Tim Ashley M(240) Review

[algrove]

Advertisement (gone after registration)

Tim- I think you are on to something with EWARB. As I posted somewhere when one pushes way to the right by looking at the histogram as one should, there is a chance that some of the RGB could get off the histogram to the right where you do not see it.

 

I have had more than one person express these reservations about ETTR. I am just not sure how the clipping settings work versus that thought even though I do as you and set the high end at 253. What if only one channel is over the clipping #, will the 253 catch that?

 

One question which I do not recall if you covered. Have you tried to manually set the WB? With my M9 I would routinely shoot a shot of my tiny WhiBal card to set it like Thorsten recommended some time ago.

[Rick]

Tim -

 

Having used the RX1 for a while now, exposing to the right doesn't seem to be needed or at least it doesn't produce results like it does on other sensors. I try to just expose being careful to not blow highlights. That's all. Seems to produce good files in most situations.

 

Rick

[ptomsu]

Thank you Peter. I think there's actually not a lot to stop people buying the M (apart from price!). They tend to hold their value reasonably well, especially whilst current, and MM aside the 240 is still the best way of using M glass. It is also pretty usable despite the quirks. The question for me, to be tested in my upcoming M 'system' reviews, is whether the M glass is as good as we think it is once sensor resolutions get to the 25mp class and possibly higher.

 

A word on ETTR. I am pretty familiar, as are you, with the math of it but I do have a feeling that some of these Sony sensors don't obey the rules as much as we think. I think that pushing exposure so far to the right as to just avoid clipping important areas, thereby creating a file that needs significant adjustment in RAW conversion to look like a JPEG of the scene would have looked if metered normally, might acutally lose some mid-tone detail. I haven't tried to prove this but I wonder if these chips and their associated in-camera processing are somehow re-allocating the bits of data across the tone curve. I intuitively feel that I get better results with both RX-1 and D800E in most normal lighting conditions if I "Expose With A Right Bias" (EWARB to coin a nasty acronym) rather than if I fully ETTR.

 

What do other people think?

 

Tim,

 

I made similar observations that what you call EWARB in many cases gives better results with the D800E than ETTR. Sorry I did not explain that exact enough in my post.

 

What I usually do with the D800E is to measure pure white (if available in the frame) and then overexpose by 1.5 - 2 stops. In most cases the 1.5 gives something like EWARB while 2.0 gives real ETTR - at least in my D800E.

 

As I tend to mostly use manual metering an a Leica M digital, this should be easy to set. The M (even the NEW M) for me is not a super fast action camera (although it can be used as such with appropriate lenses and setting) so I am just happy with the "slow" speed of manual operation, which in many cases turns out to be faster finally than always keeping auto measurements in mind and thinking about corrections. Also all files from one shoot will then need the same adjustments if so.

 

Peter

 

PS: still not 100% convinced if I should not wait a bit longer to get the M till Leica has figured out what else they can improve in FW. And meanwhile again get a Fuji XE1 and X100S for the next year or so. My M glass has rested now for 4 years, so if it does for 1 more year it would not hurt

[frasier]

Thank you Peter. I think there's actually not a lot to stop people buying the M (apart from price!). They tend to hold their value reasonably well, especially whilst current, and MM aside the 240 is still the best way of using M glass.

 

It is funny ... your statement reminds me of a posting in this froum from a few years ago. Somebody explained, that Leica is for digital cameras, what Harley Davidson is for motorcycles. The other day I read in may FAZ (German newspaper) that from the top 25 motorcycles, which hold their value best after 3 years, on position 19 there is a BMW. All other motorcycles from the list are Harley Davidsons ... So I am pretty happy owning a Leica and a Harley, even if it is "old fashioned" and not state of the art technique

 

Best, Jochen

[120]

"The rangefinder itself seems to be redesigned..."

"...the RF knows how to focus this lens so as not to be fooled by its slight residual focus shift--and in a very precise manner..."

 

Is there somewhere Leica says the rangefinder is redesigned?

 

I don't understand the other comment...

[algrove]

I just looked through my M240 brochure and I could not find anywhere saying it was redesigned or whatever. It sure does seem to focus better for me, but then again I want it to focus better too. Whether it's me(which I doubt) or the camera, it works very well.

[Tonki-M]

Advertisement (gone after registration)

Is there somewhere Leica says the rangefinder is redesigned?

 

not sure about the word 'redesigned', but Leica did altered with the original design of the rangefinder when they removed the illumination window (similarly done to the M9 Titanium?), but mechanically, it should performs the same...as it was not stated by Leica anywhere of any intention to alter it.

[UliWer]

The rangefinder of the M (240) shows some novelties, which might altogether justify the term "redesign":

 

• LED-illumination of frames instead of ambient light illumination,
  

• calibration of the frames for a distance of 2m - like it was for the M8-2,
  

• leaving out the lever for frame-selection - like it was for the very first M3.
  

 

Though all this cannot not have any influence on the precision for focussing as this depends on the coupling between lens and the rangefinder mechanism. And we know that the rangefinder cannot be more or less precise if the lens is off calibration.

 

As long as nobody can tell us that there has been altered anything else in the mechanism any reports that it was more precise than the traditional M-rangefinder should meet some scepticism.

[tashley]

Sceptical or not, I can tell you with a strong degree of certainty that something significant has been done to the design or mechanics of the RF. It is a very great deal more accurate, and a great deal more consistent than the M8 or M9 regarding near/far performance. Also, I have shot about 1,000 frames so far and in my experience the old one would have probably drifted by now, just a bit, particularly after a trip.

 

Put it this way - if your previous design was problematic but still in some shipping units, would you make the change loudly or quietly?

[tashley]

"The rangefinder itself seems to be redesigned..."

"...the RF knows how to focus this lens so as not to be fooled by its slight residual focus shift--and in a very precise manner..."

 

Is there somewhere Leica says the rangefinder is redesigned?

 

I don't understand the other comment...

 

Please see my comment in the post above this one...

 

Regarding residual focus shift, I am referring to the previous version of this lens, which was plagued by focus shift (a tendency of the field of focus to shift further away as the aperture is stopped down). This was acknowledged as a problem by Leica, and was revealed much more clearly by use on digital sensors than ever on film. The new version was designed in part to combat this issue but it still does have a little shift, though the point of focus normally stays within the field of focus as that field shifts further away on stopping down. As for what this can mean for focussing the new M, I will be covering that in an upcoming piece but in short, with this lens, focussing wide open and then stopping down before shooting seems to give better results. To say more would be to write a great deal and to pre-empt the upcoming piece.... and I am still investigating it, too!

[UliWer]

Sounds interesting since a significant change of the "heart" of the M should not take place unnoticed.

 

Have you noticed anything which differs from earlier models on the mechanical parts one can see in the camera's lens throat? One cannot see much, though if there were any changes - maybe looking insignificant - this could lead to a clue that more was changed.

[jaapv]

Still, I fail to see how the mechanical rangefinder can be able to detect the mechanical aperture of the lens. This would be prerequisite for a "knowing" and compensating focus shift.

I think the only magic going on here is a simple reduction of the mechanical tolerances of the rangefinder mechanism, allowing the user to focus more accurately. In other words, the Indian got handed an arrow that flies more true.

[tashley]

Still, I fail to see how the mechanical rangefinder can be able to detect the mechanical aperture of the lens. This would be prerequisite for a "knowing" and compensating focus shift.

I think the only magic going on here is a simple reduction of the mechanical tolerances of the rangefinder mechanism, allowing the user to focus more accurately. In other words, the Indian got handed an arrow that flies more true.

 

That would appear intuitively true but my best guess is that it is placing the POF at just the right place in the very thin DOF such that the POF is still in focus wide open, then remains at a particular place within the developingly deep zone of focus so as to keep the edges of the frame within that zone. I will give examples when I publish the piece but this lens has a complex MTF that results in complex and 'tricky' behaviour of its field of focus as you stop down. I think the M's RF handles this better, though of course it is going on the same information from the lens as it always had.

[wattsy]

Sceptical or not, I can tell you with a strong degree of certainty that something significant has been done to the design or mechanics of the RF. It is a very great deal more accurate, and a great deal more consistent than the M8 or M9 regarding near/far performance.

 

Have you taken the top off the camera to have a look to see what differences there might be, Tim? The new camera has the same measuring baselength and viewfinder magnification as the M8/9. As you know, the RF mechanism is quite simple (and crude) and I'm not sure it lends itself to some significant design change (apart from increasing the length of the measuring base or VF magnification) that will noticeably increase accuracy. Other than being better calibrated (which is possible – my M9-P came back from a sensor repair a couple of weeks ago and seems much more accurate than ever before throughout the focussing range), I can't imagine what Leica could have done to make the RF "a very great deal more accurate". I also think, if Leica had genuinely improved the RF mechanism, they would have trumpeted it in the recent marketing.

[Maarten]

My hypothesis is that by removing the frame line illumination window a source of false light was removed and the contrast of the RF patch has been improved.

As we know the contrast in the RF patch determines to a large extend the final accuracy of the eye to judge correct focus.

This could explain the phenomenon Tim is describing.

Perhaps someone owing both a M8/M9 and M can check if such a difference in contrast is visible?

 

Maarten

[mjh]

Whatever Leica may have done to improve rangefinder accuracy, a lens- and aperture-dependent compensation of focus shift with a mechanical rangefinder is virtually impossible.

[tashley]

Well, all must remain speculation in the absence of any confirmation from Leica but I am really quite sure that some significant change has been made to the mechanics themselves. The finder is more accurate across a range of lenses and regarding near/far behaviour and it even feels different.

 

Regarding lens and aperture dependant changes, I think that it could be done but I don't think it has. I think the change is purely mechanical and is very evident in results but don't forget that the camera knows which lens is in use if coded and also has an educated guess at aperture. But as I say, I don't think it's as complex as that. I think they knew that the old one was hard to calibrate accurately and prone to drift and I think something has changed to mitigate those issues, and that that something is likely purely mechanical.

 

Time will tell and I am not a one for "I told you so" but I am telling you so!

[jaapv]

I agree that it has likely been tweaked, but I disagree that even if the camera knows the lens and can guess the aperture that it would be possible to transfer those data to a purely mechanical focus mechanism other than through a highly complicated (and vastly expensive) servo mechanism.

[tashley]

I agree that it has likely been tweaked, but I disagree that even if the camera knows the lens and can guess the aperture that it would be possible to transfer those data to a purely mechanical focus mechanism other than through a highly complicated (and vastly expensive) servo mechanism.

 

Electro mechanical. It could certainly be done but it would not be the best way of doing it....

[Advertisement]

Hallo,

Look here: Leica M

[jaapv]

Nothing so crude; Leica could adapt this patent into the rangefinder optical system.