M9- full frame, thought it was impossible?

[kelly_flanigan1]

<p>The 15mm VC LTM lens I have here from when they first came out is not a retrofocus lens; it goes deep inside the lens mount; off axis rays hit the film/sensor at a steep angle. One can measure it and see it using wax paper.<br>

<br /> With the old 16mm F3.5 Nikkor; its a radically different lens; way larger with alot of clearance for the Nikon F mirror. Off axis rays that hit the corner are WAY less steep compared the the 15mm VC lens.<br>

<br /> One take a piece of wax paper and see the image of a far off axis say lightbulb's image and see the steep angle of the 15mm VC versus the 16mm. Its not a subtle thing; but way way different. If you want to say the 15mm is a retro focus; then maybe it is mildly one; and the 16mm is a radical one; like comparing the weight of a mouse to an elephant.<br>

<br /> Or maybe the take is that the 15mm is redesigned?<br>

<br /> Here with my 1970's 16mm Nikkor for the Nikon F it is a retrofocus; its design is radically different than my 15mm VC lens; it which mine appears not to be retrofocus by actual experiments.<br>

One can place a small sensor at the corner of a 24x36mm frame and the light that hits it from the 16mm retrofocus lens is more normal than the 15mm's in a radical way; since one is way more retrofocus in design than another.<br>

<br /> Wide angle lenses for the Nikon F are retrofocus to clear the swinging instant return mirror.(non mirror lockup modern types)</p>

 

[andy_piper2]

<p>Kelly F. - lens can be more or less retrofocus. It's a continuum. If the lens' exit pupil is more or less the same distance from the image plane (film, sensor) as the focal length, it is not RF. If the exit pupil (not necessarily the rear element) is further from the image plane than the effective focal length, then it IS retrofocus.</p>

<p>An SLR 15mm is MORE retrofocus than the C/V 15, in order to clear the mirror, but the C/V is still RF itself, too.</p>

<p>Here are the C/V superwides, including the optical diagrams: http://www.cosina.co.jp/seihin/voigt/english/s-wide-e.html</p>

<p>Retrofocus lenses (the 15, the 12) tend to be asymmetrical - shaped like a funnel from outer end to inner end. Non-retrofocus lenses (21/25) tend to be much more symmetrical, funnels coming in from both side about equally.</p>

<p>Here's a link showing Nikkor and Zeiss 21s which are also non-retrofocus, and again the symmetrical design is the clue, if you don't have the precise exit pupil measurement. http://www.mir.com.my/rb/photography/companies/nikon/nikkoresources/RF-Nikkor/RF21mm/index.htm</p>

<p>By contrast, here is the layout of a Zeiss Distagon, which is definitely retrofocus, and has the same funnel shape as the C/V 15 and 12: http://ogiroux.files.wordpress.com/2009/07/21linsenschnitt.jpg More extreme, since its for SLR use, and the barrel construction is obviously different from the C/Vs.</p>

[andy_piper2]

<p>Hmm- can't edit. Disregard reference to 25 c/v in above. You are right that the c/v 15 optics did not change with the new barrel design, though, as the diagrams show.</p>

[leif_goodwin8]

<p><a href="../photodb/user?user_id=40204">Andy Piper</a> said: "What Leica originally said (over 5 years ago) was that a digital M camera was impossible with current sensor architecture. "<br>

As someone else indicated, had they stated that they would bring out an M9 with a full frame and integral IR filter in a few years time, I wonder if the M8 would have sold? Or would many people have 'struggled' along with a cheaper alternative until the M9 appeared?</p>

[andy_piper2]

<p>Three years wait? No.</p>

<p>I did "struggle along" with a Sony R-1 for one year (not really the right term - the Sony was nice, and cheaper than getting into an SLR system for a year), on the assumption that the M8 would arrive at PhotoKina 2006 (which it did).</p>

<p>I don't like having a screen (groundglass or EVF) between me and my subjects, and won't do it if there is an alternative - so getting the M8 as soon as it was available was an easy choice, even though I assumed Leica would eventually get to no-IR/full-frame. It looks like Leica (maybe) sped up the M9 development by a year in order to get cash now - leaving out some improvements based on S2 technology that may show up in the next model.</p>

<p>But my philosophy is that a digital rangefinder in the hand today is always better than pie-in-the-sky 1-2-3-? years from now.</p>

[joseph_wisniewski]

<p>Matt, I can see why one might want an exit pupil so close to the sensor on a fast lens, in order to avoid vignetting. So a 50mm f1.0 ends up being built like a telephoto.</p>

<p>Andy, thank you. That's nailed it. Rangefinder lenses that are retrofocsu are typically "a lot less retrofocus" than comparable SLR lenses. People always confuse the rear element location with the lens being retrofocus. I've used some high end industrial lenses that were close to image side telecentric (chief rays parallel to the sensor) but had rear elements almost toughing the sensor. This si also very common in P&S lenses.</p>

<p>Not too surprised you'd have to dial in a bit of compensation for a 135. That just means that they've got the microlenses canted in enough so that really parallel light is on the "wrong side" of optimum. And that the beamwidth is a little narrower than I'd have hoped.</p>

[geoffs1]

<blockquote>

<p ><a href="../photodb/user?user_id=2347092">Arthur Plumpton</a> <a href="../member-status-icons"><img title="Subscriber" src="http://static.photo.net/v3graphics/member-status-icons/sub3.gif" alt="" title="Subscriber" /> <img title="Frequent poster" src="http://static.photo.net/v3graphics/member-status-icons/3rolls.gif" alt="" title="Frequent poster" /> </a> , Sep 12, 2009; 08:40 a.m.<br>

... compromises that must be made with sensor and microlens design that are virtually absent with film use. ...</p>

</blockquote>

<p>Not 100% true...<br>

From the <a href="http://www.gigapxl.org/technology-realized.htm">Gigapixel Project's technology page</a> , "Optical Filters" section:</p>

<blockquote>

<p>Ultra-wide-angle cameras invariably suffer from a fall-off in film exposure as one moves off axis. This effect is due to three separate phenomena, all of which are geometric in nature. First, as one moves off axis, the apparent shape of the pupil within the lens becomes increasingly elliptical; i.e. its projected dimension in one axis becomes smaller. Secondly, the distance from lens to film increases as the field angle increases; thereby causing the effective f-number to increase. Finally, as the field angle increases, so does the obliquity of light reaching the film surface. This causes the light to spread over an ever-larger area of film and, in so doing, causes that area to receive less exposure. At 45-degrees off axis, the collective effect of these phenomena can amount to the equivalent of two stops in exposure. In the Asymmagon™, the first effect is offset somewhat by the deliberate introduction of pupil distortion. Nevertheless, the fall-off at 49.9 degrees (corresponding to the extreme corners of a 9"x18" format) still approximates two full stops.</p>

</blockquote>

[aplumpton]

<p>If I read your refernce closely enough, Geoff, they are simply indicating what causes vignetting, or light fall-off towards the frame edges and corners. This is optical and independent of the light capture mechanism, whether film or digital camera sensor. Correcting vignetting is probably easier in digital, through software, whereas extreme cases of strongly vignetting lenses for film use require a center filter (with the price of losing the advantage of a fast lens).</p>

<p> </p>

[kelly_flanigan1]

<p>Andy and Joseph; here I have used the 16mm F3.5 Nikkor and the 15mm VC on an optical bench settup to fool around with digital sensors off axis. The sensor mule(s) are an old cameras that of course hold LTM, M or F lenses. I have M/LTM to Nikon F adapter that allows infinity focus.<br /> <br /> When one has a sensor or piece of one at the corner of a 24x36mm film plane; the light the sensor or film sees is at a WAY different angle of incidence with the 15mm compared to the 16mm; not just due to the focal length.<br /> <br /> Thus I can conclude that the 15mm VC is a radically different bird than the retrofocus 16mm F3.5.<br /> <br /> With a sensor has issues with off axis rays; the 16mm has less issues than the 15mm. With the 15mm the light hits the corner at a FAR greater off normality than the 16mm; since the 15mm is either non retro' or a mild one.<br /> <br /> In no way is the 15mm like the retrofocus 16mm.<br /> <br /> One can trace rays with a piece of wax paper held so on has a shining light hitting the corner of the 24x36mm frame; the 16mm is like the noon sun compared the 15mm VC.<br /> <br /> The lay public thinks retrofocus is like "FAT" with no sense of how fat.<br /> <br /> One can have a person 10 Lbs overweight; or 200 Lbs overweight.<br /> <br /> A strong retrofocus lens makes the far corner light rays hit the sensor/film more closer to nomal compared to a lessor or non retorfocus lens.<br>

This non normal issue is an old one; well known for several decades.<br>

<br /> The 15mm VC I bought long ago was not even available in the USA yet; I bought it in Hong Kong; its description then was a non retrofocus lens.<br>

What does NON mean in a translation to Engish in a photo advert; it it was fat we Americans have pump; husky; beefy; farm animal; obese; grossly obese; so obese they cannot walk; or is it like a models feeling fat? :)</p>

[geoffs1]

<blockquote>

<p ><a href="../photodb/user?user_id=2347092">Arthur Plumpton</a> <a href="../member-status-icons"><img title="Subscriber" src="http://static.photo.net/v3graphics/member-status-icons/sub3.gif" alt="" title="Subscriber" /> <img title="Frequent poster" src="http://static.photo.net/v3graphics/member-status-icons/3rolls.gif" alt="" title="Frequent poster" /> </a> , Sep 15, 2009; 01:28 p.m.<br>

If I read your refernce closely enough, Geoff, they are simply indicating what causes vignetting, or light fall-off towards the frame edges and corners. This is optical and independent of the light capture mechanism, whether film or digital camera sensor.</p>

</blockquote>

<p>My (intended) point was exactly that: remind people that the non-perpendicular angle of rays at the edge of the frame causes (sometimes substantial) vignetting, even with film. </p>

<p>In fact, it seems that angling the microlenses at the edge of the frame so they are (more nearly) perpendicular to the ray striking them can actually reduce some of the geometric vignetting by reducing the third effect, "... as the field angle increases, so does the obliquity of light reaching the film surface. This causes the light to spread over an ever-larger area of film."</p>