Digilloyd's S2 focus field test.

[eronald]

Advertisement (gone after registration)

http://diglloyd.com/prem/prot/DAP/LeicaS2/autofocus-granite.html

[Advertisement]

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

[kidigital]

Can't see it ...

[eronald]

Can't see it ...

 

 

In summary he says the lenses were superb, out of this world, while the AF performance was insufficient to take advantage of these lenses; he needed F8 to correct.

 

I don't know whether Lloyd had a bad sample, but another tester has verbally corroborated the AF issues to me, with a camera sourced on a different continent, tested by a known landscape photographer.

 

I would expect Leica to solve this issue with a firmware fix .

 

I would recommend any landscape photographer looking et Leica, Zeiss, Hasselblad or N and C equipment subscribe to Digilloyd. It's the only place I know where one can see practical images taken aperture by aperture of the same real-world scene, with tens of lenses.

 

DigiLLoyd's tests are rigorous, accurate and fair, and above all practical. He gives the Leica optics a glowing review. His tests are useless for an event/sports/fashion shooter. They seem wonderfully appropriate for the landscape artist. Lloyd is also responsive to queries by email, and his pages are not copy-locked. Which is precisely why I won't cut and paste here - any prospective buyer of the S2 will find purchasing access worthwhile. As a technical writer myself, I know good technical writing when I see it.

 

diglloyd.com blog: November 2010

 

Edmund

[jrv]

A guy on DP had trouble with his S2 focus, tested it on a focus chart, and found it off quite a bit. After sending his body back to Leica for adjustment he's happy.

 

S2 body now perfect: Leica Talk Forum: Digital Photography Review

 

I've always wondered if it would be possible for a camera to contrast-focus with the main sensor. Point the camera at a focus chart in good lighting and activate a "calibrate" function which would find the focus normally, then test around it with contrast focusing, then remember how far off the normal focus mechanism was. This would avoid the need to trust Mr. UPS any more than necessary with $$$ cameras and/or lens.

[kidigital]

In summary he says the lenses were superb, out of this world, while the AF performance was insufficient to take advantage of these lenses; he needed F8 to correct.

 

I don't know whether Lloyd had a bad sample, but another tester has verbally corroborated the AF issues to me, with a camera sourced on a different continent, tested by a known landscape photographer.

 

I would expect Leica to solve this issue with a firmware fix .

 

I would recommend any landscape photographer looking et Leica, Zeiss, Hasselblad or N and C equipment subscribe to Digilloyd. It's the only place I know where one can see practical images taken aperture by aperture of the same real-world scene, with tens of lenses.

 

DigiLLoyd's tests are rigorous, accurate and fair, and above all practical. He gives the Leica optics a glowing review. His tests are useless for an event/sports/fashion shooter. They seem wonderfully appropriate for the landscape artist. Lloyd is also responsive to queries by email, and his pages are not copy-locked. Which is precisely why I won't cut and paste here - any prospective buyer of the S2 will find purchasing access worthwhile. As a technical writer myself, I know good technical writing when I see it.

 

diglloyd.com blog: November 2010

 

Edmund

 

 

Technical writer, Edmund? I would've guessed sentimental greeting card writer.

 

Thanks for the thoughts and summations.

 

Kurt

[eronald]

A guy on DP had trouble with his S2 focus, tested it on a focus chart, and found it off quite a bit. After sending his body back to Leica for adjustment he's happy.

 

S2 body now perfect: Leica Talk Forum: Digital Photography Review

 

I've always wondered if it would be possible for a camera to contrast-focus with the main sensor. Point the camera at a focus chart in good lighting and activate a "calibrate" function which would find the focus normally, then test around it with contrast focusing, then remember how far off the normal focus mechanism was. This would avoid the need to trust Mr. UPS any more than necessary with $$$ cameras and/or lens.

 

 

What an excellent idea!

 

But I'm not so sure that Lloyd's issues with the S2 were as simple as calibration.

 

Edmund

[ddanois]

Advertisement (gone after registration)

I experienced autofocus issues with my S2 after about two months of use. Prior to that all was fine. I sent it to Germany and it cam back much improved. I'm wondering if the autofocus sensor is able to move as a result of knocks and bangs.

[mwilliamsphotography]

Perhaps Leica should have simply employed a user focus calibration capability like that found on all high meg 35mm DSLRs?

 

It's a no-brainer solution tor slight out of calibration lenses, manufacturing sample variations between cameras and lenses, even eventual micro mount wear, or a combination of all of them.

 

I've used the focus calibration to make minor adjustments to Nikon and Zeiss ZF lenses on a D3X, and have done the same with a Sony A900 and range of Zeiss ZA lenses. The camera can accept something like 15 or 20 automatic calibrations and keeps them in memory until told not to.

 

The Nikon even registered non-cpu lenses that you could call-up manually. This would be a good feature should someone wish to use a Zeiss lens on a S2 and perhaps help mitigate any adapter variations.

 

Not sure if this requires camera hardware, or if it is just a firmware possibility. Leica may not want to admit to any sample variation or less than zero tolerances in mating camera to lenses, but it sure seems it would save on kits sent off to Germany for service adjustments.

 

-Marc

[tom0511]

Perhaps Leica should have simply employed a user focus calibration capability like that found on all high meg 35mm DSLRs?

 

It's a no-brainer solution tor slight out of calibration lenses, manufacturing sample variations between cameras and lenses, even eventual micro mount wear, or a combination of all of them.

 

I've used the focus calibration to make minor adjustments to Nikon and Zeiss ZF lenses on a D3X, and have done the same with a Sony A900 and range of Zeiss ZA lenses. The camera can accept something like 15 or 20 automatic calibrations and keeps them in memory until told not to.

 

The Nikon even registered non-cpu lenses that you could call-up manually. This would be a good feature should someone wish to use a Zeiss lens on a S2 and perhaps help mitigate any adapter variations.

 

Not sure if this requires camera hardware, or if it is just a firmware possibility. Leica may not want to admit to any sample variation or less than zero tolerances in mating camera to lenses, but it sure seems it would save on kits sent off to Germany for service adjustments.

 

-Marc

 

I fully agree and dont understand why they wouldnot want to include such a function.

[ptomsu]

I fully agree and dont understand why they wouldnot want to include such a function.

 

Just because they feel they are PERFECT!

[tom0511]

Just because they feel they are PERFECT!

 

I hope not. dangerous to feel perfect.

[msadat]

limited resource, limted technical capability. but over time it will be a good product. it is not there now. mark if u remember the contax nd times, it is much much better

 

 

Perhaps Leica should have simply employed a user focus calibration capability like that found on all high meg 35mm DSLRs?

 

It's a no-brainer solution tor slight out of calibration lenses, manufacturing sample variations between cameras and lenses, even eventual micro mount wear, or a combination of all of them.

 

I've used the focus calibration to make minor adjustments to Nikon and Zeiss ZF lenses on a D3X, and have done the same with a Sony A900 and range of Zeiss ZA lenses. The camera can accept something like 15 or 20 automatic calibrations and keeps them in memory until told not to.

 

The Nikon even registered non-cpu lenses that you could call-up manually. This would be a good feature should someone wish to use a Zeiss lens on a S2 and perhaps help mitigate any adapter variations.

 

Not sure if this requires camera hardware, or if it is just a firmware possibility. Leica may not want to admit to any sample variation or less than zero tolerances in mating camera to lenses, but it sure seems it would save on kits sent off to Germany for service adjustments.

 

-Marc

[JMacD]

Lloyd's testing is a good reminder to always check your own equipment. In my case, thankfully, it has been right on. Lloyd's experience is not mine. I agree the lenses are out of this world. I disagree that the auto-focus is off with the S system. With me it is amazingly spot on. I think I should go visit him and compare work flow. I never set the S on auto-focus, always Manual Focus, but regularly use the AF button to spot set what I am focusing on.

The new firmware just makes it better, and I appreciate how easy it is to upgrade without sending the camera back in as with some other MF cameras. But Lloyd has these focusing problems with many lens and camera system he tests. I would appreciate Lloyd 2.0 please.

That said, an investment in his site is minimal and should be purchased by any S2 user. Just remember, that as with a given lens, there may be quality variations, and there is only one Lloyd site version available.

 

Attached are two hand held photos, one full frame, one cropped in on the area I manually set auto focus into. Look at the ice crystals. Razor thin depth of focus. No back focus or front focus issues with my lens or camera, but your experience may differ. What I concluded was I needed a tripod so I could change f stop to include more of the branch in focus with a higher f stop. And hey, Lloyd's critique is what assures Leica stays on their toes for the rest of us. So I appreciate his efforts, and am quite willing to invest in his site. I do want to know the name of his Leica Rep so I can talk to him about what Lloyd's problems really are.

Net, test your lenses and camera yourself.

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

[Arif]

Lloyd's testing is a good reminder to always check your own equipment. In my case, thankfully, it has been right on. Lloyd's experience is not mine. I agree the lenses are out of this world. I disagree that the auto-focus is off with the S system. With me it is amazingly spot on. I think I should go visit him and compare work flow. I never set the S on auto-focus, always Manual Focus, but regularly use the AF button to spot set what I am focusing on.

The new firmware just makes it better, and I appreciate how easy it is to upgrade without sending the camera back in as with some other MF cameras. But Lloyd has these focusing problems with many lens and camera system he tests. I would appreciate Lloyd 2.0 please.

That said, an investment in his site is minimal and should be purchased by any S2 user. Just remember, that as with a given lens, there may be quality variations, and there is only one Lloyd site version available.

 

Attached are two hand held photos, one full frame, one cropped in on the area I manually set auto focus into. Look at the ice crystals. Razor thin depth of focus. No back focus or front focus issues with my lens or camera, but your experience may differ. What I concluded was I needed a tripod so I could change f stop to include more of the branch in focus with a higher f stop. And hey, Lloyd's critique is what assures Leica stays on their toes for the rest of us. So I appreciate his efforts, and am quite willing to invest in his site. I do want to know the name of his Leica Rep so I can talk to him about what Lloyd's problems really are.

Net, test your lenses and camera yourself.

 

I agree with you since I too have not seen the issues he has mentioned and I have tried all three lenses (35, 70, 180). Having said that I too respect and subscribe to Lloyd and read his reviews to help me get data points for my decision. In the end, I will test my lenses before buying to minimize that risk.

 

Have a great 2011 and looking forward to many more S2 shots.

 

Best wishes,

Arif

[glenerrolrd]

I agree with you since I too have not seen the issues he has mentioned and I have tried all three lenses (35, 70, 180). Having said that I too respect and subscribe to Lloyd and read his reviews to help me get data points for my decision. In the end, I will test my lenses before buying to minimize that risk.

 

Have a great 2011 and looking forward to many more S2 shots.

 

Best wishes,

Arif

 

Arif

 

I haven t been able to find anywhere in lloyd s blog an explanation of the AF error other than an example of out of focus trees. I tried this with David Kippers S2 yesterday and I would tell you up to 20 ft the S2 nails the focus everytime...if the subject fills the center circle. If the subject splits the circle it will focus on the highest contrast area .

 

Try putting a cap on the grass and only focus part of the circle on the hat ..you will get front focus as the grass has greater contrast . This I could repeat at will. We did it at Starbucks with coffee in a basket ..put the + on the coffee and the edge of the circle would pick up and focus on the basket .

 

The spot focus is too big ...but I could have adjusted easily if I was familiar with the camera.

 

Every AF camera I have had including all the Nikon s hunts for focus sometimes...but with an S2 how could anyone not know the camera wasn t focusing on that tree. The ability to see if the focus is accurate is a big deal and it never gets discussed.

 

I would like to see the focus work on the + and be narrow ..then because I can see the result it might be the most accurate system out there (assuming you aren focus tracking).

 

There are clearly situations where the AF would cause you difficulties but since I shoot M9s ..how could I complain.

 

Roger

[mwilliamsphotography]

Arif

 

I haven t been able to find anywhere in lloyd s blog an explanation of the AF error other than an example of out of focus trees. I tried this with David Kippers S2 yesterday and I would tell you up to 20 ft the S2 nails the focus everytime...if the subject fills the center circle. If the subject splits the circle it will focus on the highest contrast area .

 

Try putting a cap on the grass and only focus part of the circle on the hat ..you will get front focus as the grass has greater contrast . This I could repeat at will. We did it at Starbucks with coffee in a basket ..put the + on the coffee and the edge of the circle would pick up and focus on the basket .

 

The spot focus is too big ...but I could have adjusted easily if I was familiar with the camera.

 

Every AF camera I have had including all the Nikon s hunts for focus sometimes...but with an S2 how could anyone not know the camera wasn t focusing on that tree. The ability to see if the focus is accurate is a big deal and it never gets discussed.

 

I would like to see the focus work on the + and be narrow ..then because I can see the result it might be the most accurate system out there (assuming you aren focus tracking).

 

There are clearly situations where the AF would cause you difficulties but since I shoot M9s ..how could I complain.

 

Roger

 

Roger, I think with knowing how the AF works on the S2 you can manage it to some degree ... maybe ... in controlled situations. The question I asked (as I used the camera) is whether that is acceptable when shooting more spontaneously?

 

I understand your notion that the AF area is too large, however, Lloyd's point wasn't that ... it was that the camera consistently back focused at medium to far distances ... using more than one sample body and various lenses. It wasn't a matter of lens calibration as they all focused spot on using manual focus. His blog doesn't go into all the details, but his paid site does.

 

I've experienced oddly inconsistent AF results even with the subject filling the AF circle ... which I initially attributed to my error somehow ... but his report makes me now doubt that was the case.

 

On Dec.14th, 2010 Lloyd revisited the S2, a different body loaded with the latest firmware (a Leica USA representative went along with him), and he repeatedly demonstrated at will, that the very dramatic back-focusing issues remain. The illustrated sample showed the AF point and the actual AF point was well out of the AF circle many feet behind the intended spot. He has some shocking mouse over comparisons between the AF and MF results from the same scene.

 

This was not as easy to spot even with larger screen images than an LCD would show. However, when enlarged are obvious. His point was that some people may not notice it right away, but it is there and basically spoils the incredible performance of the ultra expensive lenses.

 

As reported else-where, I have also experience camera lock-up three times even with the latest firmware ... Lloyd also reported repeated lock-ups with an S2 that had the previous firmware installed. His was in August, so my original suspicion that it was the 20º cold during my shoots may be incorrect.

 

If nothing more, these kinds of reports serve as alerts when checking out a S2 and lenses.

 

-Marc

[Nick Rains]

I have not see Lloyds focussing errors either, and I have looked real close. Neither the 70mm nor the 35mm are causing focussing problems for me.

 

I'm sure he is correct in his results, it's just that I don't see it with mine.

[mwilliamsphotography]

I have not see Lloyds focussing errors either, and I have looked real close. Neither the 70mm nor the 35mm are causing focussing problems for me.

 

I'm sure he is correct in his results, it's just that I don't see it with mine.

 

That's encouraging.

 

-Marc

[Advertisement]

Hallo,

Look here: Interesting Leica gear

[glenerrolrd]

Roger, I think with knowing how the AF works on the S2 you can manage it to some degree ... maybe ... in controlled situations. The question I asked (as I used the camera) is whether that is acceptable when shooting more spontaneously?

 

I understand your notion that the AF area is too large, however, Lloyd's point wasn't that ... it was that the camera consistently back focused at medium to far distances ... using more than one sample body and various lenses. It wasn't a matter of lens calibration as they all focused spot on using manual focus. His blog doesn't go into all the details, but his paid site does.

 

I've experienced oddly inconsistent AF results even with the subject filling the AF circle ... which I initially attributed to my error somehow ... but his report makes me now doubt that was the case.

 

On Dec.14th, 2010 Lloyd revisited the S2, a different body loaded with the latest firmware (a Leica USA representative went along with him), and he repeatedly demonstrated at will, that the very dramatic back-focusing issues remain. The illustrated sample showed the AF point and the actual AF point was well out of the AF circle many feet behind the intended spot. He has some shocking mouse over comparisons between the AF and MF results from the same scene.

 

This was not as easy to spot even with larger screen images than an LCD would show. However, when enlarged are obvious. His point was that some people may not notice it right away, but it is there and basically spoils the incredible performance of the ultra expensive lenses.

 

As reported else-where, I have also experience camera lock-up three times even with the latest firmware ... Lloyd also reported repeated lock-ups with an S2 that had the previous firmware installed. His was in August, so my original suspicion that it was the 20º cold during my shoots may be incorrect.

 

If nothing more, these kinds of reports serve as alerts when checking out a S2 and lenses.

 

-Marc

 

Marc

 

I now have an S2 and will soon have the 3 lenses that are released. While I have not been able to test to exhaustion the AF issue ...I can tell you its very real and as Lloyd describes.

 

The key learning is that the circle defines the area being evaluated by the AF system. In most situations it will nail the focus if the subject covers most of the circle . The accuracy when this happens is excellent similar to my D3S/X.

 

However the system can be easily fooled when a high contrast edge is found in the circle . The trees in the forrest is a good test . Start close to the tree and fill the frame then progressively walk back. The first error occurs when the edge of the tree has more contrast than the center. Slight bias toward back focusing. The second error occurs when a higher contrast background item enters the circle ...like a branch that is backlit.

 

This can easily be demonstrated with wicker basket . Put something in the basket that has low contrast and include just a little of the basket in the circle. Front focus on the basket.

 

This of course can create some erratic behavior making the AF unsuitable for certain shooting situations. On a N/C/S you have a smarter AF system and the option to use a very narrow spot .

 

The second situation are examples of very small targets at distances . The tree on the ridge. I did the same tests with a beach umbrella at distance. The sensor is not picking up enough edge contrast to confirm focus . Its works exactly as LLoyd describes . It is hard to see because the subject is so small . If you hear the system hunt ..forget it ..thats inadequate subject contrast to confirm focus. In my case I get the waters edge not the umbrella.

 

If the AF circle was smaller (more like spot metering for exposure) the photographer could control and would get fewer AF errors. But the AF process would require more skill because you could easily spot focus on a detail rather than the desired point. I believe you need both.

 

If the AF sensor reads the correct /desired point of focus ..its is exceptionally accurate ..but its the large circle and the lack of any interpretation of the content that will cause errors.

 

I also found that understanding the various AF settings ..especially if you use the real AF button and not the shutter ..is important..just like on a C/N/S you can shoot yourself in the foot in several ways . Follow focus works but is slow for example and its less accurate that using it without follow focus.

 

If AF is very important for the type of work you do ..then testing is a must . For weddings I completely get it ..for landscapes with complex detail ...I am scratching my head asking how is this important ? But then I shoot primarily M9 s and the S2 AF is much better .

[mwilliamsphotography]

Thanks Roger.

 

Now what is a bit perplexing is that the new S2P I just received seems more accurate shooting the same subjects as I did with the previous camera. Maybe I'm learning ... LOL!

 

I'll be testing it more this weekend, but of the shots I've done to date with the new body, it seems to be getting it right ... close up and far away ... with a few shots that should have confused the AF sensor.

 

I'll try a few of your suggested methods shooting the tree, etc.

 

What I need to do is shoot some people which is what the majority of the shots with this camera will be. If I can get the eyes right I'll be good to go.

 

-Marc