Hi,
Most users are pretty much ignorant about any issues their equipment has. For sure, I am. Some lenses are better than others, bayonets and sensor are not in alignment, and so on. Very clearly, BC (James Russel) is not very much about pixel peeping, he is pretty clear on that. But, if you want to do really large prints some pixel piping may be helpful. Critical users test five lenses and cherry pick the best one. I am not there.
I have seen Eric Hiss comments on Lloyds testing, but I would suggest that Eric my be slightly partial. It may be that Eric doesn't understand testing. For instance, he uses two stages of sharpening before calculating MTF using Imatest. Serious users measure MTF without sharpening. Why, because with sharpening you can achieve any result you want. The best way to do this is to use a raw processor that applies no sharpening.(Now, he who is without sin may throw the first stone. I have also used some sharpening in many of my tests. But I now realise that correct testing needs to be without sharpening, not least because raw processors apply different amounts of sharpening for different systems.)
BTW, have you seen any raw image posted by Mr. Eric Hiss? I strongly feel that we need to see raw images before jumping conclusions. I would say that Lloyd should also release raw images. The best source of raw images now is Imaging Resource and DPReview but they don't test many high end MFD systems. Personally I
try to post raw images when it may be
relevant. I have posted well over one hundred.
Just to say, I have read both Eric's comments and Lloyd's test pretty carefully. I would be clear, Lloyd would be better not publishing that test. Why? Because he did not spend enough time with the camera. Would more time with the camera affect the review? I don't think so. Very clearly, he would be able to do more tests, like the church mosaic at Stanford. With the Pentax 645z he spent several months and so did he with several samples of the Leica S2.
But, the tests he has done show issues that only show up in correctly made test. Very accurate detail in on focus detail. Aliasing effects that only arise with accurate focus and so on.
Have you read the test and looked at all of the images? Did you analyse them at actual pixels? If you did not, you actually don't know what you talk about. Focusing may matter little, you find the part that is actually in focus and analyse surrounding detail. There is always a curvature of field, that is the reason Lloyd shoots 3D subjects.
Just to make a point, Lloyd complains about AF accuracy on almost all systems. Canon 5DII, Nikon D800, Pentax 645D, Leica M, Sony Alpha 900 just to mention a few. Truth is simply that AF systems don't used to be that accurate. It may seem that some recent generation AF systems combined with some recent generation lenses may be accurate. (Roger Ciala at LensRentals has an article on that).
Getting back to focusing accuracy, one of the frequent posters here shooting an MFD back on a Linhof uses a 30X loupe for accurate focusing, but still feels the need to stop down to f/11 for optimal sharpness. On the other hand he stops down to f/16 to eliminate aliasing.
So, what I say is that Lloyd Chambers is a bit over-analytic, but he may know what he is doing.
Personally, I feel that manual focusing with magnified live view is the only way to achieve exact focus. Contrast Detect AF on sensor is a good alternative.
If you feel that Lloyd Chambers is less then competent, please post some proof demonstrating the accurate focusing of at least two samples of the Leica S under demanding conditions.
These two articles are a couple years old but they may still have some relevance for today's systems:
http://www.josephholmes.com/news-medformatprecision.htmlhttp://www.josephholmes.com/news-sharpmediumformat.htmlBest regards
Erik
Ps. Just to explain, Lloyd Chambers prefers magnified live view as focusing method. He found, however, that the Sony A7 series do correct AF using contrast detection AF (CDAF). Similarly, Roger Ciala has found that CDAF does match manual focus at live view. This makes a lot of sense, as CDAF uses the actual signal from the sensor instead of the signal from an AF device that has two moving mirrors in it's optical path.
I would expect that Roger's findings apply to all CDAF systems, they are accurate but slow and they may miss correct focus at all.
Not to play devil's advocate, but all that indicates is precisely nought other than, possibly, a lack of competence on Chamber's part. Cooter uses this camera extensively and has never reported such an issue - nor has Tom Munro, to name but two, (if you don't know who Tom is, try doing a google image search). I could add more , these are professional photographers who make a living out of using cameras. I could name at least a dozen others.
Eric Hiss also, IIRC, spent a day with him and wasn't impressed by his testing procedures.
Were you meant to and again even if you were and did, what does that prove ?
