Erwin Puts reviews the Leica M8

Discussion in 'Digital Photography' started by Annika1980, Sep 17, 2006.

  1. Annika1980

    Annika1980 Guest

    Erwin Puts, the noted Leica shill, has posted part 1 of his M8 review
    here:
    http://www.imx.nl/photosite/leica/m8report/t006.html

    Since it was first posted yesterday he has taken a few things out of
    his preliminary report. The version I read last night had some
    comparisons between the M8 and the Canon 5D, but those have since been
    removed from the report probably because Leica has asked the reviewers
    not to comment on the actual quality of the photos taken from early
    samples.

    "Yes, Mr. Putz, we'll let you review our new camera just so long as you
    don't tell anybody how bad it is."

    In the report I read last night, Puts said that the M8 and the 5D were
    "about equivalent" in image quality even though his testing showed that
    the 5D was superior. He even noted the 5D's superior high-ISO
    performance as compared to the M8. And of course he claimed that the
    Leica lens he was using, a "2/75" as he called it, was far superior to
    the Canon 24-105 f/4L IS that he was using on the 5D. Yeah, whatever.
     
    Annika1980, Sep 17, 2006
    #1
    1. Advertising

  2. Annika1980

    John Bean Guest

    On 17 Sep 2006 06:45:23 -0700, "Annika1980"
    <> wrote:
    >And of course he claimed that the
    >Leica lens he was using, a "2/75" as he called it, was far superior to
    >the Canon 24-105 f/4L IS that he was using on the 5D.


    Well that part is absolutely true. Any modern Leica prime
    rangefinder lens will give better performance than a zoom.
    *Any* zoom.

    >Yeah, whatever.


    Do some research, open your mind.

    --
    John Bean
     
    John Bean, Sep 17, 2006
    #2
    1. Advertising

  3. "John Bean" <> wrote:
    > On 17 Sep 2006 06:45:23 -0700, "Annika1980"
    > <> wrote:
    >>And of course he claimed that the
    >>Leica lens he was using, a "2/75" as he called it, was far superior to
    >>the Canon 24-105 f/4L IS that he was using on the 5D.

    >
    > Well that part is absolutely true. Any modern Leica prime
    > rangefinder lens will give better performance than a zoom.
    > *Any* zoom.


    I wonder if that's not largely counting angels on heads of pins. The cheap
    Tamron 28-75/2.8 coughs up amazingly sharp images on the 5D. To the point
    that asking for anything better is seriously silly. (Or, more particularly,
    being willing to give up price or weight or the convenience of being a zoom
    for images that might be better on a sensor with 16 times as many pixels but
    are indistibuishable at 12.7MP is seriously silly.)

    David J. Littleboy
    Tokyo, Japan
     
    David J. Littleboy, Sep 17, 2006
    #3
  4. Annika1980

    John Bean Guest

    On Mon, 18 Sep 2006 00:41:23 +0900, "David J. Littleboy"
    <> wrote:

    >
    >"John Bean" <> wrote:
    >> On 17 Sep 2006 06:45:23 -0700, "Annika1980"
    >> <> wrote:
    >>>And of course he claimed that the
    >>>Leica lens he was using, a "2/75" as he called it, was far superior to
    >>>the Canon 24-105 f/4L IS that he was using on the 5D.

    >>
    >> Well that part is absolutely true. Any modern Leica prime
    >> rangefinder lens will give better performance than a zoom.
    >> *Any* zoom.

    >
    >I wonder if that's not largely counting angels on heads of pins. The cheap
    >Tamron 28-75/2.8 coughs up amazingly sharp images on the 5D. To the point
    >that asking for anything better is seriously silly. (Or, more particularly,
    >being willing to give up price or weight or the convenience of being a zoom
    >for images that might be better on a sensor with 16 times as many pixels but
    >are indistibuishable at 12.7MP is seriously silly.)


    I don't disagree that there's probably no advantage to the
    optical superiority of the Leica lens in this situation, but
    that doesn't alter the reality that it *is* optically
    superior... which only a Canon fanboy like Brett could
    dispute.

    A lens' optical perfomance is nothing to do with
    convenience, nor on the imaging medium - as well you know -
    and my comment was about lens performance, nothing more.


    --
    John Bean
     
    John Bean, Sep 17, 2006
    #4
  5. "Annika1980" <> wrote in message
    news:...
    > Erwin Puts, the noted Leica shill, has posted part 1 of his M8
    > review
    > here:
    > http://www.imx.nl/photosite/leica/m8report/t006.html
    >
    > Since it was first posted yesterday he has taken a few things out of
    > his preliminary report.


    That might explain why I couldn't access that review when I tried.

    > The version I read last night had some comparisons between the
    > M8 and the Canon 5D, but those have since been removed from
    > the report probably because Leica has asked the reviewers not
    > to comment on the actual quality of the photos taken from early
    > samples.


    Which is of no importance to real Leica shills, the logo matters. The
    most important feature for photography, the image quality, remains
    unknown. I do wonder how much of the lack of AA-filter they were able
    to disguise in postprocessing, especially with sharp Leica lenses.

    --
    Bart
     
    Bart van der Wolf, Sep 17, 2006
    #5
  6. Annika1980

    Annika1980 Guest

    Bart van der Wolf wrote:
    >
    > > The version I read last night had some comparisons between the
    > > M8 and the Canon 5D, but those have since been removed from
    > > the report probably because Leica has asked the reviewers not
    > > to comment on the actual quality of the photos taken from early
    > > samples.

    >
    > Which is of no importance to real Leica shills, the logo matters. The
    > most important feature for photography, the image quality, remains
    > unknown. I do wonder how much of the lack of AA-filter they were able
    > to disguise in postprocessing, especially with sharp Leica lenses.


    >From what I can remember from the original review, Puts claimed that

    the superiority of the Leica lenses didn't really reveal itself in his
    tests. I took that to mean that the M8 was so bad that it couldn't
    really test the lens to it's capabilities. Contrast that with the top
    Canon bodies like the 5D or the 1DsMKII, which are known for their
    abilities to reveal even the slightest flaw in the lens.

    One wonders why Leica would release a product that obviously isn't
    ready for "prime" time.
     
    Annika1980, Sep 17, 2006
    #6
  7. "Annika1980" <> wrote:
    > Bart van der Wolf wrote:
    >>
    >> > The version I read last night had some comparisons between the
    >> > M8 and the Canon 5D, but those have since been removed from
    >> > the report probably because Leica has asked the reviewers not
    >> > to comment on the actual quality of the photos taken from early
    >> > samples.

    >>
    >> Which is of no importance to real Leica shills, the logo matters. The
    >> most important feature for photography, the image quality, remains
    >> unknown. I do wonder how much of the lack of AA-filter they were able
    >> to disguise in postprocessing, especially with sharp Leica lenses.

    >
    >>From what I can remember from the original review, Puts claimed that

    > the superiority of the Leica lenses didn't really reveal itself in his
    > tests. I took that to mean that the M8 was so bad that it couldn't
    > really test the lens to it's capabilities. Contrast that with the top
    > Canon bodies like the 5D or the 1DsMKII, which are known for their
    > abilities to reveal even the slightest flaw in the lens.
    >
    > One wonders why Leica would release a product that obviously isn't
    > ready for "prime" time.


    Sheesh, what a bunch of obnoxious negativity. You guys sound like me!

    By the way, in real life, the 5D is the ultimate cheap glass camera; the fat
    pixels mean that even cheap consumer zooms produce great images if you stop
    them down a bit to minimize vignetting (which is really obnoxious wide open
    on the plastic fantastic (US$100 used) Canon 55-200). It's only fairly
    extreme wide angle lenses that have problems in the corners.

    David J. Littleboy
    Tokyo, Japan
     
    David J. Littleboy, Sep 17, 2006
    #7
  8. "Annika1980" <> wrote in message
    news:...
    SNIP
    > The version I read last night had some comparisons between the
    > M8 and the Canon 5D, but those have since been removed from
    > the report [...]


    Of course Google cached the page with images ;-) :
    Search for <http://www.imx.nl/photosite/leica/m8report/t006.html> ,
    without the angled brackets, and select the "Cached" page. Not
    surprising mostly Black and White images (which hides False Color
    moire), and False Color moire on the color ones.

    The vignetting seems a bit decentered, the angled microlenses do add
    another complication to the manufacturing process.

    Resolution, a bit overstated for both due to the type of target used,
    is 60 cy/mm on sensor for the Canon, and 70 cy/mm for the Leica (which
    reduces to 53 cy/mm if output magnification differences are
    equalized). The lenses used differ, as does the postprocessing, so
    there may be other differences caused by that.

    My prediction for the DPreview resolution test based on these images
    would be: 5D = 2356 LPH, and M8 = 2062 LPH, after correction for
    output size differences and target type used.

    --
    Bart
     
    Bart van der Wolf, Sep 17, 2006
    #8
  9. "Bart van der Wolf" <> wrote:
    >
    > Of course Google cached the page with images ;-) :
    > Search for <http://www.imx.nl/photosite/leica/m8report/t006.html> ,
    > without the angled brackets, and select the "Cached" page. Not surprising
    > mostly Black and White images (which hides False Color moire), and False
    > Color moire on the color ones.


    http://vgrin.front.ru/M8/

    > The vignetting seems a bit decentered, the angled microlenses do add
    > another complication to the manufacturing process.
    >
    > Resolution, a bit overstated for both due to the type of target used, is
    > 60 cy/mm on sensor for the Canon, and 70 cy/mm for the Leica (which
    > reduces to 53 cy/mm if output magnification differences are equalized).
    > The lenses used differ, as does the postprocessing, so there may be other
    > differences caused by that.


    I was surprised that the sweater image looks so bad. Also interesting was
    how well the 5D handles that star chart (no surprises about how badly the M8
    does, though). Of course, that star chart is a square wave pattern, not a
    sine wave pattern, so is problematic.

    > My prediction for the DPreview resolution test based on these images would
    > be: 5D = 2356 LPH, and M8 = 2062 LPH, after correction for output size
    > differences and target type used.


    FWIW, Dpreview calls the 5D at 2000 lph. Dpreview resolution numbers seem
    totally random. For starters, the lpw should be 1.5x larger than lph for 2:3
    cameras and 1.333x larger for 3:4 cameras, but it never is...

    David J. Littleboy
    Tokyo, Japan
     
    David J. Littleboy, Sep 17, 2006
    #9
  10. "David J. Littleboy" <> wrote in message
    news:eekgv8$6tk$...
    > "Bart van der Wolf" <> wrote:

    SNIP
    > I was surprised that the sweater image looks so bad. Also
    > interesting was how well the 5D handles that star chart (no
    > surprises about how badly the M8 does, though). Of course, that star
    > chart is a square wave pattern, not a sine wave pattern, so is
    > problematic.


    Yes, such a square wave target overstates the sinusoidal wave (cycle)
    resolution by a factor of 4/Pi, so I corrected for that. Since unknown
    postprocessing in CS2 was applied it complicates a direct comparison.
    Maybe the sweater was Red in the full color version. The brick example
    on the other hand looks blocky at the given screen resolution. The
    Siemens star target is also less critical (fewer sectors) than the
    original Siemens star, I'm not sure if that was on purpose to reduce
    the False Color artifacts.

    > FWIW, Dpreview calls the 5D at 2000 lph. Dpreview resolution numbers
    > seem totally random. For starters, the lpw should be 1.5x larger
    > than lph for 2:3 cameras and 1.333x larger for 3:4 cameras, but it
    > never is...


    That is partly because the cut-off point is somewhat arbitrary and the
    target is not really suited (nor intended) for quantification.
    Different sampling apertures in Hor/Ver dimensions can also account
    for resolution differences.

    The 9-line hyperbole targets are meant to be visually compared. Since
    they are also square wave patterns, (mis-)alignment with the sensel
    array will change the results. That's why I suggest a sinusoidal star
    target (for limiting resolution), or a slanted edge target (for 4x
    sub-sampled sensel MTFs), for more robust measurements.

    --
    Bart
     
    Bart van der Wolf, Sep 17, 2006
    #10
  11. Annika1980

    Doug Robbins Guest

    This is the equivalent of using 21st century technology to build a very
    sophisticated robot horse to pull your buggy around.

    Doug


    "Annika1980" <> wrote in message
    news:...
    > Erwin Puts, the noted Leica shill, has posted part 1 of his M8 review
    > here:
    > http://www.imx.nl/photosite/leica/m8report/t006.html
    >
    > Since it was first posted yesterday he has taken a few things out of
    > his preliminary report. The version I read last night had some
    > comparisons between the M8 and the Canon 5D, but those have since been
    > removed from the report probably because Leica has asked the reviewers
    > not to comment on the actual quality of the photos taken from early
    > samples.
    >
    > "Yes, Mr. Putz, we'll let you review our new camera just so long as you
    > don't tell anybody how bad it is."
    >
    > In the report I read last night, Puts said that the M8 and the 5D were
    > "about equivalent" in image quality even though his testing showed that
    > the 5D was superior. He even noted the 5D's superior high-ISO
    > performance as compared to the M8. And of course he claimed that the
    > Leica lens he was using, a "2/75" as he called it, was far superior to
    > the Canon 24-105 f/4L IS that he was using on the 5D. Yeah, whatever.
    >
     
    Doug Robbins, Sep 18, 2006
    #11
  12. Annika1980

    jeremy Guest

    "Doug Robbins" <> wrote in message
    news:lvwPg.12566$...
    > This is the equivalent of using 21st century technology to build a very



    Well-said!
     
    jeremy, Sep 18, 2006
    #12
  13. Annika1980

    tomm42 Guest

    Doug Robbins wrote:
    > This is the equivalent of using 21st century technology to build a very
    > sophisticated robot horse to pull your buggy around.
    >
    > Doug
    >


    Since I assume you have never used an M series Leica, that is a fairly
    harsh comment. Leica makes some of the best 35mm lens made - ever. An M
    series camera are just about the smoothest operating camera ever
    designed. Until you use one you won't get it.
    I couldn't find images everyone was talking about, does anyone have
    the address for the images? Shame that they didn't put an AA filter on
    the camera, going for max resolution ala Kodak with their DCS cameras
    and (ugh) Sigma with the foveon sensor. Kodak at least had an
    anti-moire setting in their Photodesk software.

    Tom
     
    tomm42, Sep 18, 2006
    #13
  14. "David J. Littleboy" <> writes:

    >FWIW, Dpreview calls the 5D at 2000 lph. Dpreview resolution numbers seem
    >totally random. For starters, the lpw should be 1.5x larger than lph for 2:3
    >cameras and 1.333x larger for 3:4 cameras, but it never is...


    No, they shouldn't. Those resolutions are in lines per picture
    *height*, no matter whether horizontal or vertical resolution is
    being measured. So the horizontal and vertical resolution should match
    if the sensor has square pixels, no matter what its aspect ratio.

    On the other hand, if you want to know how many vertical lines can be
    resolved across the width of your image, you need to convert lines per
    picture height into lines per picture width by multiplying by 3/2 or
    4/3, whatever the image aspect ratio.

    But the horizontal resolution numbers on the chart and in the data
    tables are not lines per picture width.

    Dave
     
    Dave Martindale, Sep 19, 2006
    #14
  15. "tomm42" <> writes:

    > I couldn't find images everyone was talking about, does anyone have
    >the address for the images? Shame that they didn't put an AA filter on
    >the camera, going for max resolution ala Kodak with their DCS cameras
    >and (ugh) Sigma with the foveon sensor. Kodak at least had an
    >anti-moire setting in their Photodesk software.


    At least some of the Kodak cameras had a removable AA filter, so you
    could choose to use it or not depending on subject material.

    Dave
     
    Dave Martindale, Sep 19, 2006
    #15
  16. "Dave Martindale" <> wrote:
    > "David J. Littleboy" <> writes:
    >
    >>FWIW, Dpreview calls the 5D at 2000 lph. Dpreview resolution numbers seem
    >>totally random. For starters, the lpw should be 1.5x larger than lph for
    >>2:3
    >>cameras and 1.333x larger for 3:4 cameras, but it never is...

    >
    > No, they shouldn't. Those resolutions are in lines per picture
    > *height*, no matter whether horizontal or vertical resolution is
    > being measured. So the horizontal and vertical resolution should match
    > if the sensor has square pixels, no matter what its aspect ratio.


    OK. But since they never match (they're always larger in the horizontal
    direction), there's still something fishy going on.

    David J. Littleboy
    Tokyo, Japan
     
    David J. Littleboy, Sep 20, 2006
    #16
  17. "David J. Littleboy" <> writes:

    >> No, they shouldn't. Those resolutions are in lines per picture
    >> *height*, no matter whether horizontal or vertical resolution is
    >> being measured. So the horizontal and vertical resolution should match
    >> if the sensor has square pixels, no matter what its aspect ratio.


    >OK. But since they never match (they're always larger in the horizontal
    >direction), there's still something fishy going on.


    Yeah. Take dpreview's test of the 30D as an example:
    http://www.dpreview.com/reviews/canoneos30d/page28.asp

    The text says it resolves 1850 lph horizontal and 1650 lph vertical. But
    if you look at the chart, the vertical resolution (measured from the
    horizontal lines) is clearly greater than the horizontal resolution
    (measured with the vertical lines). Perhaps Phil is swapping labels, so
    "vertical" means vertical lines, which measure horizontal resolution,
    and vice versa. His numbers certainly don't match what my eyes see.

    As for why they are different, I don't know. It could be that the
    vertical resolution target (horizontal lines) are arranged radially
    outward from the centre of the target, while the horizontal resolution
    target is off to one side. If the lens has any aberrations that cause
    the image spot to be stretched radially into an ellipse rather than a
    disc, the vertical target is positioned where it wouldn't be affected
    while the horizontal target would. But that's just a wild guess.

    Dave
     
    Dave Martindale, Sep 20, 2006
    #17
  18. Annika1980

    mark Guest

    David J. Littleboy schreef:

    > "John Bean" <> wrote:
    > > On 17 Sep 2006 06:45:23 -0700, "Annika1980"
    > > <> wrote:
    > >>And of course he claimed that the
    > >>Leica lens he was using, a "2/75" as he called it, was far superior to
    > >>the Canon 24-105 f/4L IS that he was using on the 5D.

    > >
    > > Well that part is absolutely true. Any modern Leica prime
    > > rangefinder lens will give better performance than a zoom.
    > > *Any* zoom.

    >
    > I wonder if that's not largely counting angels on heads of pins. The cheap
    > Tamron 28-75/2.8 coughs up amazingly sharp images on the 5D. To the point
    > that asking for anything better is seriously silly. (Or, more particularly,
    > being willing to give up price or weight or the convenience of being a zoom
    > for images that might be better on a sensor with 16 times as many pixels but
    > are indistibuishable at 12.7MP is seriously silly.)
    >


    hahahahahaha

    > David J. Littleboy
    > Tokyo, Japan
     
    mark, Sep 20, 2006
    #18
  19. Annika1980

    mark Guest

    David J. Littleboy schreef:

    > "Annika1980" <> wrote:
    > > Bart van der Wolf wrote:
    > >>
    > >> > The version I read last night had some comparisons between the
    > >> > M8 and the Canon 5D, but those have since been removed from
    > >> > the report probably because Leica has asked the reviewers not
    > >> > to comment on the actual quality of the photos taken from early
    > >> > samples.
    > >>
    > >> Which is of no importance to real Leica shills, the logo matters. The
    > >> most important feature for photography, the image quality, remains
    > >> unknown. I do wonder how much of the lack of AA-filter they were able
    > >> to disguise in postprocessing, especially with sharp Leica lenses.

    > >
    > >>From what I can remember from the original review, Puts claimed that

    > > the superiority of the Leica lenses didn't really reveal itself in his
    > > tests. I took that to mean that the M8 was so bad that it couldn't
    > > really test the lens to it's capabilities. Contrast that with the top
    > > Canon bodies like the 5D or the 1DsMKII, which are known for their
    > > abilities to reveal even the slightest flaw in the lens.
    > >
    > > One wonders why Leica would release a product that obviously isn't
    > > ready for "prime" time.

    >
    > Sheesh, what a bunch of obnoxious negativity. You guys sound like me!
    >
    > By the way, in real life, the 5D is the ultimate cheap glass camera; the fat
    > pixels mean that even cheap consumer zooms produce great images if you stop


    hahahahahaha

    > them down a bit to minimize vignetting (which is really obnoxious wide open
    > on the plastic fantastic (US$100 used) Canon 55-200). It's only fairly
    > extreme wide angle lenses that have problems in the corners.
    >
    > David J. Littleboy
    > Tokyo, Japan
     
    mark, Sep 20, 2006
    #19
  20. mark wrote:
    >
    > hahahahahaha


    Now, this is the sort of intelligent exposition most of us come here for.

    Please, say it in German rather than this baby talk.

    --
    john mcwilliams
     
    John McWilliams, Sep 20, 2006
    #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. P$YCH0 78

    New power supply puts out a funny smell...

    P$YCH0 78, Mar 4, 2005, in forum: Computer Support
    Replies:
    21
    Views:
    2,145
    Wizard
    Mar 5, 2005
  2. Ola Theander

    Prblm: Nikon Coolpix 5700 puts focus on wrong object.

    Ola Theander, Aug 17, 2003, in forum: Digital Photography
    Replies:
    3
    Views:
    525
    Ed Ruf
    Aug 17, 2003
  3. Philip Homburg

    Erwin Puts on AA filters (in his M8 review)

    Philip Homburg, Nov 14, 2006, in forum: Digital Photography
    Replies:
    6
    Views:
    421
    Philip Homburg
    Nov 17, 2006
  4. John Navas

    Is Lumix Leica real Leica?

    John Navas, Nov 17, 2007, in forum: Digital Photography
    Replies:
    1
    Views:
    669
    Dennis Pogson
    Nov 18, 2007
  5. TJ
    Replies:
    13
    Views:
    1,803
    Tony Polson
    Dec 23, 2007
Loading...

Share This Page