    > looked
    > at Photoshop, for example, it gave you a choice between simple
    > bilinear
    > interpolation versus a more complex bicubic (slower, but
    > better-quality)
    > interpolation.
    >
    > With the better algorithms, and with typical photographic images, it
    > makes
    > very little difference whether the scaling ratio is an integer or
    > not. And
    > also, if the pixel density remains above 150dpi, it's going to be
    > rather
    > hard to tell the difference anyway.


    Hi Lawrence,

    Once the image has been created (in whatever application), if I send
    it to print, there will be further processing though, with the
    exception of whatever the printer driver on the PC, and then the
    printer itself does (I think!)

    If that is the case, then the algorithm used to get to the end image
    won't come into it per se?

    So, if I take the same image, and resample it to 600dpi and also
    500dpi, then send each of those to print to a printer with a 150dpi
    maximum resolution, from a pure image viewing / printing application,
    will the 600dpi image come out better than the 500dpi image (in
    general)?

    Thanks,

    --

    Alan.

    The views expressed are my own, and not those of my employer or anyone
    else associated with me.

    My current valid email address is:



    This is valid as is. It is not munged, or altered at all.

    It will be valid for AT LEAST one month from the date of this post.

    If you are trying to contact me after that time,
    it MAY still be valid, but may also have been
    deactivated due to spam. If so, and you want
    to contact me by email, try searching for a
    more recent post by me to find my current
    email address.

    The following is a (probably!) totally unique
    and meaningless string of characters that you
    can use to find posts by me in a search engine:

    ewygchvboocno43vb674b6nq46tvb
    Alan, Nov 9, 2008
    #8
  9. Alan

    Alan Guest

    "Enkidu" <> wrote in message
    news:4916b0ab$...
    > Alan wrote:
    >>
    >> Thanks for replying.
    >>
    >> I may do that, but since I posted here, I'll engage the answers I
    >> get first, then see how it goes.
    >>

    > Hmm, Roger's replies are, by rule of thumb, less than correct or
    > helpful. I'll just say that same model printers *can* produce vastly
    > different results. It might be best to experiment.
    >
    > Cheers,
    >
    > Cliff
    >
    > --
    >
    > Tax is not theft.


    Hi Cliff,

    Absolutely! For a single printer, and given time, I would definately
    just try the various options.

    However, I was looking for a more generic approach that would work
    across images and printers, such that it would be *likely* (no
    guarantees of course) to give the best quality outcome from a given
    image / printer.

    Hope that makes sense!

    --

    Alan.

    The views expressed are my own, and not those of my employer or anyone
    else associated with me.

    My current valid email address is:



    This is valid as is. It is not munged, or altered at all.

    It will be valid for AT LEAST one month from the date of this post.

    If you are trying to contact me after that time,
    it MAY still be valid, but may also have been
    deactivated due to spam. If so, and you want
    to contact me by email, try searching for a
    more recent post by me to find my current
    email address.

    The following is a (probably!) totally unique
    and meaningless string of characters that you
    can use to find posts by me in a search engine:

    ewygchvboocno43vb674b6nq46tvb
    Alan, Nov 9, 2008
    #9
  10. Alan

    Your Name Guest

    "Alan" <> wrote in message news:gf6arq$8fq$...
    >
    > Hi,
    >
    > Whilst it may be a waste of time, will I get a worse outcome if I send
    > a 2400dpi image to a printer that actually only does 150dpi?
    >
    > Is it different if I send a 2400dpi image to a 250dpi printer?
    >
    > I chose those numbers to ask the generic question as to whether I will
    > get a better outcome by the dpi of the image being an exact integer
    > multiple of the dpi of the printer, as opposed to a non-integer
    > multiple?


    There's no point in going above 150dpi with a desktop printer. The printer
    will simply throw away the extra dots, so you're just extending the time it
    takes to send all those extra dots across the printer connection. As to
    whether a higher dpi image will give a "better" printout, you are not going
    to notice any real difference.

    Changing the resolution of the image doesn't affect the image's data in any
    way (assuming you are keeping the number of X-Y pixels the same). It's when
    you scale the image that you lose pixels and it gets a little blurrier.
    Your Name, Nov 9, 2008
    #10
  11. Alan

    Me Guest

    Alan wrote:
    <snip>

    Most modern inkjet photo printers will resolve detail finer than 300dpi
    (300dpi is a fairly common print standard, and default setting for
    professional wet-process photo printing using equipment like Fuji
    Frontier etc).
    At resolution greater than 300dpi, any improvement (ie to 600 dpi) is
    just visible to the naked eye at closest distance.
    Inkjet printer "headline" DPI figure like 4880x2400 dpi are droplets of
    individual coloured ink per inch (horizontal and vertical) but not
    resolvable dpi, but there is a relationship between the two. A common
    rule of thumb is to feed image data to Epson printers at divisors of
    native "best photo" resolution (2880) - usually 360 dpi. For Canon and
    HP (2400dpi) usually 300dpi.
    There are programs (QImage the best known?) that carry out interpolation
    resizing images to this "native" resolution automatically and feeding
    the printer this "optimised" data.
    So that said - what difference does it make?
    Tests I have done using Epson photo printers seem to indicate that
    maximum half-tone colour resolution is around or slightly over 600dpi
    horizontal, 450dpi vertical. For best printer performance, make sure
    that heads are aligned correctly - and recheck periodically.
    There is IMO a very slight improvement either feeding 360dpi by resizing
    and carefully sharpening the image manually (ie in photoshop), or by
    using QImage to do this automatically. That was based on my tests with
    an Epson printer, and it may be that there is more (or less) difference
    using other printers / drivers. After careful comparison of several test
    prints, I decided that the printer driver actually did a great job of
    resampling, and not to bother resizing images - except when printing
    large images (15x10 inches and larger). There may be valid reasons to
    use QImage for other capabilities of the program (software colour
    management) if that capability isn't built in to image editing software
    that you may use.
    For the best/sharpest prints, I'd suggest:
    Shooting in raw format and post-converting to tiff or jpeg using raw
    conversion software. Sharpening (typically "USM" or "High Pass"
    methods) should be the last step in software image processing. If
    shooting in jpeg format, then turn in-camera sharpening as low as
    possible so that you can sharpen the image as the last step. Sharpening
    of the image should be done with final print size in mind. The quality
    of in-camera jpeg conversion varies - with many cameras there is quite a
    large difference between in-camera jpeg and post-converted raw files.
    There's some experience / trial and error involved. For preview on
    screen for a large (12x18" print) and a 12 megapixel file, then I find
    that 100% pixel view on screen is too great - 50% pixel view at normal
    screen viewing distance gives me a good impression of how a 12x18 print
    will look - but YMMV. Selective area sharpening is recommended -
    particularly when processing a digital camera file taken at high ISO, or
    a film scan, where sharpening out of focus areas will only accentuate
    film grain or noise - but provide no benefit. Sharpening twice can be
    effective - sharpen at large radius to accentuate larger detail, then
    sharpen at small radius to bring out fine detail.
    For up-scaling images, I find that Lanczos interpolation gives the best
    result - marginally better than bi-cubic. There are other and more
    exotic interpolation methods that might be better than Lanczos, but I
    haven't seen any that live up to extraordinary claims sometimes made for
    them.
    For down-scaling images, bi-cubic is usually okay, Lanczos is no good as
    it will often result in aliasing of detail, and that can be a serious
    attention grabbing flaw in a finished print.

    Acceptable sharpness of images is subjective. Subject matter makes a
    big difference - and while <150 pixels per inch (ie a 6 megapixel
    digital image printed at 15x10") can look great if the subject is for
    example a portrait - it may look comparatively crappy as a landscape
    image if the image draws the viewer's eye in very close.
    Me, Nov 9, 2008
    #11
  12. Alan

    Me Guest

    Your Name wrote:
    > "Alan" <> wrote in message news:gf6arq$8fq$...
    >> Hi,
    >>
    >> Whilst it may be a waste of time, will I get a worse outcome if I send
    >> a 2400dpi image to a printer that actually only does 150dpi?
    >>
    >> Is it different if I send a 2400dpi image to a 250dpi printer?
    >>
    >> I chose those numbers to ask the generic question as to whether I will
    >> get a better outcome by the dpi of the image being an exact integer
    >> multiple of the dpi of the printer, as opposed to a non-integer
    >> multiple?

    >
    > There's no point in going above 150dpi with a desktop printer. The printer
    > will simply throw away the extra dots, so you're just extending the time it
    > takes to send all those extra dots across the printer connection. As to
    > whether a higher dpi image will give a "better" printout, you are not going
    > to notice any real difference.
    >
    > Changing the resolution of the image doesn't affect the image's data in any
    > way (assuming you are keeping the number of X-Y pixels the same). It's when
    > you scale the image that you lose pixels and it gets a little blurrier.
    >


    Desktop photo inkjet printers are capable of resolving much more than
    150dpi, so your claim that there's no point going above 150dpi is incorrect.
    Me, Nov 9, 2008
    #12
  13. In message <gf6e1j$ml4$>, Alan wrote:

    > "Lawrence D'Oliveiro" <_zealand> wrote in
    > message news:gf6b5j$7k7$...
    >>
    >> And also, if the pixel density remains above 150dpi, it's going to be
    >> rather hard to tell the difference anyway.

    >
    > So, if I take the same image, and resample it to 600dpi and also
    > 500dpi, then send each of those to print to a printer with a 150dpi
    > maximum resolution, from a pure image viewing / printing application,
    > will the 600dpi image come out better than the 500dpi image (in
    > general)?


    Like I said, you're probably not going to be able to tell the difference.
    Lawrence D'Oliveiro, Nov 9, 2008
    #13
  14. In message <gf7k0t$1u6$>, Me wrote:

    > Desktop photo inkjet printers are capable of resolving much more than
    > 150dpi, so your claim that there's no point going above 150dpi is
    > incorrect.


    Thereby misunderstanding the difference between dpi in a full-colour RGB
    image, and dpi in individual ink dots, which was already explained
    upthread.
    Lawrence D'Oliveiro, Nov 9, 2008
    #14
  15. Alan

    Alan Guest

    "Lawrence D'Oliveiro" <_zealand> wrote in
    message news:gf7p60$3fl$...
    > In message <gf6e1j$ml4$>, Alan wrote:
    >
    >> "Lawrence D'Oliveiro" <_zealand> wrote in
    >> message news:gf6b5j$7k7$...
    >>>
    >>> And also, if the pixel density remains above 150dpi, it's going to
    >>> be
    >>> rather hard to tell the difference anyway.

    >>
    >> So, if I take the same image, and resample it to 600dpi and also
    >> 500dpi, then send each of those to print to a printer with a 150dpi
    >> maximum resolution, from a pure image viewing / printing
    >> application,
    >> will the 600dpi image come out better than the 500dpi image (in
    >> general)?

    >
    > Like I said, you're probably not going to be able to tell the
    > difference.


    Okay - thanx!

    --

    Alan.

    The views expressed are my own, and not those of my employer or anyone
    else associated with me.

    My current valid email address is:



    This is valid as is. It is not munged, or altered at all.

    It will be valid for AT LEAST one month from the date of this post.

    If you are trying to contact me after that time,
    it MAY still be valid, but may also have been
    deactivated due to spam. If so, and you want
    to contact me by email, try searching for a
    more recent post by me to find my current
    email address.

    The following is a (probably!) totally unique
    and meaningless string of characters that you
    can use to find posts by me in a search engine:

    ewygchvboocno43vb674b6nq46tvb
    Alan, Nov 9, 2008
    #15
  16. Alan

    Alan Guest

    "Lawrence D'Oliveiro" <_zealand> wrote in
    message news:gf7pa4$3fl$...
    > In message <gf7k0t$1u6$>, Me wrote:
    >
    >> Desktop photo inkjet printers are capable of resolving much more
    >> than
    >> 150dpi, so your claim that there's no point going above 150dpi is
    >> incorrect.

    >
    > Thereby misunderstanding the difference between dpi in a full-colour
    > RGB
    > image, and dpi in individual ink dots, which was already explained
    > upthread.


    Hi Guys,

    I think I may have said in my OP, but if not, I am focusing solely on
    Laser printers here.

    Does that change anything?

    Thanks,

    --

    Alan.

    The views expressed are my own, and not those of my employer or anyone
    else associated with me.

    My current valid email address is:



    This is valid as is. It is not munged, or altered at all.

    It will be valid for AT LEAST one month from the date of this post.

    If you are trying to contact me after that time,
    it MAY still be valid, but may also have been
    deactivated due to spam. If so, and you want
    to contact me by email, try searching for a
    more recent post by me to find my current
    email address.

    The following is a (probably!) totally unique
    and meaningless string of characters that you
    can use to find posts by me in a search engine:

    ewygchvboocno43vb674b6nq46tvb
    Alan, Nov 9, 2008
    #16
  17. Alan

    Me Guest

    Lawrence D'Oliveiro wrote:
    > In message <gf7k0t$1u6$>, Me wrote:
    >
    >> Desktop photo inkjet printers are capable of resolving much more than
    >> 150dpi, so your claim that there's no point going above 150dpi is
    >> incorrect.

    >
    > Thereby misunderstanding the difference between dpi in a full-colour RGB
    > image, and dpi in individual ink dots, which was already explained
    > upthread.


    The common misunderstanding is that droplet placement in the "headline"
    dpi figure of the printer (ie 4800x2400 for some printers) can be simply
    back-calculated - by presuming a constant droplet size and placement to
    extrapolate a meaningful resolution "dpi" capability.
    Droplet size is variable, and placement effectively random - a part-tone
    colour isn't made up of adjacent droplets of fixed size, placement, and
    density. Tones are made by overlapping droplets as well as white-space
    between droplets. Ink - at the densities applied - is relatively
    transparent. Looking at my test prints under a microscope of a primary
    colour gradient, the dot size stays constant but density reduces from
    full-tone up to a certain point, then dot size reduces with white-space
    between. The dots aren't round, but oval-shaped. IIRC with my printer
    (Epson) droplet size can be be of 9 different values, with 6 colours
    applied. Watching the printer in action, the image is built up in
    layers with multiple passes of the print head over the paper.
    Better photo printers exceed the colour gamut of wet-process colour
    photographic prints, and exceed the projected longevity under
    standardised testing. Droplets are effectively invisible to the naked
    eye. The main down-sides are that (OEM) ink prices are outrageous,
    getting a great colour-accurate result isn't easy, and water-fastness
    and abrasion resistance may not be as good - or may take several hours
    or days to develop after printing.
    Me, Nov 9, 2008
    #17
  18. Alan

    Me Guest

    Alan wrote:
    > "Lawrence D'Oliveiro" <_zealand> wrote in
    > message news:gf7pa4$3fl$...
    >> In message <gf7k0t$1u6$>, Me wrote:
    >>
    >>> Desktop photo inkjet printers are capable of resolving much more
    >>> than
    >>> 150dpi, so your claim that there's no point going above 150dpi is
    >>> incorrect.

    >> Thereby misunderstanding the difference between dpi in a full-colour
    >> RGB
    >> image, and dpi in individual ink dots, which was already explained
    >> upthread.

    >
    > Hi Guys,
    >
    > I think I may have said in my OP, but if not, I am focusing solely on
    > Laser printers here.
    >
    > Does that change anything?
    >
    > Thanks,
    >

    Yes.
    Laser printers truly suck for quality photographic images.
    Me, Nov 9, 2008
    #18
  19. In message <gf7r2u$ajd$>, Me wrote:

    > Alan wrote:
    >
    >> I think I may have said in my OP, but if not, I am focusing solely on
    >> Laser printers here.


    Whoops, didn't notice that.

    >> Does that change anything?
    >>

    > Yes.
    > Laser printers truly suck for quality photographic images.


    Agreed!
    Lawrence D'Oliveiro, Nov 10, 2008
    #19
  20. Alan

    Your Name Guest

    "Alan" <> wrote in message news:gf7ql2$6k4$...
    >
    > Hi Guys,
    >
    > I think I may have said in my OP, but if not, I am focusing solely on
    > Laser printers here.
    >
    > Does that change anything?


    Not really. There's no inherent difference between the way a laser printer
    and an inkjet prints the image - they both still use dots (although inkjet
    dots do spread more than laser dots because the injet is using wet ink as
    against dry toner).
    Your Name, Nov 10, 2008
    #20
    1. Advertising

Want to reply to this thread or ask your own question?

It takes just 2 minutes to sign up (and it's free!). Just click the sign up button to choose a username and then you can ask your own questions on the forum.
Similar Threads
  1. Rindler Sigurd

    Test with weird outcome!!!

    Rindler Sigurd, Sep 3, 2003, in forum: Computer Support
    Replies:
    20
    Views:
    830
    Thund3rstruck
    Sep 3, 2003
  2. Moromax

    Betting with a known outcome

    Moromax, Jan 21, 2004, in forum: Computer Support
    Replies:
    4
    Views:
    444
    wisefool
    Jan 22, 2004
  3. Birk Binnard

    What's best for resizing images for printing?

    Birk Binnard, Aug 10, 2003, in forum: Digital Photography
    Replies:
    1
    Views:
    545
    Joseph Meehan
    Aug 10, 2003
  4. Benjamin Flogged Daley

    What is the best way to resizing images in Photoshop 7 before printing?

    Benjamin Flogged Daley, Aug 15, 2004, in forum: Digital Photography
    Replies:
    9
    Views:
    705
    Roger Troost
    Aug 22, 2004
  5. pete

    Outcome of removeable drive problem

    pete, Apr 27, 2007, in forum: Digital Photography
    Replies:
    0
    Views:
    504
Loading...

Share This Page