M8 Viewfinder

[ho_co]

Advertisement (gone after registration)

This LFI article referred to DoF in a confused way IMHO. Apertures don't change with crop factors....

LCT--

I found the LFI article tough reading as well, but I think Karbe proved his point, showing why a smaller imaging surface must of its very nature give increased DoF--that's why a Digilux 2 has more DoF at "28mm equiv" than a 28 Elmarit on an M6. (Also check http://www.leica-camera-user.com/showthread.php?t=473 for a simpler explanation of the phenomenon.)

 

But we already knew that smaller sensors brought more depth of field. The genius of Karbe's presentation is the concept of "equivalent aperture." It strikes me as a seminal idea, but why would someone do the work to come up with it? I'm sure he doesn't expect us to start using the term in describing digicam lenses. Again, I admittedly jump to the conclusion that the original impetus might have been consideration of the digital M's finder.

 

"Apertures don't change with crop factors. A f/2 lens on a M6 remains a f/2 lens on the R-D1 and will remain a f/2 lens on the M8...." That was my original feeling as well. But if you say that, then you then need to go the next step and say that a 90mm on the M6 is still a 90mm on an APS-C sensor: the smaller sensor simply crops the edges. If I put a 90mm on the M6 and enlarge only the center of the image, it's identical to using a smaller sensor to start with. It's still a 90mm, and I still shot it at f/4; but the image has the field of view of a 120mm (assuming M8 sensor size)--but the depth of field is greater than that of a 120 at f/4 on full frame. And you can explain that fact in two ways: One, invoke Karbe's "equivalent aperture"; or two, point out that the image has the DoF of a 90mm at f/4 because that's what it was in fact shot with. The two explanations are identical.

 

Do me a favor: Apply your rangefinder baselength formula to the figures that I derive from the Stefanus LFI article: a 120/2.7 and a 180/3.7 (in the latter case, remember the 1.5x goggle magnification). (This all assumes I understand his point, and I'm not sure I do. But then, as Einstein said, "Fantasy is better than knowledge--because knowledge is limited.")

 

AND--I want you to know that it was because of the calculations you posted earlier that I included my fallback, the second part of my original post, saying that if the first part was wrong for assuming the 0.72x VF, then Leica would likely go to the 0.85x.

 

The Germans have a saying, "When I write my Faust...." I think for this forum the saying should be, "When I build my M8...."

 

--HC

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[ho_co]

 

It's going to be interesting because there's a conflict between achieving what most of us would regard as the ideal solution and compatability with what is out there now. The more you enforce compatability with every widget which has gone before, the less likely you will achieve your ideal solution....

 

More generally, I really hope there is something about the M8 which will surprise and delight us beyond what we think we know and assume (top quality images). Something we haven't thought about or didn't think they'd do, something to create a bit of a stir, something beyond "It's just Leica's take on the R-D1".

 

Mark--

It will indeed be interesting! Look at the table that Nikon includes with the D200 to show which lenses are usable with which body functions--very clear, but it shows clearly the problems you get into trying to grandfather in your legacy. (That chart may or may not appear below; I'm not doing well figuring that out. ))

 

There may be something unexpected to "surprise and delight." I certainly hope so. One very respected Leica aficionado opined that when the M digital arrives, it might be simply a digital camera based on a 1950's concept. Let's hope for surprise and delight instead.

 

--HC

Picture 1.pdf

[lct]

..Do me a favor: Apply your rangefinder baselength formula to the figures that I derive from the Stefanus LFI article: a 120/2.7...

I don't mind Howard but to focus a 120mm lens at f/2.7, the M8 rangefinder would need an effective base length of 70.93mm against 53.87mm for a 90mm lens at f/2, which is quite different.

You wrote "The question in both cases of rangefinder magnification is whether the rangefinder base length would be accurate enough for the 90/2 (120/2.7 equiv)" and i can't agree with you.

To calculate the RF magnification you must take into account the actual focal length and aperture of the lens, i.e. 90/2 and not 120/2.7, otherwise you'll get a false value (70.93mm) instead of the right one (53.87mm).

[marknorton]

Howard, not sure your link is working but I know the Nikon chart you are referring to. I think there will be similar chart for the M8. Lens compatability is for me the most important for them to achieve - that's where the money is tied up. Other accessories are much less important.

[ho_co]

You wrote "The question in both cases of rangefinder magnification is whether the rangefinder base length would be accurate enough for the 90/2 (120/2.7 equiv)" and i can't agree with you.

To calculate the RF magnification you must take into account the actual focal length and aperture of the lens, i.e. 90/2 and not 120/2.7, otherwise you'll get a false value (70.93mm) instead of the right one (53.87mm).

LCT--Thanks for checking!

 

"...you must take into account the actual focal length and aperture of the lens...." That was my point all along. Must not have expressed it well.

 

But now I'm confused: Earlier, you wrote "According to formulas i referred to previously, a .78x VF magnification would be just enough for 90/2 and 75/1.4 lenses to be focussed accurately at full aperture." But we already use those lenses comfortably with 0.72x, and here you say that 53.87mm is adequate for the 90/2.

 

So I'm still sticking with my assumptions regarding the 0.72x finder and frame choices, though we can still hope for a solution other than the current frame mask mechanism, as Mark and you and others suggested!

 

Anyway, we can speculate for another two months!

 

--HC

[lct]

...But now I'm confused: Earlier, you wrote "According to formulas i referred to previously, a .78x VF magnification would be just enough for 90/2 and 75/1.4 lenses to be focussed accurately at full aperture." But we already use those lenses comfortably with 0.72x, and here you say that 53.87mm is adequate for the 90/2. ...

Didn't i tell you that the LFI article was confusing?

DoF depends on the circle of confusion which is reduced in APS cameras because of the crop factor.

A .72x magnification is enough for full frame cameras but not for 'cropped' ones for this simple reason.

We discussed this already here Howard.

[ho_co]

Advertisement (gone after registration)

Didn't i tell you that the LFI article was confusing?

DoF depends on the circle of confusion which is reduced in APS cameras because of the crop factor.

A .72x magnification is enough for full frame cameras but not for 'cropped' ones for this simple reason.

We discussed this already here ...

 

LCT--

Yeah, you gave me those forumlas but that doesn't mean I use them!

 

Thanks! This puts the two ends together. As I told you in the original exchange, I knew that I needed to do more thinking.

 

Okay. We've made the new finder. Where next?

 

--HC

[ho_co]

Just learned from another thread that the 24mm lens keys the 35 + 135 frames on current cameras. Boy, did Leica plan ahead or what!

 

Now everything falls into place as I said above in post 23 this thread, with the 24mm keying the 32 + 47 frames.

 

Only problem left are LCT's pesky persnickety but perfectly presentable and doggedly direct depth-of-field calculations!

 

Simple solution: Throw in with each M8 the 1.25x magnifier.

 

But update it to a higher eye-point first, and redesign the rangefinder so that it has a larger exit pupil with higher eyepoint and maybe a live sensor to connect the view live to a computer and maybe a laser rangefinder, with larger battery pack of course to handle it, the LEDs and the MP3 player.

 

Guys and Gals, I think we've got it!!!

 

--HC

[adan]

I think the concepts of "equivalent" apertures and focal lengths should be avoided like rattlesnakes - they cause more confusion than they eliminate.

 

The widest setting on a Digilux 2 zoom is a 7mm lens - of course it delivers more DOF at any given aperture than a 28mm lens.

 

The main function of aperture settings is for exposure control - DOF is secondary. So the absolute value is most important, not an "equivalent" value as it applies to DOF. Any lens at f/2.0 projects the same light intensity per mm^2 of film or sensor area, allowing for the same shutter speed at a given ISO.

 

F/2 is a specific value - it means a lens opening of the focal length F divided ("/") by the physical diameter of the aperture. A 7mm lens set to f/2 is set such that the diameter of the opening is f(=7mm) divided by (/) 2 = 3.5mm. A 28mm lens set to f/2 is set such that the diameter of the opening is f(=28mm) divided by (/) 2 = 14mm.

 

The nice thing about this is one can use, say, f/2 and 1/250th second, for a given amount of light, without having to worry about the actual diameter of the opening. In terms of light flow, all f/2 lenses are the same. Much easier than having to remember (for exposure purposes) that a 45mm aperture in a 90mm lens is the same as a 25mm aperture in a 50mm lens or a 3.5mm aperture in a 7mm lens.

 

A 90mm f/2 lens has a full aperture 45mm wide - this does not change, regardless of the size of the imaging surface behind it. Putting the lens in front of a 1.33x cropped sensor such that it frames like a 120mm on film does not magically enlarge the opening to 60mm (f(120mm)/2) nor does it magically reduce the amount of light available for exposure to the amount from an f/2.67 (120mm/45mm) lens.

 

For a given absolute focal length/aperture (say, 90mm/f/4), the DOF will be reduced somewhat if one crops out the middle of the image and enlarges it more - some things that looked sharp at a 10x enlargement may start to be noticeably fuzzy at a 13.33x enlargement, and the already-obvious blur circles in the foreground/background will be 1.333x larger as well.

 

This is true with film as well as digital. And it is the same thing LCT's equations state - put into words instead of numbers.

 

In anticipation of the digital-M, I just swapped a 90 f/2 for a 50 f/1.4. I can use the extra f/stop for exposure, and I know the 50 f/1.4 will blur backgrounds nicely while allowing a bit more leeway for critical focus, whatever the finder magnification of the M8 is.

 

Worrying about whether it will be exactly equivalent to a "68.98mm f/1.67254" as regards DOF is silly - it will be a nice portrait/short tele lens.

[adan]

Ho_co: Yes, Leica did plan ahead to some extent - the 21 Elmarits key the 28/90 frame set.

 

I would not, however, bet heavily yet on there being 21 or 24 framelines built into the M8. 21/28 and 24/35 pairs would be roughly as close together (and visually distracting/confusing) as the 50/75 pair is now. I think the M8 will frame for 28-to-90 lenses, roughly the same FOV choices Leica used from 1960-1980 (M2, M4, M4-2). The 28mm frame in the M4-P through M7/MP is (IMHO) rather useless.

 

We should remember that when the M6ttl .58x "wide-angle" finder was introduced, everyone thought the 24 frame would be added. Didn't happen.

 

I could be wrong, of course...

 

But I do expect the "whole viewfinder", ignoring framelines, will be a fair match for either the 21 or 24 (which one will depend on the final magnification Leica has chosen).

 

Possibly, in an M9 with hypothetical LCD frames, where the pairings could be eliminated by using the zebra-code to flag the frame selection, there could well be 21 and 24 framelines, uncomplicated by additional frames.

[ho_co]

I think the concepts of "equivalent" apertures and focal lengths should be avoided like rattlesnakes - they cause more confusion than they eliminate.

 

The main function of aperture settings is for exposure control - DOF is secondary.

 

This is true with film as well as digital. And it is the same thing LCT's equations state - put into words instead of numbers.

 

Worrying about whether it will be exactly equivalent to a "68.98mm f/1.67254" as regards DOF is silly - it will be a nice portrait/short tele lens.

 

Andy--

 

Please don't misunderstand my meanderings on the topic; I agree completely with what you say except that I feel one may choose equally to use aperture either primarily for exposure or for DoF control.

 

I found the Karbe argument interesting because it came to the question of depth of field completely differently from the usual approach, and the Stefanus article lightly humorous for tacking on the thought that anyone who used (full-frame) Leica lenses' aperture control simply for exposure adjustment rather than for depth-of-field consideration wasn't using them to their full potential--a thought which seemed not to fit into what was otherwise a non-tendentious piece.

 

And as I said, I think there is likely more than just a lens designer's "Aha! moment" behind the presentation.

 

--HC

[ho_co]

I would not ... bet heavily yet on there being 21 or 24 framelines built into the M8. 21/28 and 24/35 pairs would be roughly as close together (and visually distracting/confusing) as the 50/75 pair is now.... The 28mm frame in the M4-P through M7/MP is (IMHO) rather useless.

 

But I do expect the "whole viewfinder", ignoring framelines, will be a fair match for either the 21 or 24 (which one will depend on the final magnification Leica has chosen).

 

Andy--it's all speculation, of course, and an exercise in keeping the brain awake while awaiting photokina.

 

I'm bothered only a bit by the similarity in field of the 32 and 47 frame lines in my plan, though I don't personally find use of the current 50 and 75 frames troublesome. I'm bothered more by the addition of a (third) set of 28 frame lines to the 37 and 120 set. If we're still using frame masks (as I consider likely), I wonder about losing contrast with all the slits. (Please don't be annoyed by my use of equivalent fields of view; I'm just following custom. )

 

As you point out, the full frame of all the post-M3 M's has been (un)usable for the field of view of the 28mm lens. As an eyeglass wearer with a high nose bridge, I find even the 35mm frame a bit uncomfortable, and with you the 28mm frame virtually unusable.

 

I think we'll all be delighted to see the camera. And one thing we know is that it won't be all the things we think we've designed into it!

 

--HC

[burkey]

It would seem to me that eventually some sort of accomodation will be needed for us other folks with CV or Zeiss M-mount lenses. I'm strongly leaning towards the M8 but doubt whether or not I'll ever be able to replicate the CV and Zeiss M-mount inventory I have with all new Leica glass. 'Just hoping, that's all.

[lct]

It would seem to me that eventually some sort of accomodation will be needed for us other folks with CV or Zeiss M-mount lenses...

I don't think so otherwise the same accomodation would be necessary for non-zebra Leica lenses as well.

[burkey]

LCT - what do you think will show in the M8 viewfinder, as far as frame lines is concerned, when M-mount lenses without the indicator marks are mounted on the camera?

[lct]

...I'm bothered only a bit by the similarity in field of the 32 and 47 frame lines...

There are no 32mm or 47mm lenses, Howard, so we can't get 32mm or 47mm frame lines either.

[lct]

LCT - what do you think will show in the M8 viewfinder, as far as frame lines is concerned, when M-mount lenses without the indicator marks are mounted on the camera?

Framelines are determined by the length of this flange so far then i don't expect any change in that respect.

 

[Bob Ross]

I like your comment about the 32mm & 47mm lenses. If we start talking FOV instead of FL, we'll all be addled...:-)

Let me try a picture posting (first time) of my scribbles about the frame line possibilities.

[Advertisement]

Hallo,

Look here: Interesting Leica gear

[Bob Ross]

I like your comment about the 32mm & 47mm lenses. If we start talking FOV instead of FL, we'll all be addled...:-)

Let me try a picture posting (first time) of my scribbles about the frame line possibilities.

Well, that didn't work so here is the link http://www.pbase.com/rossrtx/image/61593471

[ho_co]

There are no 32mm or 47mm lenses, Howard, so we can't get 32mm or 47mm frame lines either.

LCT--

Check our earlier correspondence. These are the equivalent fields of the lenses listed. I pointed out how each lens would key (as now) two or (new) three sets of frame lines. You may restate them as you will.

 

--HC