Discussion in 'Australia Photography' started by Joe @ Seaham, Sep 22, 2003.

  1. Joe @ Seaham

    Brenton Guest

    Are you actually saying that a 10x8 picture printed at 150 DPI looks
    that same quality as a 10x8 printed at 300dpi. (assuming that the
    pictures have the same uncropped perspective (not Miro's version of
    perspective) and that you are printing on photographic paper.

    Now that is just illogical... why would ANYONE bother buying a better
    printer if the resolution is the same as a low end printer.
    PS... Your not the same person who keeps complaining about the poor
    focusing ability of your Canon 10D are you???
    Brenton, Sep 23, 2003
    1. Advertisements

  2. Joe @ Seaham

    Cameron Guest

    Good question Brenton...

    If you are dealing with a 'small' file like we are shooting with our dig
    cameras then it seems as though there is a point of diminishing returns when
    it comes to resampling up. The 'breaking up' of a file by resampling a 6mp
    file to 20"x30" at 300dpi is pretty bad but no worse than the 'breaking up'
    of a 20"x30" print at 150dpi. The file needs to be resampled a lot less
    (from 18Mb to 38.6Mb instead of from 18Mb to 154Mb - Approx only) making it
    a LOT smaller and there easier to send to the lab via ftp or email. On the
    wall however the print looks identical considering the viewing distance.

    It is a lot different if like Ken Duncan or similar panorama/large format
    photogs you are shooting 6x17 or larger and getting drum scans done. The
    resulting raw files are massive and are normally resampled down or kept
    about the same. Why not have a pin sharp file when viewed from a 10 inches
    when you have the raw data there to achieve it.

    Sorry, it was a bit hard to put into words but I hope this helps to answer
    your question a bit.

    Cameron, Sep 23, 2003
    1. Advertisements

  3. Joe @ Seaham

    Cameron Guest


    See the smily face????? It ain't personal....

    All I am saying is that there is a big difference between photo quality
    inkjet prints with 100dpi or 150dpi files compared with 300dpi files.
    Anyone else care to share their opinion?

    (.......trying not to mention your previous backflip on the greatness of the
    Sigma 120-300 2.8 lens v Canon 300 2.8L.......)

    Cameron, Sep 23, 2003
  4. Joe @ Seaham

    Snaps! Guest

    Hey Brenton...
    See the finger?
    The first D10 I bought was faulty and out of focus.
    It was replaced under Canon's 14 day replacement policy.
    I have since bought another 10D and it had a back focus issue which has been
    fixed. Now, both of my 10Ds perform as I would have expected in the first
    place. Perhaps the fact that the serial numbers are only 630 apart has
    something to do with that?
    I can't answer dumb questions like 'why would- if''

    It might interest you to know that most magazine photos are 150 LPI which
    roughly equates to 100~150 dpi from an inkjet. I instantly think of
    justifying my statements when some joker says it can't be done by offering
    to send some output. Then I reflect back to when Rudi had an issue with an
    s900 and I did a heap of research on my s9000 printer's capability. I sent
    him (at my expense) quite a few examples and told him how I'd managed to
    duplicate his problems and what I did to avoid them. His promise to keep me
    informed of the outcome imortalised the attitude of most of the poster
    here... All gunnado and no willdo. Does anybody here actually get off their
    arse and do anything?

    Different Inkjet technologies produce different results. Epson technology is
    more precise at drop placement and construction than Canon's. If you follow
    that theory then, For an Epson to produces better photographs, It MUST
    therefore put down more drops of ink to cover a given area to arrive at a
    'better' picture. Epson then, need to print 200 dpi images to achieve the
    same density and coverage as a 150 dpi Canon which can't control the
    construction of a droplet as well as Epson.

    Canon have developed a system around the shortcommings of it's thermo pumps
    so that there is little difference to the human eye in a photo printed on
    either brand. Just the same, Canon can deliver a better quality picture at
    lower resolution than Epson can.

    Snaps!, Sep 23, 2003
  5. Joe @ Seaham

    Snaps! Guest

    I'm starting to think you two have never had much to do with a 'real' photo
    lab in your lives.

    "are normally resampled down or kept about the same" What the hell does that

    This is a clip of the instruction for one of Australia's largest,
    professional, digital print labs. for photographers intending to provide
    their work for printing at any size up to 1,200mm wide.

    "Resolution - The Lambda is capable of imaging at 400dpi (continuous-tone
    resolution is effectively 4000dpi when comparing it to half-tone
    reproduction). If you require the ultimate in reproduction you should supply
    your files at 400dpi. However, because of the practical limitations of
    working at these file sizes we suggest preparing your files at 250-300dpi
    and applying sharpening which will still produce excellent results for
    Wedding, Portrait and Commercial photographic images. NOTE: L2 Overnight
    files must be prepared at 250dpi."

    Further to this... Give me a file saved at 150 dpi and I'll print it on an
    Agfa printer.
    Give me the exact same image at 300 dpi, I'll do the same and I'll bet you
    both, you won't be able to pick with the naked eye, which one is the lower
    resolution .

    If you jokers ever get your heads out of your arse long enough to look
    around, you'll discover it's not sunglasses that stoped you seeing clearly
    at all!
    Snaps!, Sep 23, 2003
  6. Joe @ Seaham

    Cameron Guest

    I thought you were the expert? When you scan a 6cmx17cm tranny at 4000dpi
    do you think the file is quite large??? Do you think that you would have to
    resample the file downwards to make a 30" wide print at 300dpi???
    Read it before, used them a few times but really, they were pretty poxy....
    F-Stop that is...
    I ain't giving you any of my files... However if someone else, next time
    they are sending a few pics to a lab, wants to test it out.... Best thing to
    do is what I have done in the past. Get a nice pic from your camera (about
    9 Meg). Crop it to a 5"x7" at 300dpi then save that as as, say,
    test_300dpi.tif, undo the crop and then crop it to a 5"x7" at 250dpi then
    save that as, say, test_250dpi.tif, undo the crop...... continue on until
    you have 5 files 300dpi, 250dpi, 200dpi, 150dpi, 100dpi. Send them to the
    lab. Get the results back. Realise that 150dpi is

    A while ago I made this comment

    "For what it is worth, I am not 100% sure that I agree with many others on
    here that think you are a troll..... however time will tell and in the
    meantime I will endeavour to answer your last question"

    I have now come to the conclusion that you are a troll who appears to know
    very little about digital imaging and unless you make a massive turn around
    (like you did re: Sigma 120-300 2.8 being just as good as the Canon 300
    2.8L) you can join two others on the 'never to be replied to' list.


    Cameron, Sep 24, 2003
  7. resolution ONLY exists WHEN it's printed. Before that they're only
    pixels, and pixels are all that matter.

    a 1200x600 image, 'saved' at 150dpi or 300dpi still has the same
    number of pixels (1200x600). It's only the instructions to the
    printer that vary.

    So if you save your 1200x600 at 150, and tell the printer to print at
    300, you'll have a 4x2 image.
    if you save your 1200x600 at 300, and tell the printer to print at
    300, you'll have a 4x2 image. They'll be the same image.

    if, however, you tell the printer to print it at 150, you'll end up
    with an 8x4 with less resolution.

    So it's possible you're assumption is correct, and it's possible for
    Cameron & everyone elses assumptions to be correct - first you need to
    understand the terminology and learn how things work, before you go
    slamming people.
    You may find it hard to comprehend, but others do know this stuff and
    have more experience than you do.

    Andrew :eek:)
    Andrew Hennell, Sep 24, 2003
  8. spot on - because you're qualifying the DPI with a physical size.

    5" x 7" at:
    300 = 1500 x 3500
    250 = 1250 x 1750
    200 = 1000 x 1400
    150 = 750 x 1050

    1500x3500 ([email protected]) vs 750x1050 ([email protected])...
    5,250,000 pixels vs 787,500 pixels...
    an 85% decrease in the information available.

    Now if Doug can't see the difference in quality between these two,
    then he needs to think of another hobby :eek:)

    - or -

    Doug's 'saving' the 1500x3500 file at "150dpi", but printing it at 300
    anyway, and is sadly mistaken about what he's actually doing.
    Andrew Hennell, Sep 24, 2003
  9. Joe @ Seaham

    Andrew Mc Guest

    Looking at his site, Ken Duncan can scan at up to 11000dpi, and prints at
    400dpi. So, for a 6x17 tranny and a 30" print, that'd be 6233dpi. So yeah, you'd
    imagine a bit of downsampling would be required... Even at a "low" 4000dpi,
    you'd have 2267dpi.

    This is the site in question: http://www.createdforlife.com/techinfo.htm
    Andrew Mc, Sep 24, 2003
  10. Joe @ Seaham

    Happy Guest

    Hey Doug, One question When you print these two images one at 150 and one at
    300DPI are they printed to the same physical size?

    Happy, Sep 24, 2003
  11. Joe @ Seaham

    Snaps! Guest

    Snaps!, Sep 24, 2003
  12. Joe @ Seaham

    Snaps! Guest

    You seem to have a problem with recollections here Cameron. You obvious have
    some other issues too.

    I never said the Sigma was as good as the Canon. What I said was: "From some
    reports, this big Sigma is quite capable of worrying a 300 Canon prime". DO
    the sunglasses thing before you start claiming I said something someone else
    reported on a US lens test site.

    For 'Andrew Hennel' who seems to think his pubic hair entitles him to claim
    some great experience...
    If you started your career as a photographer after January 1960, you don't
    have more experience than me.
    If you started your career as a photo lab technician after August 1965, you
    don't have any more experience than me either. If you keep talking down to
    people, one day you could find yourself looking up at them.

    Keep in mind too...
    It was Cameron who hijacked this thread and turned it into his own private
    circus. All I did was give the OP some advise I have found to be usefull...
    Who then, is the Troll?

    Snaps!, Sep 24, 2003
  13. Oh Doug, I'm sorry - I didn't realise it was YOU! Oh my god, how DARE
    I even hint at anything other than your word being gospel.

    Here's a hint - you may have taken photo's of Noah's animals, but
    digital technology's just a wee bit newer than that.

    If I talked 'down' to you, then I apologise - I try not to sink to the
    level of those such as you, who defend your wrong claims by attacking

    You're now nothing but an ill-informed, "I was taking photos when you
    were in nappys sonny, don't you tell me what's what" troll.

    Andrew Hennell, Sep 24, 2003
  14. Joe @ Seaham

    Snaps! Guest

    Snaps!, Sep 24, 2003
  15. When you resize a picture from (for example) 100cm x 70 cm at 72 dpi, to
    150mm x 100mm (roughly 6"x4") you can, at the same time, alter the dpi of
    the image which will also alter the pixel dimensions. What others have been
    saying about this is that the pixel dimensions control the final image dpi
    when in fact, it is the dpi plus picture dimensions which control the pixel
    count. You can work entirely with dpi print size without ever referring to
    the pixel dimensions and produce excellent results provided you understand
    your printer's capability.

    It is mostly irrelevant which dimension you work in - pixel size or Dots Per
    Inch (dpi) what is important is that you understand which dimension you
    choose to use. Andrew Hennel suggested I had no comprehension of digital
    photography because my career was (for it's most part) film and chemicals.
    He was wrong in the extreme.

    Computers and software writers have managed to convert the mass of computer
    users to their way of description. The most interesting is in the naming
    conventions and use of dimensions in the print industry which of course
    encompasses much of the Photographic industry too.

    Desktop publishers (DTP) no longer refer to cases of typeface with differing
    fonts. They are all just fonts to non-traditional publishers. Photography
    never was about pixels as a measurable description of size. This is
    something that eventuated from the use of DTP software.

    With the advent of cheap, high quality photo capable printers, pixels are
    even more irrelevant. Pixels are a measurement of a computer screen and a
    digital dimension. The presumption here is that 72 dpi is the resolution a
    computer monitor and everything else will relate to that. 1024x768 pixels at
    72 dpi is about what the average computer user sees on their screen. I know
    of no freely available printer that will print such an image with any degree
    of quality unless it is resized to a higher dpi which reduces the size of
    the image while keeping the number of pixels the same.

    When it comes to printing a photograph ...pixels mean absolutely nothing.
    What is a photograph if it is not a print? So when you resize a file for
    printing or saving and decide to save it as an 8x10, 150dpi image the pixel
    size will conform to those dimensions wether you want it too or not. If you
    later decide to resize the image to 300 dpi, the software used to
    interpolate the image so it's pixel count goes up instead of down, will
    determine the quality of the final print. There are some really good
    interpolation tool about.

    The answer to your question is yes - on my S9000 printer. On an Agfa L2
    printer, yes but the image gets interpolated by the printer's software.
    Whenever you use a printer's control panel to resize an image I.E. "fit to
    page" printing, many printers interpolates the output data. This blows some
    people's theory that pixels rule, right out of the water because the 2
    constants here are dpi and picture measurements. Pixels are an uncontrolled
    Not the Troll again, Sep 24, 2003
  16. Joe @ Seaham

    Andrew Mc Guest

    Surely someone so experienced as yourself would be able to figure this out. I'm
    joining Cameron and Andrew Hennell and not bothering with your trolling from now
    Andrew Mc, Sep 25, 2003
  17. Joe @ Seaham

    Happy Guest

    Unsure what the change of name means Doug but anyway ....

    Thanks for the description Doug but I have been using Photoshop for many
    years and studied a fair bit on it, Some of what you say just does not make
    sense and i asked my question in order to try and understand what you were
    Now i think i know
    ???? See my example at the foot of this post. You have lost me here
    A slide ? :)

    It is becoming clear now, Why would you use software from your printer to
    interpolate your image. This is the last item i would allow to touch my
    file. Either straight from the camera or from Photoshop after i resize using
    Genuine fractuals.

    You are the first person i have met who would use that feature to print an
    image. ( I am talking about people who understand digital and try to sell
    there work )

    Hmmmm I see what you are saying but the prints i have done and have saw do
    not relate what you say ( Maybe i have picked you up wrong ) also i do not
    allow my printer to manipulate my image at all. it prints what PS tells it
    to print.
    True PS was not intended as an out and out Photographers tool. But it HAS
    evolved into a very god tool for a photographer.

    Example to test my understanding of your theory to the extreme.

    If you take a picture on your 1 MB camera and want to print this at A4 size
    at any DPI you want i would bet it looks BAD :) Using you theory it
    should be as good as my D100 at A4 because the only difference is PIXELS. I
    have the same printer ,the same paper and set to the same DPI as you .
    Surely you must see this is not the case and PIXELS are a major requirement
    for any digital print?

    Anyway I gate crashed this thread to try and understand what you were doing.
    Doug i am happy to leave it there. I now see exactly how you go about
    printing your images and that method ( although it will work to a point )
    is definitely
    not for me.

    Happy, Sep 25, 2003
  18. pixels are dots are pixels.
    dpi is how many of those you place per inch on paper.
    pixel numbers / printed size = dpi.
    They are proportional, and intrinsically tied.
    I believe I have already said that. a dpi image is meaningless
    without either pixels or finished size quoted. pixels, on the other
    hand, can exist alone - and for this reason they are the constant
    reference to be used.
    I responded to your assertions that because I had not begun
    photography (or even been born) before you had, then my comments were
    You continually prove otherwise, so I'll stand by my suggestion.
    Photography? you're correct.

    Digital photography? - you're not. Digital files are ALL about
    pixels - that's all there is.

    Ever wondered why camera manufacturers talk about megapixels, and
    pixel sizes of images - and NOT dpi figures? because to a digital
    file (sourced from scan or digital camera) the dpi is irrelevant - it
    matters not until ink (or laser or dye) hits paper.
    A laughable assumption, and completely wrong.
    screens being 72dpi is completely false.
    See http://www.scantips.com/no72dpi.html for more information.
    That's fine, except you're printing a digital file, not a photograph.
    And pixels (or dots) per inch is ALL there is, nothing else.
    You can't be serious, and suggest that interpolation is a substitute
    for having original data?

    How about before you go a) changing your nick, and b) sprouting more
    incorrect information, you do some research.

    A good newbie site for this stuff is http://www.scantips.com/ Come
    back when you've read & comprehended the information there, or when
    you've finished arguing with the website authors about the information
    they present - whichever is first.

    Know that many of us do have experience and knowledge in these
    matters, and be willing to learn. A closed mind learns nothing,

    My only reason for replying is to put your incorrect assumptions to
    rest for others reading. Until you're willing to accept others input,
    and learn, you're nothing but a troll.

    Andrew Hennell, Sep 25, 2003
  19. Let's see -
    The file has a defined, set number of pixels.
    The final print size is variable.
    The final print dpi is variable.

    Allow me to blow your theory right out of the water - I'll remain with
    the constant value, while you base your life on 2 independant

    Cheers idiot,
    Andrew :eek:)
    Andrew Hennell, Sep 25, 2003
  20. Let's say a 1mp image is 1250x800 ( = 1,000,000 pixels)

    A4 = 297x210mm, or 11.69" x 8.26"

    So your final print resolution *regardless of your printer* is:
    1250/11.69 = 106dpi
    800/8.26 = 96dpi

    a 4mp image, of say 2500 x 1600 ( =4,000,000 pixels)
    2500/11.69 = 213dpi
    1600/8.69 = 184dpi

    If we take the first line of this (2500/11.69=213):
    Your printer *can* do, say, 1200dpi - all the printer will do here is
    place 1200 'dots' of ink per inch to make up the 213 pixels you send
    it to put in that inch. ie: the printer uses 5.6 of it's 'dots' to
    make up each of your pixels.

    As I say, pixels are all that matters to the file, and dpi is only a
    function of how many pixels you place over what distance on paper.

    a 8x10 of 150dpi is a 1200x1500 image, regardless of how I print it.
    a 8x10 of 300dpi is a 2400x3000 image, regardless of how I print it.

    if I vary the output dpi, or the output size, the 'dpi' changes - but
    the number of pixels dont.

    It's something that's hard to get your head around to begin with -
    something that took me some time to understand, but since doing so
    I've not had a problem with printing anything here, or via commercial
    photo printers.

    If only there was a way to have imparted this knowledge without the
    likes of the Doug troll. oh well, ce la vie.

    Andrew Hennell, Sep 25, 2003
    1. Advertisements

Ask a Question

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

You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.