Jump to content

New Summarits - Puts review part II


thrid

Recommended Posts

Advertisement (gone after registration)

I don't remember having been bothered by this when i used old Contax rangefinders but i may be wrong. Any opinion from Zeiss Ikon users? A neat optical solution could be to choose a longer RF base a la ZI and a higher VF magnification a la R-D1. This would imply a larger VF window though, hence the difficulty or impossibility to fit some Visoflex and lenses with goggles like 35/1.4, 35/2, 35/2.8 and 135/2.8.

 

So then make a variant for those of us who don't use a Visoflex or spectacles lenses, in other words about 99.9% of M8 users.

 

Wilson

Link to post
Share on other sites

  • Replies 91
  • Created
  • Last Reply

Top Posters In This Topic

Top Posters In This Topic

Posted Images

LCT, I agree with you, but this all assumes the lens and camera are correctly calibrated..

I agree of course but i wonder if we don't overestimate the issue of lenses calibration.

Don't get me wrong, i don't discuss the fact that some late Leica lenses have not the same quality as some earlier ones but i remember myself 3 years ago with my first R-D1. As soon as i've had the latter properly calibrated all my Leica lenses that i thought more or less out-of-focus became spot on.

But i did not demand too much to them.

I mean we have to know the limits of the RF we use folks.

Expecting a high success rate without magnifier from a 90/2 or a 75/1.4 at full aperture is a waste of time with the M8, the same way as expecting such a success from a Noctilux at f/1 with the R-D1.

Link to post
Share on other sites

So like me you are not only blind but deaf as well ;-}}. When I had a hearing test a few months ago as part of a general medical, the lady looked at the hearing test and said "hmm rifle shooter eh?"

 

Wilson, I guess that's exactly why we prefer typing on the Internet. LOL

Link to post
Share on other sites

Peter, I think you are correct. One of the messages that has come through loud and clear from the M8 is that the M rangefinder as a focussing aid is at the limit of its capabilities. Add the triple whammy of long focal length, fast aperture and close focus and you run of out depth of field to hide your mistakes and the camera's limitations. 75mm Summilux at f1.4 and 1.5m? Difficult and what might have been fine for a soft-ish 5cm Elmar on an M3 50 years ago doesn't cut it now.

 

The rangefinder has no margin for error, so mis-adjustment of the camera or lens or user limitations due to their eye-sight, hangover, state of mind or sleep deprivation will all contrive to spoil the images. It's been a wake-up call for me to put a manual focus Nikon 50mm f1.2 on a D3 (not the aspheric 58mm f1.2 Noct-Nikkor, can't find one at a sensible price) and use a single focus patch and focus confirmation to get the focus exactly right. Just like focussing an M with a Noctilux but more accurate.

 

The rangefinder design needs to be reworked for a future camera so that the feedback in the viewfinder is based on actual focus, not "we-think-this-is-what-it-should-be-if-the-camera and-lens-are-correctly-adjusted" focus by proxy we have now. That suggests using the sensor in some sort of live view mode - do away with the shutter completely? - to determine focus and have the rangefinder patch image injected into the viewfinder and shifted left or right to mimic the current rangefinder action. Add electronic framelines which are accurate for the lens mounted and the focus distance and variable finder magnification to complete the picture.

 

Most of all, we need to get rid of the need to calibrate lenses beyond a simple infinity stop reference. We wouldn't get rid of focussing cams because they would still be used for parallax correction but no longer for focussing.

 

Purists might not be happy but let's be clear. The current focussing method of the M8 is not consistently good enough to get the image quality out of the camera the lenses are capable of providing.

 

I need to fully back these comments. Owning almost all actual glass for the M8 I had to go through a adjustment of all my focal length above and equal 50mm on the M8. And I am using now the right diopter correction lenses on the M8 to make it really work. But it takes a long time to focus. No way of real dynamic photography at open apertures.

 

This is really bad, since the fast longer focal length on the M8 are really marvels if they are focused right. But they are not fully usable with that constraints.

 

So I do hope in the M9 we will see a new focussing system, because the current one is simply not up to digital photography.

Link to post
Share on other sites

Whether Leica recognised these limitations when the M8 was being developed is an open question but I feel certain there was no development budget to do anything else - carry the viewfinder/rangefinder over from the M7, adjust (incorrectly) the frame lines and change the viewfinder magnification was the only option open to them at the time.

 

I'd willingly give up the ability to use a Visoflex - an anachronism if ever there was one, goggled lenses and the macro adapter in return for being able to focus more accurately.

Link to post
Share on other sites

Whether Leica recognised these limitations when the M8 was being developed is an open question but I feel certain there was no development budget to do anything else - carry the viewfinder/rangefinder over from the M7, adjust (incorrectly) the frame lines and change the viewfinder magnification was the only option open to them at the time.

 

I'd willingly give up the ability to use a Visoflex - an anachronism if ever there was one, goggled lenses and the macro adapter in return for being able to focus more accurately.

 

Mark, one can only hope that you are correct and that they had very little option but to carry over the M7 design due to time and financial constraints. But the M8 has shown, beyond all doubt, that there is a need for a digital rangefinder camera able to work, as all previous Leica cameras worked, with high aperture moderate telephoto lenses in available light at intermediate subject distances. To me this is the very essence of the M camera.

 

There have been reports that the M8 rangefinder was “reworked” to significantly increase its precision because the problem was recognised. I understand that this related mainly to manufacturing tolerances.

 

What I, and I think you, are saying is that the existing design is not capable of performing to the required level of accuracy and a radical rethink is required. I agree that they should drop any idea of compatibility with goggles etc. These were anachronisms even in their day.

 

I for one would not buy an upgraded M8, no matter what size or specification of sensor or improvement in firmware it had unless it clearly has an improved viewfinder / focusing mechanism. To me this is the M8’s Achilles heel. I can produce A3 prints that are superior to anything I produced using film with the existing camera under “easy” conditions so that is not the limitation as far as I am concerned.

 

There are, as you have commented, many ways that this could be done. I’m still intrigued by my electro-mechanical engineering contacts assertion that the position of the rangefinder cam could be measured with exquisite accuracy, < 1 micron, with readily available position sensors. I’m sure that this could be combined with software and a redesigned long base line rangefinder to solve this problem and maintain compatibility with all existing lenses.

Link to post
Share on other sites

I partially agree with the fact that with M8 and lenses like the recent 1,4s (and Nocti of course) the RF has been put into a sort of crisis: but I am not convinced that a completely new design would be the best way... on the M8 body a longer base would be a problem, and, after all it CAN work right when lenses and RF are finely adjusted... for M8, I think they would better improve the tolerancing of assembly... after all, many tech capabilities have improved a lot from the times when the small-base RF of the LTMs worked well even with 135s at f4...

 

For a future RF camera in FF (or whatsoever bigger format) they anyway will be compelled to design a new body, and this would be the occasion to design a completely new RF, maybe based with some other kind of coupling, of viewing... number of possibilities exist with electronics, even if the old mechanical coupling is so logical and simple (even if delicate to machine & assembly) that is difficult to imagine something better and as durable.

Link to post
Share on other sites

To give an idea of the precision required, think of the Noctilux at f1 and focussed at 2m, say. Depth of field at this point is +/- 5cm. The question is, what angle does the rangefinder view move through as you focus through this distance so that any focussing error is covered by depth of field?

 

At 1.95m and a physical base line of 69.25mm, the rangefinder is looking 2.03 degrees away from the straight-ahead position. At 2.05m, it's looking 1.93 degrees away or 0.10 degrees difference; the mirror moves half this amount, so 0.05 degrees, which means the focussing mechanism has to set the effective mirror position to +/- 0.025 degrees, 1/14400 of a full circle.

 

In moving from 0.7 meters to infinity, the mirror rotates just 2.82 degrees and in moving from 2 meters to infinity, it moves less than a degree. The rangefinder roller moving backwards and forwards about 3mm brings about these tiny rotations, so it's surprising it works as well as it does.

 

However, I can unscrew the lens barrel of a pre-ASPH Summilux and pretend nothing has happened - the viewfinder says the image is in focus when there isn't even a lens there, and as long as there's this disconnect between the viewfinder feedback and the actual focus, mis-calibration is going to be part of the problem.

 

Peter, it's certainly possible to sense mechanical position very accurately using, for example, interference grids, and you can buy calipers and micrometers which do just this without relying on extra-ordinary mechanical precision. The Leica M focussing problem will not be cracked though until they start sensing actual focus. At the moment, it's this focussing by proxy...

Link to post
Share on other sites

I partially agree with the fact that with M8 and lenses like the recent 1,4s (and Nocti of course) the RF has been put into a sort of crisis: but I am not convinced that a completely new design would be the best way... on the M8 body a longer base would be a problem, and, after all it CAN work right when lenses and RF are finely adjusted... for M8, I think they would better improve the tolerancing of assembly... after all, many tech capabilities have improved a lot from the times when the small-base RF of the LTMs worked well even with 135s at f4...

 

For a future RF camera in FF (or whatsoever bigger format) they anyway will be compelled to design a new body, and this would be the occasion to design a completely new RF, maybe based with some other kind of coupling, of viewing... number of possibilities exist with electronics, even if the old mechanical coupling is so logical and simple (even if delicate to machine & assembly) that is difficult to imagine something better and as durable.

 

Luigi,

 

It is not just the RF in the M8 that has been put into crisis but Leica's ability to manufacture and service the focusing mechanism of lenses to the required tolerances. Of the five new/just serviced Leica lenses I have bought over the last 12 months, two were spot on - my WATE and the 20 year old but unused 135 Tele-Elmar. My 50 Summilux front focuses a bit but that is OK as the aperture shift towards back focus then works in your favour. The focus point is still within the depth of focus at f1.4 so that also is acceptable. The just serviced Noctilux was totally unusable with a focus error of 2 meters at 8 meters. Finally my new Summarit 75 is about 5% wrong at close focus and is being exchanged tomorrow. I am taking my M8 with me to London and I am going to test the replacement before I leave the shop.

 

I have posted before that I think Leica's crude and antiquated manual/optical/by eye focus checking method is proving to be both inadequate and very error prone. How many bad Summilux 35's have come out of the factory over the last 18 months - lots and they seem to have great difficulty in getting the bad ones right. Where we have the combination of a barely adequate RF in the camera together with inaccurately made or serviced lenses, it is a recipe for customer dissatisfaction. At least they could solve the lens problem by buying a modern computerised optical bench system like the Zeiss K8 or one of the Japanese equivalents. It beggars belief that they still have not done this. Of the two new Zeiss and two new CV lenses I bought, how many were wrong when new - none. Not a statistically significant sample I know but indicative.

 

Wilson

Link to post
Share on other sites

While I appreciate the erudition and validity of many of the arguments made here, I don't want to have the issue overstated and leave the impression that getting accurate focus quickly, in tough conditions, at open aperture on longer lenses is not something we can expect to achieve. Here's a 75 cron at f2, 1250 iso, 1/15, composed and focused very quickly.....Peter

original.jpg

 

91659239.jpg

Link to post
Share on other sites

While I appreciate the erudition and validity of many of the arguments made here, I don't want to have the issue overstated and leave the impression that getting accurate focus quickly, in tough conditions, at open aperture on longer lenses is not something we can expect to achieve. Here's a 75 cron at f2, 1250 iso, 1/15, composed and focused very quickly.....Peter

original.jpg

 

91659239.jpg

 

We can expect to achieve it - and sometimes do - but it can't be done reliably. Too many shots are near misses. The 75mm ASPH at f/2 is a good test and I get about one in four spot on.

Link to post
Share on other sites

Peter,

 

Fair enough and I have very sharp images taken with my 90 Elmarit and 135 Tele-Elmar but I also have quite a few, snapped in a hurry, that are not as sharp as I would like them to be. I have found that I have to have the 1.25X magnifier on for the 90 and 135 and that in itself is a pain. I would be interested at some point to try a Zeiss Ikon, which is reputed to have a better rangefinder - has anyone any experience of this? If Zeiss ever get round to "digitorum extractum" and produce a digital RF with a markedly better RF performance, then life will get interesting for Leica. Of course Zeiss may well repeat their previous thinking, come to the conclusion that optical convergence is just not accurate enough and go for an autofocus system along the lines of their AX SLR, with moving imaging medium. As AF systems have improved hugely since the AX and G2 days, there is just a chance they could get it right this time.

 

Wilson

Link to post
Share on other sites

Peter- This was one of a series of 4, all requiring slight focus change, all in focus.

Wilson- I just leave the 1.35 on all the time, it's definitely worth it to me and accuracy is way up since I made the commitment to it. best...Peter

 

I have just ordered a 1.15 from them. If I don't like it, it will be in the "For Sale" area in a couple of weeks.

 

Wilson

Link to post
Share on other sites

...interesting timing, Wilson... I just ordered a 1.35 from the same source. Wouldn't I like it, we could first try a mutual exchange, to have a 2nd choice... :)

 

Luigi,

 

Might take you up on that but I think 1.15 about right for me. If the Megaperls works, I will also have a Leica 1.25X for sale with leather case.

 

Someone asked if the hood on the 75 would reverse. I think people may have felt it would not. Well I am pleased to able to report that it does reverse on the lens for a neater package while in a bag. The fit of the lens hood is not quite as secure as one might like if like me, you have a UV/IR filter fitted but then this is only for storage. Also glad to be able to say that my replacement Summarit 75 looks to be focusing OK but will have to run full test with chart tomorrow.

 

Wilson

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...