View Full Version : Solms has spoken:
Some more tid-bits of information. The USA market will have priority in the first deliveries of the M8, both in number of cameras and different types of camera's [black vs chrome??]. The 3.4 apo 135 cannot be used at all as the focal point is not on the sensor :confused: :mad: :confused: .
I hope the word could spread at light speed so I can grab a Apo135 for 50 bucks.
Mark Norton
06-08-2006, 09:06
Where did these gems come from? Don't tell me! They've only ordered US power cords for the charger and ones for Europe, UK and Oz are on a 2 year lead time!
If the lens will not focus onto the sensor, how did it ever focus on film?
I thought they had said it would be black only to start with. Maybe the variants refer to finder magnifications.
The 3.4 apo 135 cannot be used at all as the focal point is not on the sensor :confused: :mad: :confused: .
All Leica lenses have to be calibrated to the same register. There's no way that lens could focus on the film plane of an M-film body but not on the sensor plane of the M-digital body.
What Jaap has posted is true ... after reading his post, I called my dealer in Tokyo and they already have the official words from Leica Japan.
Something must be getting lost in translation, or someone at Leica decided to have a little fun with the speculators on internet forums :D Having some lenses that were able to focus on film but not on the digital sensor is physically impossible.
Mark Norton
06-08-2006, 09:28
The only problem surely is going to be focussing the thing which we all expected anyway. I agree with Ben Z, the scope of the message getting lost in the translation is very great. It's not a lens for me anyway. 90mm is just fine.
To be precise, what I heard was that because of the 1.33x cropping factor, the picture angle would be too narrow so you can't effectlvely focus at all.
That's just as nonsensical to me. If right now I can focus my 135 accurately enough that I can crop the negative to the effective angle of a 180mm lens (which I can and have, many times), then there's no reason it matters whether I crop it after the fact or the camera crops it for me due to a smaller area of capture. In fact, if Leica has constructed the viewfinder so that the framelines brought up by the 28mm lens occupies the same outermost position as it does on a .72x film body, they would actually have needed to increase the magnification of the optics so that it would encompass the narrower angle of view of a 37mm lens, thus making the effective baselength longer and therefore more capable of focusing accurately! And, 135mm framelines of the same apparent size as the current ones in the .72 finder would automatically show the coverage of a 180mm lens. So I'm completely confused as to why the 135mm is being blacklisted by Leica.
Bob Ross
06-08-2006, 12:53
To be precise, what I heard was that because of the 1.33x cropping factor, the picture angle would be too narrow so you can't effectlvely focus at all.
This may be the case. The 135mm has a FOV of a 180mm and in the finder this may be smaller than the focus patch. Depending on the finder magnification factor, it could be near impossible to see the focus difference. The only rumor that I have read about the finder magnification is that it would be .68, which I haven't quite digested yet as to how small the field of a 180mm eqv would be. The 135mm should at least work at infinity :)
If the finder mag is even .58, let alone .68, and the eyepiece threads are the same as the current M bodies and will accept the 1.25x magnifier, that would increase the finder to .72, which is currently adequate for the 135/3.4 and /4.
To focus accurately a 135mm lens at f/3.4, the effective base length of the rangefinder should be more than 70mm for a finder magnification of 0.72x due to the smaller size of the circle of confusion caused by the crop factor.
Best,
LCT
Given the crop factor a magnification of 0.85x makes more sense to me.
Bob Ross
06-08-2006, 17:47
Given the crop factor a magnification of 0.85x makes more sense to me.
I agree with you on the .85x. I wonder what kind of feedback they are getting on the viewfinder from their beta testers? There has to be a few .85x types amonst them.
the crop factor is not going to make people want a .85x viewfinder when they like .72x.
0.85x with a crop factor of 1.33 is equivalent (for lens framig prpose) to 0.64 on full frame, assuming that the physical size of the viewfinder is left unchanged.
The RD-1 has a 1:1 viewfinder with frames for 28, 35 and 50mm lenses, but that has a 1.5 crop factor, equivalent to 0.666.
So a 0.58 will probably have brightlines for 21mm but the brightline for 50mm is going to be really small, and a 75mm will become extremely difficult to use. This could explain why a 135mm lens is considered unusable with the M8.
If this is true I have yet another reason to hold on my RD-1 for a while.
All Leica lenses have to be calibrated to the same register. There's no way that lens could focus on the film plane of an M-film body but not on the sensor plane of the M-digital body.
I agree, See my :confused: :confused: :confused: :confused: :confused:
you're not making sense. the crop factor has no effect on viewfinder magnification.
imagine you're using a 35mm lens on a .72x film body. when you get an m8, you want the same thing, so you'll use a 28mm lens on a .72x body. get it?
you're not making sense. the crop factor has no effect on viewfinder magnification.
imagine you're using a 35mm lens on a .72x film body. when you get an m8, you want the same thing, so you'll use a 28mm lens on a .72x body. get it?
The crop factor together with the viewfinder magnification, has an effect on which brightlines can be shown in the viewfinder, i know that with the 1.33 cropfactor 21mm have the same field of view as 28mm, however since the depth of field of a 21mm cropped by 1.33 is greater than the depth of field of a 28mm, i'd rather have useful brightlines for the standard 28, 35 and 50mm lenses (like in my RD-1) and use an external viewfinder for a 21mm (guess focusing is soo much easier on 21mm than on 50mm) than having a brightline for 21mm and a too small one for 50mm.
That's why Epson's choice makes more sense to me, also a 1x viewfinder helps framing quite a bit (I love to be able to keep both my eyes open)
Essentialy so far the only thing the M8 seems to do significantly better than the RD-1 is the wider rangefinder base, and this, to me, does not justify upgrading.
Of course your priorities may be different.
That's why Epson's choice makes more sense to me, also a 1x viewfinder helps framing quite a bit (I love to be able to keep both my eyes open)
We don't even know yet which choice Leica has made, Francesco.
you wouldn't feel the same way if the r-d1 had a lower crop factor. plus you're counting on people going through the hassle of external viewfinders for things as mundane as a 28mm fov, and even a 35mm fov if you're wearing glasses on a .85x body.
.72x ain't king for no reason.
We don't even know yet which choice Leica has made, Francesco.
You are right of course, point taken.
But there is some fun in speculating :D
The only problem surely is going to be focussing the thing which we all expected anyway. I agree with Ben Z, the scope of the message getting lost in the translation is very great. It's not a lens for me anyway. 90mm is just fine.
I doubt the Dutch importer would have much difficulty translating German. However, I cannot see how that could be either, but then the optical science behind building diffraction limited lenses is esoteric to say the least.
I don't think we need to be scientists to understand that all existing Leica lenses were designed to focus on the same plane, the same distance from the body flange. So they will all focus on the same plane, the same distance from the body flange in the M8. Just as any lens that could focus on a Nikon F or Canon EOS body focuses on the plane of their digital sensors.
And even if for the sake of argument we suspend common knowledge and sense and allow that the 135 APO doesn't focus on the M8 plane, it would be a simple matter to have the rangefinder cam adjusted so that it does. That would mean it wouldn't focus correctly on a film body, but nonetheless it would be usable on the M8.
As I said, I can't believe this either,I won't be selling mine, and I will be posting whatever (I think great!) results I get, whenever I have the M8 in my hands.....
...nonetheless it would be usable on the M8.
Yes but not at f/3.4 with a .72x mag VF is the effective base length of the rangefinder is (most probably) shorter than 70mm.
Best,
LCT
And the 1.25 Okular? Let's hope that fits!
With the same RF and a 1.25 magnifier the effective base length would be 69.25 (mechanical base length) * 0.72 * 1.25 = 62.33 mm.
Best,
LCT
Well, at least it is 25 % better.....
Since the old Elmarit-M 135mm f/2.8 (11829) with goggles is quoted as compatible with the M8 D [iesel ;)] the view/range-finder system has probably not changed much. My bet is on a x.85 version at first.
I would love to handle the new Apo beast, though, because I cannot imagine better wide-open performance than my 'cheap' Tele-Elmarit...
Yes but not at f/3.4 with a .72x mag VF is the effective base length of the rangefinder is (most probably) shorter than 70mm.
Best,
LCT
The physical baselength of the M8 rangefinder is the same as the film M bodies, this we know for a fact since the googled lenses will fit. So the adjustment for the cropped fov must be controlled by the finder magnification. If the outermost frameline is the one activated by a 28mm lens, then the finder magnification would be increased so those framlines would frame the correct, cropped view. However if the finder is to have a built-in frameline at the outermost border for the 21mm lens (28mm fov) then they could use the same .72 magnification and simply put in a corrected mask set. Either way there is no reason the cropped sensor necessitates demagnifying to the point that a 135mm lens can't be reliably focused. Leica's only reason for doing so might be to increase the eyepoint of the finder, such as the .58 does.
Mark Norton
06-09-2006, 11:54
Keep in mind that the bayonet lug of the 21mm currently activates the 28 and 90 framelines and if we are losing the 135mm frame, the 35 will be on its own. I'm wondering whether the zebra coded lens mount for the 21 also changes the lug position so that the frame lines would be 21/35, 28/90 and 50/75.
And then there's the runt of the litter, the 24mm. Where does that fit in?
There's so much about the viewfinder and rangefinder we do not know...
Due to the crop factor which reduces the size of the circle of confusion, a.72x magnification won't be large enough to focus accurately a 90mm lens at f/2 or a 75mm one at f/1.4 if the mechanical base length of the rangefinder remains the same as classic M cameras IMHO.
Best,
LCT
expect sales of the 75/2.5 color heliar to skyrocket.
A 1.33 crop factor amounts to approx. 75% of the 24x36mm frame, yes? I've made crops of as much as 50% from a slide taken with my 90/2 or 135/4 and (ignoring the enlargement of grain) never found one that showed up a mis-focus compared to the full-frame shot enlarged to the same end size.
Bob Ross
06-09-2006, 18:23
Keep in mind that the bayonet lug of the 21mm currently activates the 28 and 90 framelines and if we are losing the 135mm frame, the 35 will be on its own. I'm wondering whether the zebra coded lens mount for the 21 also changes the lug position so that the frame lines would be 21/35, 28/90 and 50/75.
And then there's the runt of the litter, the 24mm. Where does that fit in?
There's so much about the viewfinder and rangefinder we do not know...
I did read a rumor somewhere that the coding was done by replaceing the mount, so changing the frame lug would be possible and for the 21mm & 24mm it wouldn't affect the film bodies. I did fool around with the .72x frame line pictures from the Leica site, adding the new FOV lines. When I did this, I just left the 28mm frame as the 21mm frame, so there would be three frames. The 24mm could go with the 35mm to keep it company :)
humanized_form
06-09-2006, 19:35
expect sales of the 75/2.5 color heliar to skyrocket.
yep. i also anticipate the 28mm Ultron will become very popular, as it will be the fastest 36mm lens available for the M8.
A 1.33 crop factor amounts to approx. 75% of the 24x36mm frame, yes? I've made crops of as much as 50% from a slide taken with my 90/2 or 135/4 and (ignoring the enlargement of grain) never found one that showed up a mis-focus compared to the full-frame shot enlarged to the same end size.
Just a question of math IMHO.
The minimum EBL (effective base lenght) of a rangefinder is given by the formula b' = (e * f^2) / (k * z) where b' is the EBL, e the visual acuity (0.0003 at approx. 1 arcmin), f the focal length, k the aperture and z the circle of confusion (CoC).
For 35mm cameras, the generally admitted CoC value is 0.030mm so taking into account the 1.33x crop factor of the Digital M, the size of its CoC would become 0.03 : 1.33 = 0.023mm.
This way, the minimum EBL for a 90mm lens at f/2 is 53.87mm when the actual EBL of a .72x finder is only 49.86mm.
Would be even worse if calculation is based on a CoC value of 0.025mm (0.019mm with 1.33x crop factor) sometimes associated to Leica lenses.
Best,
LCT
This said, we don't know yet which mechanical base length will be chosen for the rangefinder of the Digital M.
Suffice it to choose a larger one a la Zeiss Ikon to resolve most problems i guess.
Best,
LCT
This said, we don't know yet which mechanical base length will be chosen for the rangefinder of the Digital M.
Yes we do. It's the same as the rest of the M film bodies. We know that because they've already stated that the googled lenses (like 135/2.8 and Macro-Elmar attachment) will work. So the windows have to be the same distance apart. What we don't know is the magnification factor of the viewfinder, which multiplies against the physical (mechanical) baselength to get the effective baselength.
Ah yes! I forgot the goggles, you're right Ben.
Then i would not be surprised if the VF mag were at least .78x and the finder window somewhat larger to fit 21mm (28mm FoV) frame.
Best,
LCT
Bob Parsons
06-11-2006, 03:42
I don't think we need to be scientists to understand that all existing Leica lenses were designed to focus on the same plane, the same distance from the body flange. So they will all focus on the same plane, the same distance from the body flange in the M8. Just as any lens that could focus on a Nikon F or Canon EOS body focuses on the plane of their digital sensors.
And even if for the sake of argument we suspend common knowledge and sense and allow that the 135 APO doesn't focus on the M8 plane, it would be a simple matter to have the rangefinder cam adjusted so that it does. That would mean it wouldn't focus correctly on a film body, but nonetheless it would be usable on the M8.
I suspect the real reason for Leica saying the 135 is not compatible is due to the offset microlenses in the sensor corners. Those most certainly have been optimised for what will be the most popular wide angles and would probably give corner darkening with long focal lengths.
Bob.
The light rays coming in from a 135/2.8 @ f/4-22 would be the same as a 135/4 @ f/4-22 and that lens is on the codable list. I'm as certain as I can be that the issue is related to focusing accuracy, which would indicate that the M8 will have an effective baselength shorter than the 0.72 M finder.
I suspect the real reason for Leica saying the 135 is not compatible is due to the offset microlenses in the sensor corners. Those most certainly have been optimised for what will be the most popular wide angles and would probably give corner darkening with long focal lengths.
Bob.
The whole idea of microlenses, as I understand it, is to compensate for non-telecentric designs, especially with WAs. So, without being an optical engineer, I doubt this is the reason. Even if falloff for this lens were the case, encoding the lens could provide the information necessary to compensate in the firmware.
The light rays coming in from a 135/2.8 @ f/4-22 would be the same as a 135/4 @ f/4-22 and that lens is on the codable list. I'm as certain as I can be that the issue is related to focusing accuracy,This makes no sense. The 2.8/135, which is in, requires even more focusing accuracy than the 3.4/135, which is out. The reason must have something to do either with the optical characteristics of the Telyt's construction or with the lens coding system.
(Edit: I forgot that the 2.8/135 has googles. So the accuracy argument does make sense. But then again, in 2006 who would use a 180 mm FOV lens on a rangefinder at all?)
I'm interested in the specifics of the coding system because that will be interesting when adapting screw mount lenses to the M8. The system apparently supports lenses with at least eight focal lengths (135, 90, 75, 50, 35, 28, 24, 21), possibly some more (a 15 or 17, maybe), and five different opening apertures (1, 1.4, 2, 2.8, 4, note that only full apertures are supported). Not every single of these combinations makes sense, of course. It could also be that the system has extra provisions for encoding some optical parameters such as vignetting. There have to be some extra codepoints for the M macro adapter and coding the Tri-Elmar(s), too - it would be possible to have the a future Tri-Elmar announce itself differently at different focal lengths, but with the old one I don't think this would be possible. Just assigning every lens an individual codepoint would be possible, too, but that way they would have wasted 45 of their precious 64 codepoints already without presenting any new lenses at all! Maybe they simply couldn't fit the Telyt or its 3.4 opening aperture in there?
But that's a lot of speculation, of course.
Philipp
in 2006 who would use a 180 mm FOV lens on a rangefinder at all?
I'm not sure what the year has to do with it but 180mm has always been and continues to be a very useful and popular focal length. I have cropped hundreds of 135mm shots to approximately the 180-200 fov.
Bob Parsons
06-11-2006, 13:20
The whole idea of microlenses, as I understand it, is to compensate for non-telecentric designs, especially with WAs. So, without being an optical engineer, I doubt this is the reason. Even if falloff for this lens were the case, encoding the lens could provide the information necessary to compensate in the firmware.
As I understand, microlenses can give an overall sensitivity increase including avoiding some of the light loss due to high angles of light incidence in the corners. Further improvements in the corners are possible by offsetting the corner microlenses as is done on the DMR sensor. However the offset is only optimum for a particular angle of incidence. The angle depends on where the lens exit pupil is placed in relation to the sensor, which in turn depends on the lens design.
If the offsets are optimised for wide angle RF lenses which tend to be symmetrical designs with the exit pupil near the film plane then it's possible performance would not be good with a more telecentric design such as a 135mm.
The light rays coming in from a 135/2.8 @ f/4-22 would be the same as a 135/4 @ f/4-22 and that lens is on the codable list. I'm as certain as I can be that the issue is related to focusing accuracy, which would indicate that the M8 will have an effective baselength shorter than the 0.72 M finder.
The positions of the exit pupils for the two 135 lenses could be different due to their optical design. So one may give acceptable corner performance with offset microlenses and not the other.
We'll have to wait and see if the answers will come from the horse's mouth. :D
Bob.
I'm not sure what the year has to do with it but 180mm has always been and continues to be a very useful and popular focal length.Well, not with rangefinder lens makers. Has any new >135 mm rangefinder lens been developed ever since SLRs came about?I have cropped hundreds of 135mm shots to approximately the 180-200 fov.Cropping is an entirely different story. You still use a 135 mm lens for framing. Using a >135 mm lens with a rangefinder is awkward, lenses are big, focusing is potentially inaccurate, framing is difficult and parallax is a problem. I think we can all agree on this. Given the development of SLRs over the last fifty years or so rangefinders really have very few advantages..
Philipp
[QUOTE=jaapv] " The 3.4 apo 135 cannot be used at all as the focal point is not on the sensor."
What about the older but still great Tele-Elmar 135/4 lens?
I have re-read both tables on Leica's recent 6-bit lens coding memo and don't find this good lens mentioned on either the "current" or "discontinued" tables. Did Solms just forget about this optic?
The positions of the exit pupils for the two 135 lenses could be different due to their optical design. So one may give acceptable corner performance with offset microlenses and not the other.
We'll have to wait and see if the answers will come from the horse's mouth. :D
Bob.
Yep, so far it's all been speculation that might as well have come from the other end of the horse :D However the exit pupil of any of the >35mm lenses is quite far forward. The rear elements themselves are all rececessed up into the mounts.
Cropping is an entirely different story. You still use a 135 mm lens for framing. Using a >135 mm lens with a rangefinder is awkward, lenses are big, focusing is potentially inaccurate, framing is difficult and parallax is a problem
But cropping is exactly what will happen when a 135 is used on an M8. It'll be the same size lens as it always was, focus as it always did. Just the subject inclusion within the frames will be smaller. The frame itself could be bigger, depending on the finder magnification.
What about the older but still great Tele-Elmar 135/4 lens?
That's the one I have, mine's the 39mm version with the removable lens head. I've already used it on my 20D (via OTZFO short focusing mount + 14167 V-R adaptor + R-EOS adaptor) and it's great (by definition the exit pupil must be the same distance from the sensor as it is from film and would be on an M8). The 20D has a 1.6 vs 1.33 crop, but no offset microlenses. If I can find someone with a 1D-II or a 5D I'm going to see how it works. I think the only reason Leica isn't considering the 135 as viable on the M8 is because they must have lowered the finder magnification so the effective base is the same or less than a 0.58 M body, probably to give more eye relief. No other explanation makes sense.
Due to the reduction of the circle of confusion (0.03/1.33=0.023mm) a 135mm lens would need a 60mm RF effective base length (EBL) to be focused accurately at f/4.
Assuming that the RF of the digital M will have the same mechanical base length as that of classic M camera, such an EBL would need at least a .87x finder magnification.
Best,
LCT
Or a .72 and the 1.25 magnifier (0.9x).
Yes but .72 wouldn't be enough to focus a 'cron 90 or a 'lux 75 at full aperture and we can assume that Leica will not ask us to use the magnifier for those lenses IMHO.
Best,
LCT
Is it possible to have two magnifications in one viewfinder that could be switched according to the type of lenses one wants to use. One that works with WA and one for the longer end.
Having no idea about the technical side of rangefinders, I think I remember somebody saying that the magnification can be changed by just exchanging one lens in the VF.
Is it possible to have two magnifications in one viewfinder that could be switched according to the type of lenses one wants to use. One that works with WA and one for the longer end....
I don't know but it would be a great idea if the magnifications could switch automatically when changing lenses.
But how to do this for zebra and non-zebra lenses?
Best,
LCT
Mark Norton
06-23-2006, 14:54
This is a hobby horse of mine because I'd like to see the viewfinder handle the full range of lenses from 21 - 90 mm without recourse to aux finders, magnifiers and especially the need to choose the magnification when you buy the camera. They're all fudges to some degree to improve on the basic design.
A dual range viewfinder would go a long way to improving the usability of the viewfinder. The thread about new lenses on this forum has the brain numbing detail.
I don't think you'd need to switch finder magnifications automatically, it would be fine to simply select "wide" and "normal" as you wish.
...I don't think you'd need to switch finder magnifications automatically, it would be fine to simply select "wide" and "normal" as you wish...
Medium frames (28, 35, 50) would remain visible with both magnifications i guess.
Great idea, Mark.
Best,
LCT
vBulletin® v3.6.8, Copyright ©2000-2013, Jelsoft Enterprises Ltd.