Advice on buying camera

Discussion in 'Photography' started by Jack, Dec 29, 2013.

  1. Jack

    Savageduck Guest

    On 2013-12-30 15:19:13 +0000, "Jack" <> said:

    > Well thankyou all yet more. I have saved some of these posts to study
    > properly for when I make my final decision.


    To add to what has been said here is my 2 cents worth.

    I agree, that unless you are planning large prints you can capture more
    than adequate images with a 7MP DSLR, especially if you are only
    considering online presentation. These days I would look at something
    along the lines of the D7100 in APC-S or D610/D800 for FF.
    This is a Raymond Han I shot back in 2004 with my lowly D70 at the
    Munson, Williams, Procter Art Institute in Utica, NY under available
    light.
    < https://dl.dropbox.com/u/1295663/FileChute/DSC_0548-E1c.jpg >

    The most critical thing is getting white balance right, and to do that
    ypu should be shooting in RAW and have a WB reference shot of a WB
    reference card under the same light conditions as your subject.
    I use the WhiBal card which just simplifies the WB problem under
    problematic or mixed light sources.
    < http://michaeltapesdesign.com/whibal.html>

    If this is going to be a primary purpose for your camera I would look
    into buying, or at least first renting a tilt-shift lens.
    Take a look at what an establishment such as LenProToGo has to offer.
    < https://www.lensprotogo.com >

    --
    Regards,

    Savageduck
     
    Savageduck, Dec 30, 2013
    #21
    1. Advertisements

  2. Jack

    Jack Guest

    "Peabody" <> wrote in message
    news:...
    > Just to follow up on a couple things.
    >
    > It's impossible to predict how a lens will behave in a specific situation,
    > so
    > you have to experiment. But at the most basic level, if your camera is
    > positioned opposite the center of the painting, then the center of the
    > painting is closest to the lens, and every point farther away from the
    > center
    > of the painting is farther from the lens. Specifically, the mid-points of
    > each side are closer to the lens than the corners. This alone can lead to
    > some barrel distortion.


    I found your post really interesting but what you say at this point
    particularly intrigued me. Geometry in isolation tells me that there should
    be no issue, so is there some empirical criterion at play?
     
    Jack, Dec 30, 2013
    #22
    1. Advertisements

  3. Jack wrote:
    > Hi,
    >
    > I need a camera to photograph artwork, so it doesn't need much in the way
    > of gizmos; it really just needs reasonbly good picture quality and high
    > resolution.
    >
    > I was thinking of getting a Nikon Coolpix L27 or (since I don't really
    > want to go beneath 20 Mpx, unless I am persuaded otherwise) a S3500.
    > However, these cameras are so much cheaper than the vast majority (or so it
    > seems) of others of similarly high res that I am wondering about whether the
    > picture quality will really keep up with the resolution. I don't know about
    > these things but if there's a trade-off between res and picture quality for
    > any given price band, I'll aim to maintain a balance as I don't want to end
    > up with a lot of superfluous pixels; but want to keep the price as low as
    > possible.
    >
    > In case it makes a difference price-wise, I have already got a
    > perfectly good lens from my old pre-digital Canon EOS.
    >
    > Also would prefer a remote shutter release socket; and - less importantly,
    > perhaps - remote flash socket.
    >
    > Can anyone offer any advice?
    >
    > With thanks in advance.


    I have a very nice point and shoot, Panasonic DMC-ZS8, that I
    use for work photos at refineries and chem plants world wide.
    It took me months of handling cameras at stores and
    reading reviews to find one that exactly fit my needs.

    What I have discovered is that P&S large zooms are
    not dust proof. I take about 500 photos per month in dusty
    environments and spend about 4 hours every 4 weeks with a large
    magnifying glass and microscope to take the cam apart with it's
    27 micro screws and 7 micro cables and cleaning the sensor,
    and trying to clean the inside of the lenses that don't come apart
    but have slots and holes for dust entry.
    I got a waterproof case (as suggested by Mr. Duck) for it but found
    it's not practical about 50% of the time.
    So, if you get dust on the sensor that's the end of your P&S
    unless you are really technical and are able to clean it and
    have time and tools to do so.
     
    Paul in Houston TX, Dec 31, 2013
    #23
  4. Jack

    Savageduck Guest

    On 2013-12-30 18:29:29 +0000, Savageduck <savageduck1@{REMOVESPAM}me.com> said:

    > On 2013-12-30 15:19:13 +0000, "Jack" <> said:
    >
    >> Well thankyou all yet more. I have saved some of these posts to study
    >> properly for when I make my final decision.

    >
    > To add to what has been said here is my 2 cents worth.
    >
    > I agree, that unless you are planning large prints you can capture more
    > than adequate images with a 7MP DSLR, especially if you are only
    > considering online presentation. These days I would look at something
    > along the lines of the D7100 in APC-S or D610/D800 for FF.
    > This is a Raymond Han I shot back in 2004 with my lowly D70 at the
    > Munson, Williams, Procter Art Institute in Utica, NY under available
    > light.
    > < https://dl.dropbox.com/u/1295663/FileChute/DSC_0548-E1c.jpg >
    >
    > The most critical thing is getting white balance right, and to do that
    > ypu should be shooting in RAW and have a WB reference shot of a WB
    > reference card under the same light conditions as your subject.
    > I use the WhiBal card which just simplifies the WB problem under
    > problematic or mixed light sources.
    > < http://michaeltapesdesign.com/whibal.html>
    >
    > If this is going to be a primary purpose for your camera I would look
    > into buying, or at least first renting a tilt-shift lens.
    > Take a look at what an establishment such as LenProToGo has to offer.
    > < https://www.lensprotogo.com >


    Here are two more 2005, D70 gallery shots, a Picasso print and a Philip
    Guston, Porch No.2 (1947)
    < https://dl.dropbox.com/u/1295663/FileChute/DSC_0543-E1c.jpg >
    < https://dl.dropbox.com/u/1295663/FileChute/DSC_0549-E1c.jpg >

    --
    Regards,

    Savageduck
     
    Savageduck, Dec 31, 2013
    #24
  5. Jack

    Peabody Guest

    Jack says...

    >> It's impossible to predict how a lens will behave in a
    >> specific situation, so you have to experiment. But at
    >> the most basic level, if your camera is positioned
    >> opposite the center of the painting, then the center of
    >> the painting is closest to the lens, and every point
    >> farther away from the center of the painting is farther
    >> from the lens. Specifically, the mid-points of each
    >> side are closer to the lens than the corners. This
    >> alone can lead to some barrel distortion.


    > I found your post really interesting but what you say at
    > this point particularly intrigued me. Geometry in
    > isolation tells me that there should be no issue, so is
    > there some empirical criterion at play?


    I'm not sure what you're asking. I just meant that if you
    are close to the subject painting, the corners will appear
    to recede into the distance because they are significantly
    farther away, just as the top of a tall building would do if
    you are close to it. But if you move back away from the
    painting, the distances from each point on the painting to
    the lens become more equal in optical terms, and the
    distortion goes away, just as the tall building becomes more
    purely vertical if you get far enough away from it.

    At least that's what the geometry looks like to me.
     
    Peabody, Dec 31, 2013
    #25
  6. Jack

    Savageduck Guest

    On 2013-12-31 06:12:17 +0000, Peabody <> said:

    > Jack says...
    >
    > >> It's impossible to predict how a lens will behave in a
    > >> specific situation, so you have to experiment. But at
    > >> the most basic level, if your camera is positioned
    > >> opposite the center of the painting, then the center of
    > >> the painting is closest to the lens, and every point
    > >> farther away from the center of the painting is farther
    > >> from the lens. Specifically, the mid-points of each
    > >> side are closer to the lens than the corners. This
    > >> alone can lead to some barrel distortion.

    >
    > > I found your post really interesting but what you say at
    > > this point particularly intrigued me. Geometry in
    > > isolation tells me that there should be no issue, so is
    > > there some empirical criterion at play?

    >
    > I'm not sure what you're asking. I just meant that if you
    > are close to the subject painting, the corners will appear
    > to recede into the distance because they are significantly
    > farther away, just as the top of a tall building would do if
    > you are close to it. But if you move back away from the
    > painting, the distances from each point on the painting to
    > the lens become more equal in optical terms, and the
    > distortion goes away, just as the tall building becomes more
    > purely vertical if you get far enough away from it.
    >
    > At least that's what the geometry looks like to me.



    Many times, no matter what precautions you might take there is always a
    problem with top/bottom vertical "keystoning" distortion, and if the
    camera is not square to the flat surface, but level, left/right
    horizontal distortion.
    There are a few ways of dealing with that problem today beyond the
    standard advice of using a lens with minimum barrel/pincushion
    distortion, and proper positioning of the lens face relative to the
    subject painting surface.

    One way is to use a tilt-shift lens if available.

    Both Photoshop CC and Lightroom 5 have a new "Upright Filter" which
    along with using the correct lens profile does an amazing job of
    correcting horizon level, vertical lines, vertical
    (top/bottom)distortion, and horizontal (left/right)distortion. It is so
    good you can get away with not using a tilt-shift lens.

    Here is how it works in Lightroom, PS CC works the same way.
    <
    http://tv.adobe.com/watch/whats-new-in-lightroom-5/lightroom-5-correcting-perspective-using-upright/
    >


    --
    Regards,

    Savageduck
     
    Savageduck, Dec 31, 2013
    #26
  7. Jack

    PeterN Guest

    On 12/29/2013 11:30 PM, Peabody wrote:
    > Jack says...
    >
    > > I need a camera to photograph artwork, so it doesn't
    > > need much in the way of gizmos; it really just needs
    > > reasonbly good picture quality and high resolution.

    >
    > I've done all the photography for a local painter who works
    > in oils. I'm not a professional, but the pictures came out
    > quite well. In the beginning I shot with an 8mp Canon P&S,
    > the A590is, jpegs, and that was plenty of resolution for
    > medium-size posters and giclees. An oil painting is a
    > pretty "soft" subject in the first place, and the resolution
    > isn't really that important.


    It depends on the painting, and the purpose for taking the image. Ihe
    the purpose is for a viewer just to get an idea of what the paingint is,
    you are correct. However, if the purpose is to reproduce an piece of
    art, that is a different issue. We don't really now what the purpose for
    the image is.



    >
    > Moreover, as long as you have lots of light, a small camera
    > will take very nice pictures of paintings. You don't have
    > to be concerned with low light, or shallow depth of field,
    > where large cameras would have a big advantage. As long as
    > you get the alignment, focus and color right, a good P&S
    > should work fine.
    >


    Justy a minor point. As smaller sensor will give greater DOF. But you
    are correct. DOF is not an issue.

    --
    PeterN
     
    PeterN, Dec 31, 2013
    #27
  8. Jack

    Savageduck Guest

    On 2013-12-31 11:35:24 +0000, PeterN <> said:

    > On 12/29/2013 11:30 PM, Peabody wrote:
    >> Jack says...
    >>
    >> > I need a camera to photograph artwork, so it doesn't
    >> > need much in the way of gizmos; it really just needs
    >> > reasonbly good picture quality and high resolution.

    >>
    >> I've done all the photography for a local painter who works
    >> in oils. I'm not a professional, but the pictures came out
    >> quite well. In the beginning I shot with an 8mp Canon P&S,
    >> the A590is, jpegs, and that was plenty of resolution for
    >> medium-size posters and giclees. An oil painting is a
    >> pretty "soft" subject in the first place, and the resolution
    >> isn't really that important.

    >
    > It depends on the painting, and the purpose for taking the image. Ihe
    > the purpose is for a viewer just to get an idea of what the paingint
    > is, you are correct. However, if the purpose is to reproduce an piece
    > of art, that is a different issue. We don't really now what the purpose
    > for the image is.


    If a true reproduction of the painting is the purpose then there is
    only one current method which will get the job done.
    <
    http://www.designboom.com/art/oce-3...productions-of-fine-art-paintings-09-30-2013/
    >

    < http://gizmodo.com/3d-printing-fine-art-fakes-is-here-to-stay-1383456733 >

    >>
    >> Moreover, as long as you have lots of light, a small camera
    >> will take very nice pictures of paintings. You don't have
    >> to be concerned with low light, or shallow depth of field,
    >> where large cameras would have a big advantage. As long as
    >> you get the alignment, focus and color right, a good P&S
    >> should work fine.
    >>

    >
    > Justy a minor point. As smaller sensor will give greater DOF. But you
    > are correct. DOF is not an issue.


    Agreed, this is a case where DoF is not an issue. However, One has to
    be quite precise with regard to exposure settings, as even with a
    quality set up, shooting wide open can lead to edge & corner softening.

    --
    Regards,

    Savageduck
     
    Savageduck, Dec 31, 2013
    #28
  9. Jack

    Jack Guest

    "Peabody" <> wrote in message
    news:...
    > Jack says...
    >
    > >> It's impossible to predict how a lens will behave in a
    > >> specific situation, so you have to experiment. But at
    > >> the most basic level, if your camera is positioned
    > >> opposite the center of the painting, then the center of
    > >> the painting is closest to the lens, and every point
    > >> farther away from the center of the painting is farther
    > >> from the lens. Specifically, the mid-points of each
    > >> side are closer to the lens than the corners. This
    > >> alone can lead to some barrel distortion.

    >
    > > I found your post really interesting but what you say at
    > > this point particularly intrigued me. Geometry in
    > > isolation tells me that there should be no issue, so is
    > > there some empirical criterion at play?

    >
    > I'm not sure what you're asking. I just meant that if you
    > are close to the subject painting, the corners will appear
    > to recede into the distance because they are significantly
    > farther away, just as the top of a tall building would do if
    > you are close to it. But if you move back away from the
    > painting, the distances from each point on the painting to
    > the lens become more equal in optical terms, and the
    > distortion goes away, just as the tall building becomes more
    > purely vertical if you get far enough away from it.
    >
    > At least that's what the geometry looks like to me.
    >


    I had thought you were talking about some aspect of barrel distortion that
    will not be corrected by the Photoshop method; but now I don't think you
    are. I may be splitting hairs but I think I am right in saying that it's not
    distance from the lens that causes the distortion, but distance from the
    *centre* of the lens of the components of the image as they fall on the
    sensor (or, if you like, on the lens glass).
     
    Jack, Dec 31, 2013
    #29
  10. Jack

    Savageduck Guest

    On 2013-12-31 14:16:33 +0000, "Jack" <> said:

    >
    > "Peabody" <> wrote in message
    > news:...
    >> Jack says...
    >>
    >>>> It's impossible to predict how a lens will behave in a
    >>>> specific situation, so you have to experiment. But at
    >>>> the most basic level, if your camera is positioned
    >>>> opposite the center of the painting, then the center of
    >>>> the painting is closest to the lens, and every point
    >>>> farther away from the center of the painting is farther
    >>>> from the lens. Specifically, the mid-points of each
    >>>> side are closer to the lens than the corners. This
    >>>> alone can lead to some barrel distortion.

    >>
    >>> I found your post really interesting but what you say at
    >>> this point particularly intrigued me. Geometry in
    >>> isolation tells me that there should be no issue, so is
    >>> there some empirical criterion at play?

    >>
    >> I'm not sure what you're asking. I just meant that if you
    >> are close to the subject painting, the corners will appear
    >> to recede into the distance because they are significantly
    >> farther away, just as the top of a tall building would do if
    >> you are close to it. But if you move back away from the
    >> painting, the distances from each point on the painting to
    >> the lens become more equal in optical terms, and the
    >> distortion goes away, just as the tall building becomes more
    >> purely vertical if you get far enough away from it.
    >>
    >> At least that's what the geometry looks like to me.
    >>

    >
    > I had thought you were talking about some aspect of barrel distortion that
    > will not be corrected by the Photoshop method; but now I don't think you
    > are. I may be splitting hairs but I think I am right in saying that it's not
    > distance from the lens that causes the distortion, but distance from the
    > *centre* of the lens of the components of the image as they fall on the
    > sensor (or, if you like, on the lens glass).


    Today, Photoshop and Lightroom both do a very good job of correcting
    "barrel" and "pincushion" distortion. Subtle distortions are fixed
    using lens profiles. In CS4-CS6 the "Lens Correction" filter can fix
    most of those major distortion problems.
    With PS CC & Lightroom 5 there are the "Lens Correction" filter, and
    the new "Upright" filter found in ACR (or the "Camera RAW" filter) &
    LR5 and that is truly amazing.

    Here is a video explaining perspective correction with the "Upright"
    filter in LR5 (it works the same way in PS CC, ACR).
    <
    http://tv.adobe.com/watch/whats-new-in-lightroom-5/lightroom-5-correcting-perspective-using-upright/
    >


    --
    Regards,

    Savageduck
     
    Savageduck, Dec 31, 2013
    #30
  11. Jack

    Peabody Guest

    Savageduck says...

    >>> Moreover, as long as you have lots of light, a small
    >>> camera will take very nice pictures of paintings. You
    >>> don't have to be concerned with low light, or shallow
    >>> depth of field, where large cameras would have a big
    >>> advantage. As long as you get the alignment, focus
    >>> and color right, a good P&S should work fine.


    >> Justy a minor point. As smaller sensor will give
    >> greater DOF. But you are correct. DOF is not an issue.


    > Agreed, this is a case where DoF is not an issue.
    > However, One has to be quite precise with regard to
    > exposure settings, as even with a quality set up,
    > shooting wide open can lead to edge & corner softening.


    Yes, I should have been more precise. What I meant was you
    don't have to worry about blurring out the background
    and generating nice creamy bokeh, which would normally be
    very hard to achieve with a small sensor camera.

    Basically, I'm just saying that this is a task where the
    advantages of using a fancy expensive camera are not so
    apparent. You're going to set your ISO to 100, your
    aperture to the sweet spot for sharpness (somewhere in the
    middle for most lenses), and then use shutter speed to get
    the right exposure - neither the subject nor the camera is
    moving, so it's easy to give yourself plenty of light.
    Moreover, since you're photographing a painting, it's highly
    unlikely there will be any dynamic range issues - all the
    detail should be visible at both ends.

    So in a sense this task is the least demanding for the
    camera. It does need decent lens/sensor performance so as
    to produce sharp images, and it does need to get the color
    right, but even P&S cameras these days do really well at
    those tasks provided you have enough light. Most of the
    expanded capabilities you get with a high-end camera just
    aren't needed for this. But getting the lighting right, and
    positioning the camera properly, are still the difficult
    parts no matter what camera you use.
     
    Peabody, Dec 31, 2013
    #31
  12. Jack

    Peabody Guest

    Savageduck says...

    > Both Photoshop CC and Lightroom 5 have a new "Upright
    > Filter" which along with using the correct lens profile
    > does an amazing job of correcting horizon level,
    > vertical lines, vertical (top/bottom)distortion, and
    > horizontal (left/right)distortion. It is so good you can
    > get away with not using a tilt-shift lens.


    But doesn't that filter depend on cues in the picture -
    straight lines or whatever? A painting may not contain any
    such cues, or may contain some that aren't supposed to be
    straight. Well I suppose you could try to depend on the
    stretcher edges for that, but in my experience artists
    aren't so familiar with the meaning of "square" when they
    stretch canvas.

    Would the filter work on your Picasso or Guston examples?
     
    Peabody, Dec 31, 2013
    #32
  13. Jack

    Peabody Guest

    Jack says:

    > I had thought you were talking about some aspect of barrel distortion that
    >will not be corrected by the Photoshop method; but now I don't think you
    >are. I may be splitting hairs but I think I am right in saying that it's not
    >distance from the lens that causes the distortion, but distance from the
    >*centre* of the lens of the components of the image as they fall on the
    >sensor (or, if you like, on the lens glass).


    I just don't understand the point you're making. As far as correcting the
    distortion is concerned, it may well be that Photoshop can do that, but I
    think it would be more likely to succeed in that if the distortion was
    minimized in the first place.
     
    Peabody, Dec 31, 2013
    #33
  14. Jack

    Savageduck Guest

    On 2013-12-31 16:47:46 +0000, Savageduck <savageduck1@{REMOVESPAM}me.com> said:

    > On 2013-12-31 14:16:33 +0000, "Jack" <> said:
    >
    >>
    >> "Peabody" <> wrote in message
    >> news:...
    >>> Jack says...
    >>>
    >>>>> It's impossible to predict how a lens will behave in a
    >>>>> specific situation, so you have to experiment. But at
    >>>>> the most basic level, if your camera is positioned
    >>>>> opposite the center of the painting, then the center of
    >>>>> the painting is closest to the lens, and every point
    >>>>> farther away from the center of the painting is farther
    >>>>> from the lens. Specifically, the mid-points of each
    >>>>> side are closer to the lens than the corners. This
    >>>>> alone can lead to some barrel distortion.
    >>>
    >>>> I found your post really interesting but what you say at
    >>>> this point particularly intrigued me. Geometry in
    >>>> isolation tells me that there should be no issue, so is
    >>>> there some empirical criterion at play?
    >>>
    >>> I'm not sure what you're asking. I just meant that if you
    >>> are close to the subject painting, the corners will appear
    >>> to recede into the distance because they are significantly
    >>> farther away, just as the top of a tall building would do if
    >>> you are close to it. But if you move back away from the
    >>> painting, the distances from each point on the painting to
    >>> the lens become more equal in optical terms, and the
    >>> distortion goes away, just as the tall building becomes more
    >>> purely vertical if you get far enough away from it.
    >>>
    >>> At least that's what the geometry looks like to me.
    >>>

    >>
    >> I had thought you were talking about some aspect of barrel distortion that
    >> will not be corrected by the Photoshop method; but now I don't think you
    >> are. I may be splitting hairs but I think I am right in saying that it's not
    >> distance from the lens that causes the distortion, but distance from the
    >> *centre* of the lens of the components of the image as they fall on the
    >> sensor (or, if you like, on the lens glass).

    >
    > Today, Photoshop and Lightroom both do a very good job of correcting
    > "barrel" and "pincushion" distortion. Subtle distortions are fixed
    > using lens profiles. In CS4-CS6 the "Lens Correction" filter can fix
    > most of those major distortion problems.
    > With PS CC & Lightroom 5 there are the "Lens Correction" filter, and
    > the new "Upright" filter found in ACR (or the "Camera RAW" filter) &
    > LR5 and that is truly amazing.
    >
    > Here is a video explaining perspective correction with the "Upright"
    > filter in LR5 (it works the same way in PS CC, ACR).
    > <
    > http://tv.adobe.com/watch/whats-new-in-lightroom-5/lightroom-5-correcting-perspective-using-upright/


    Here
    >

    is a shot I made, before and after using the "Upright" feature in LR5.
    The horizontal perspective issue is obvious.
    Before
    < https://dl.dropbox.com/u/1295663/FileChute/screenshot_492.jpg >
    After
    < https://dl.dropbox.com/u/1295663/FileChute/screenshot_496.jpg >
    --
    Regards,

    Savageduck
     
    Savageduck, Dec 31, 2013
    #34
  15. Jack

    Savageduck Guest

    On 2013-12-31 17:14:01 +0000, Peabody <> said:

    > Savageduck says...
    >
    > > Both Photoshop CC and Lightroom 5 have a new "Upright
    > > Filter" which along with using the correct lens profile
    > > does an amazing job of correcting horizon level,
    > > vertical lines, vertical (top/bottom)distortion, and
    > > horizontal (left/right)distortion. It is so good you can
    > > get away with not using a tilt-shift lens.

    >
    > But doesn't that filter depend on cues in the picture -
    > straight lines or whatever? A painting may not contain any
    > such cues, or may contain some that aren't supposed to be
    > straight. Well I suppose you could try to depend on the
    > stretcher edges for that, but in my experience artists
    > aren't so familiar with the meaning of "square" when they
    > stretch canvas.
    >
    > Would the filter work on your Picasso or Guston examples?


    It did, with both. The canvas or frame edges are your best references.
    < https://dl.dropbox.com/u/1295663/FileChute/screenshot_497.jpg >
    < https://dl.dropbox.com/u/1295663/FileChute/screenshot_498.jpg >

    --
    Regards,

    Savageduck
     
    Savageduck, Dec 31, 2013
    #35
  16. Jack

    Savageduck Guest

    On 2013-12-31 17:23:41 +0000, Peabody <> said:

    > Jack says:
    >
    >> I had thought you were talking about some aspect of barrel distortion that
    >> will not be corrected by the Photoshop method; but now I don't think you
    >> are. I may be splitting hairs but I think I am right in saying that it's not
    >> distance from the lens that causes the distortion, but distance from the
    >> *centre* of the lens of the components of the image as they fall on the
    >> sensor (or, if you like, on the lens glass).

    >
    > I just don't understand the point you're making. As far as correcting the
    > distortion is concerned, it may well be that Photoshop can do that, but I
    > think it would be more likely to succeed in that if the distortion was
    > minimized in the first place.


    Always the best solution is going to be in the set up, but as I have
    shown sometimes taking care isn't enough. That is when PS or LR can and
    does help.

    --
    Regards,

    Savageduck
     
    Savageduck, Dec 31, 2013
    #36
  17. Jack

    Jack Guest

    >
    > Here
    >>

    > is a shot I made, before and after using the "Upright" feature in LR5. The
    > horizontal perspective issue is obvious.
    > Before
    > < https://dl.dropbox.com/u/1295663/FileChute/screenshot_492.jpg >
    > After
    > < https://dl.dropbox.com/u/1295663/FileChute/screenshot_496.jpg >
    > --


    For some reason, I find that if it's exhibiting parallax in both the
    vertical and horizontal axes, no amount of fiddling around in the lens
    correction facility of Photoshop will correct it (unless I am just
    approaching the whole thing wrongly). It seems to be to do with the fact
    that showing parallax implies being off-centre. The best bet is to make up a
    rectangle of precisely the dimensions you want the finished subject to be,
    and manipulating the image with the Edit>Transform>Distort tool. But even
    then it dosen't quite get it right, as there remains the issue of
    differences in scale caused by different proximities ot the lens.
     
    Jack, Dec 31, 2013
    #37
  18. Jack

    Savageduck Guest

    On 2013-12-31 17:26:23 +0000, Savageduck <savageduck1@{REMOVESPAM}me.com> said:

    > On 2013-12-31 16:47:46 +0000, Savageduck <savageduck1@{REMOVESPAM}me.com> said:
    >
    >> On 2013-12-31 14:16:33 +0000, "Jack" <> said:
    >>
    >>>
    >>> "Peabody" <> wrote in message
    >>> news:...
    >>>> Jack says...
    >>>>
    >>>>>> It's impossible to predict how a lens will behave in a
    >>>>>> specific situation, so you have to experiment. But at
    >>>>>> the most basic level, if your camera is positioned
    >>>>>> opposite the center of the painting, then the center of
    >>>>>> the painting is closest to the lens, and every point
    >>>>>> farther away from the center of the painting is farther
    >>>>>> from the lens. Specifically, the mid-points of each
    >>>>>> side are closer to the lens than the corners. This
    >>>>>> alone can lead to some barrel distortion.
    >>>>
    >>>>> I found your post really interesting but what you say at
    >>>>> this point particularly intrigued me. Geometry in
    >>>>> isolation tells me that there should be no issue, so is
    >>>>> there some empirical criterion at play?
    >>>>
    >>>> I'm not sure what you're asking. I just meant that if you
    >>>> are close to the subject painting, the corners will appear
    >>>> to recede into the distance because they are significantly
    >>>> farther away, just as the top of a tall building would do if
    >>>> you are close to it. But if you move back away from the
    >>>> painting, the distances from each point on the painting to
    >>>> the lens become more equal in optical terms, and the
    >>>> distortion goes away, just as the tall building becomes more
    >>>> purely vertical if you get far enough away from it.
    >>>>
    >>>> At least that's what the geometry looks like to me.
    >>>>
    >>>
    >>> I had thought you were talking about some aspect of barrel distortion that
    >>> will not be corrected by the Photoshop method; but now I don't think you
    >>> are. I may be splitting hairs but I think I am right in saying that it's not
    >>> distance from the lens that causes the distortion, but distance from the
    >>> *centre* of the lens of the components of the image as they fall on the
    >>> sensor (or, if you like, on the lens glass).

    >>
    >> Today, Photoshop and Lightroom both do a very good job of correcting
    >> "barrel" and "pincushion" distortion. Subtle distortions are fixed
    >> using lens profiles. In CS4-CS6 the "Lens Correction" filter can fix
    >> most of those major distortion problems.
    >> With PS CC & Lightroom 5 there are the "Lens Correction" filter, and
    >> the new "Upright" filter found in ACR (or the "Camera RAW" filter) &
    >> LR5 and that is truly amazing.
    >>
    >> Here is a video explaining perspective correction with the "Upright"
    >> filter in LR5 (it works the same way in PS CC, ACR).
    >> <
    >> http://tv.adobe.com/watch/whats-new-in-lightroom-5/lightroom-5-correcting-perspective-using-upright/


    Here
    >>

    > is a shot I made, before and after using the "Upright" feature in LR5.
    > The horizontal perspective issue is obvious.
    > Before
    > < https://dl.dropbox.com/u/1295663/FileChute/screenshot_492.jpg >
    > After
    > < https://dl.dropbox.com/u/1295663/FileChute/screenshot_496.jpg >


    This fix is a bit more extreme:
    < https://dl.dropbox.com/u/1295663/FileChute/screenshot_500.jpg >


    --
    Regards,

    Savageduck
     
    Savageduck, Dec 31, 2013
    #38
  19. Jack

    Savageduck Guest

    On 2013-12-31 17:40:24 +0000, "Jack" <> said:

    >>
    >> Here
    >>>

    >> is a shot I made, before and after using the "Upright" feature in LR5. The
    >> horizontal perspective issue is obvious.
    >> Before
    >> < https://dl.dropbox.com/u/1295663/FileChute/screenshot_492.jpg >
    >> After
    >> < https://dl.dropbox.com/u/1295663/FileChute/screenshot_496.jpg >
    >> --

    >
    > For some reason, I find that if it's exhibiting parallax in both the
    > vertical and horizontal axes, no amount of fiddling around in the lens
    > correction facility of Photoshop will correct it (unless I am just
    > approaching the whole thing wrongly). It seems to be to do with the fact
    > that showing parallax implies being off-centre. The best bet is to make up a
    > rectangle of precisely the dimensions you want the finished subject to be,
    > and manipulating the image with the Edit>Transform>Distort tool. But even
    > then it dosen't quite get it right, as there remains the issue of
    > differences in scale caused by different proximities ot the lens.


    I will start this response by asking; What version of PS are you using?

    If you have a version which has the "Lens Correction Filter" that can
    be used to fix perspective issues.
    Here is what can be done with that.
    < https://dl.dropbox.com/u/1295663/FileChute/screenshot_503.jpg >
    It is still tough to make the adjustment. The "Upright" filter in LR5 &
    CS CC is a great improvement and much easier to work with.

    I made the correction with LR5 which along with PS CC has the new
    "Upright" fliter.
    I was certainly not centered for that shot and there is obvious
    horizontal perspective distortion along with some "keystoning" in the
    vertical.
    Perhaps a side-by-side comparison will make the fix more obvious.
    < https://dl.dropbox.com/u/1295663/FileChute/screenshot_502.jpg >

    ....and an even more extreme shot in need of fixing.
    < https://dl.dropbox.com/u/1295663/FileChute/screenshot_500.jpg >
    --
    Regards,

    Savageduck
     
    Savageduck, Dec 31, 2013
    #39
  20. Jack

    Jack Guest

    "Floyd L. Davidson" <> wrote in message
    news:...
    > "Jack" <> wrote:
    >>Thankyou for the replies so far but to some extent you are speaking a
    >>foreign language. I learnt my photography -- exclusively in the world of
    >>the
    >>35mm (except that my EOS had a zoom as supplied on purchase) -- in the
    >>pre-digital era and have only come back to it fairly recently (and am
    >>rusty!). I have Googled APS-C but don't think I am getting very far with
    >>understanding it or knowing whether or not the Nikon Coolpix range will
    >>give
    >>me sufficiently good quality photos.

    >
    > I didn't respond to your initial article specifically because
    > you suggested that a Nikon P&S was what you might be looking
    > for.
    >
    > Most of the rest of this is just as much a waste of time and
    > energy as the idea of a P&S!
    >
    >>I am chiefly photographing oils with plenty of impasto (thickly applied)
    >>paint, and am finding that I prefer the character afforded them by
    >>sunlight
    >>catching them at a very shallow angle, or at least the character given
    >>them
    >>by daylight.

    >
    > That statement caught my attention, because it shows some truly
    > critical thought about the final product! Time for me to rethink
    > whether you are serious!
    >
    >>Artificial light gets something of a multiple-spotlights-shadow
    >>effect and even if this can be somewhat rectified, so far I still prefer
    >>daylight (not least because I haven't got any lights!).
    >>
    >>My plan is to put the artwork on an easel, at a suitable angle to the
    >>vertcal, and precisely cut and position a thin cardboard frame so as to
    >>give
    >>me guidance,

    >
    > Forget all of this. You need a permanent setup that has
    > appropriate adjustments. It has to be resetable to exactly the
    > same parameters every single time it is used. The same easel
    > and positioning mechanism, the same camera support system that
    > is stable and precisely positioned in relation to the easel.
    >
    > A "good enough" set up is a wall mounted easel with marks on the
    > floor where a relatively high quality tripod is placed.
    >
    > A remote shutter release is great, but a self timer is
    > unnecessary.
    >
    >>in the image, as to how to adjust for barrel distortion in
    >>Photoshop. This is what I have been doing with my current Optio Scorpio
    >>S4,
    >>which is 4 Mpx, but I had to devise a system of moving the painting around
    >>on a precisely marked-out grid (made more difficult by the fact that the
    >>canvass sides are never cut orthogonally, as they should be), placing it
    >>under the cardboard frame that is cut to just the right size to get a 300
    >>dpi print at life-sze (which is not far off A3), when they're all stitched
    >>together. The sheer degree of highly time-consuming stitching and
    >>adjustment
    >>required - what with the way the light catches each section differently,
    >>particularly with respect to glare from the slightly shiny regions of
    >>evenly, thinly applied paint - and the artistic corruption of the process
    >>of
    >>making the adjustments, have led me to aim for getting the entire piece in
    >>one snap.

    >
    > Stitching is nonsense for this type of work. You can go as low
    > as perhaps 200 pixels per inch and get top quality prints. To
    > make A3 prints, here is a chart of MP sensor sizes, with crops
    > to the same aspect ratio as an A3 print, and the Pixels Per Inch
    > for an A3 print
    >
    > Sensor MP Cropped Pixels Size PPI for A3 Print
    > 12 4027x2848 244
    > 16 4615x3264 280
    > 24 5656x4000 342
    > 36 6046x4912 421
    >
    > There really is no need for a sensor larger than 24MP regardless
    > of how critical one might be. It may be open to question
    > whether 244 or 280 PPI is high enough resolution, but 342
    > absolutely is!
    >



    Can I take it, then, that I could in principle get an image -- stitching
    irregularities taken out of the equation -- that is as high quality as any
    camera will achieve, by using my Optio to do close-ups, which are then put
    together in Photoshop?
     
    Jack, Dec 31, 2013
    #40
    1. Advertisements

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. Bill

    camera buying advice needed

    Bill, Nov 29, 2003, in forum: Photography
    Replies:
    3
    Views:
    517
    Marvin Margoshes
    Nov 29, 2003
  2. Robert
    Replies:
    5
    Views:
    483
    Robert
    Aug 29, 2005
  3. Unforgiven

    Looking for advice on buying first digital camera

    Unforgiven, Nov 3, 2003, in forum: Australia Photography
    Replies:
    5
    Views:
    416
  4. Replies:
    3
    Views:
    330
    AaronW
    Sep 28, 2006
  5. Stig Holmberg

    Advice on buying the best (sharpest) P+S "macro" camera

    Stig Holmberg, Oct 26, 2005, in forum: Digital Point & Shoot Camera
    Replies:
    2
    Views:
    908
    Guenter Fieblinger
    Oct 30, 2005
  6. Nige
    Replies:
    5
    Views:
    439
  7. Kevin Duggan

    Advice on buying 1st Digital camera please

    Kevin Duggan, Jan 19, 2004, in forum: Digital Cameras
    Replies:
    2
    Views:
    389
    Ray Fischer
    Jan 24, 2004
  8. Robert
    Replies:
    32
    Views:
    659
    SleeperMan
    Aug 30, 2005
Loading...