AF microadjustment - SpyderLensCal

[keith_cooper]

Is this microadjust thing just a big con ...

Well, far be it for me to point out the occasional 'Emperors New Clothes effect' that seems to accompany quite a few photographic items.  Such items can also build up their own vociferous fan clubs... ;-)

I found it useful to adjust (without the benefit of additional hardware or expense) my 1Ds3 and some lenses, but to be honest I'd not previously noticed any problems.

So, useful - yes.
Essential - probably to nowhere near as many people as might like to think it is ;-)

So, if I have a 1Ds4 by the end of the year, will I check my lenses - yes; will I make it a top priority before actually using the camera to take some pictures - no...

[Michael Tapes]

Keith,

Of course there are photographers who are "over concerned" about their gear (not that there is anything wrong with that). But the fact is that quite a lot of body/lens combinations are simply too far "out of tolerance", to meet the sharpness needs of many photographers. For proof of this I simply point out that EVERY manufacturer of DSLRs now has a Focus Fine-Tune feature in at least some of their cameras. Why would they do this if there was not a problem. I could see ONE mfg doing it, but ALL? Surely when Canon (the first to put in this feature) introduced Micro-Adjustment, the best thing Nikon and the others could have done was to simply state (hopefully truthfully) that their stuff "works" out of the box, and that is why they do not have or need this feature. But no...every mfg added it because they saw it as a needed thing. I can assure you that they would not do this casually, as their support lines are asked about micro-adjustment a lot and that support costs a lot of money.

Anecdotally, I have countless customers that have told me how much better their gear performs after their use of LensAlign (whether they could have adjusted with another method is irrelevant to this part of the discussion).

And, factually, the genesis of LensAlign (the need) came from 2 observations on my part. One was how frustrated people seemed to be (on the forums) about their lenses front/back focusing, with no relief from the mfgs. The equipment went back and they were told that it was in spec, or they were told that something was adjusted, but in either case the owners could not verify the results, because they did not have a systematic and repeatable approach to testing the performance of the AF. The other observation was in my own photography (Canon 1D at the time) and being very unhappy with the sharpness I was achieving (or not achieving to be more accurate). In the end I tracked it down to the AF system.

So LensAlign was not a child "gizmo" for the micro-adjustment crowd. Micro-adjustment did not exist when I began the LensAlign project. The purpose was simply to provide a repeatable methodology for people to test for back/front focus errors, as a help for discussion with the mfgs about their issues, and a validation (or not) when equipment came back from the service centers. Towards the end of the LensAlign design process Canon introduced Micro-adjustment, and I was immediately on the phone with Chuck Westfall telling him of my soon to be released product, and had many discussions with him to keep him in the loop and gain his insights for the final design phase.

I have also done many demos with high level Canon (and other mfgs) personnel present, where of course I discussed the issues of front/back focus. At the conclusion of these demos I invited the people from the camera companies to correct any mis-statements I might have made. Never has any camera maker disputed the existence of the front/back focus issues or contradicted any aspect of my presentations.

I believe that the issue (con or no con?) stems from this...

A specific camera/lens combination can perform "in spec" but still have a front/back "issue". By issue I mean the following...let's say that at a given distance the combination has a DOF of 6 inches. We will call the range -3 inches to +3 inches. If the camera/lens focuses anywhere within that range, it is generally considered to be in spec. That means that the point of sharpest focus could be at the -3 point with all 6 inches of DOF towards the back, or at + 3 with all of the DOF towards the front, or anywhere in-between. Many photographers are not happy with that wide range. They want there point of specific focus from the AF system to be centered as it should be for optimum overall performance (yes, I know that most people think that DOF is 1/3 front and 2/3 rear, but that is only true for distances as they approach infinity).

Canon says it like this...

"AF precision is adjusted for the camera and lens to fall within
the lens' maximum aperture's depth of focus. However, there
are users who want to adjust it more minutely. They have
had to go to a Canon Service Center to have it done.
AF micro-adjustment is a feature developed for these users.
The user himself can now finely adjust the AF focusing
position. The adjustment range is ±20 steps in front of (-) or behind (+) the point of focus."

I have spoken at length to service centers and rental companies, as well as a few :>) photographers. Front/back focus issues are real. Some can be "fixed" using the AF fine tune in the cameras, some can only be fixed by the service center, and unfortunately some cannot be fixed. And of course,  the majority of camera/lens combinations work just fine out of the box. But the minority group is large and affects a lot of people.

I hope that the above adds to the discussion of the "con" aspect of AF adjustment in DSLRs.

Michael Tapes

[keith_cooper]

Couldn't disagree with your observations on the causes (I have an engineering background long before I took up photography for a living)

It's just some of the wailing (and I choose the term deliberately) that I've seen over the lack of AF adjustment on the 60D suggests that some have bought into the idea of a tech fix for some perceived deficiency just a little more than might perhaps be warranted by actual need.

I'll use the term 'a tech fix' to describe any way of performing the adjustment - no preference to method implied BTW

Tomorrow I have a room full of people, in the property survey business, who want to learn to take better photos. The most difficult lesson for most to learn is that better photos come from looking and thinking about what they are doing, not just buying better kit.   I do enjoy the teaching side of our company's business, since like the articles and reviews I write, you can't really get an idea across until you've given it a fair bit of thought yourself.

[Michael Tapes]

...  some have bought into the idea of a tech fix for some perceived deficiency just a little more than might perhaps be warranted by actual need.

I concur. It seems like in this day of technology, so many of today's enthusiast photographers are looking for the magic bullet that some how will make their pictures "as good" as what they see from the gurus. It could be a lens, or a plug-in, or LensAlign, but what they have to realize is that there is nothing that replaces the knowledge of the basics. As said so many times, a great photographer can take a great picture with any camera, and an uneducated photographer will continue to take mediocre photographs even when handed the finest camera (or gadget or whatever).

While I am not thrilled to be getting much older than I would like, I am very pleased that in my years that I learned by shooting with a 4x5 Speed Graphic, and working in a wet darkroom, and in my pro audio life editing recording tape with a razor blade and splicing tape (hell...we even edited 2-inch 16 track masters!), and on and on (oh...the good old days! :>)

Cheers..
Michael Tapes

[deejjjaaaa]

Micro-adjustment did not exist when I began the LensAlign project.

I am not sure about Canon cameras, but the very low end and cheap Pentax *ist DL had AF adjustment in its firmware since 2005 - you had to press a certain combination of buttons while switching the camera on to get into its service menu... granted it was not described in user manual, but still it was there and it was in 2005

[Michael Tapes]

Yes...I was told later that Nikon and Canon had these in their service menus as well. But totally unaccessible from the user menu system.

[Mike Arst]

Is this microadjust thing just a big con for not making the thing right in the first place.

You could look at it that way. Then again, how much would you be willing to pay for a camera/lens combination so perfectly "tuned," and for a zoom lens of such high quality, that auto-focus is perfect first time and every time and at every focal length? In the days when dinosaurs walked the earth, nobody'd be caught dead using auto-focus gear. That stuff was slow and not reliable. The technologies have advanced to a remarkable degree now. For someone like me with aging eyesight, who would once rather have stayed home than be seen using auto-focus: auto-focusing is a complete god-send. Micro-adjustments, likewise.

When I had a 5D I'd-a killed for the ability to adjust auto-focus with a 50/1.2, which I'd found "previously owned" at a decent price at a local store. Always wanted that lens! I got it home and found right away: at distances where I knew I'd typically use it, focus was off by damned near an inch. Awful. Damn. Canon (which at the time was not admitting such a problem with the lens despite complaints) said they'd make the adjustment if I'd send the lens and camera -- and on my dime. But they were rumored not always to do a good job of this. Would I end up having to do it all over again? I decided instead to take the lens back to the store for an exchange. Sigh. So as far as I'm concerned, micro-adjustment is a "glass is half full" feature -- especially considering the loss, with d-SLRs, of truly bright viewfinders and decent microprism focusing screens. I rarely focus manually now...

my 70 -200 is sharp across the range, on the 1Ds it often missed focus at the 70 mm end.

When I had the 70-200/4.0/IS, with the 5D it never missed focus at any focal length. It must simply be a case of variation among bodies (or particular lens+body combinations). So IMO, a micro-focus adjustment feature is your friend and not a "big con."
 

[digitaldog]

To once again prove Datacolor has no new ideas and would rather steal those from others, check this out:

http://blog.david-kennedy.com/2010/09/22/datacolor-immitation/

So David Tobie, care to let us know other than undercutting the price, what competitive advantages you guys have come up with this time?

If this one the case of a photographer’s work, the photo community would rightly be up in arms with pitchforks. A similar tact in terms of ignoring DataColor seems a rational response to their total lack of innovation (and making up BS terms to unsuspecting customers like spectrocolorimeter. Shame on them!

[keith_cooper]

As someone who has looked at pre-production software for the SpyderCheckr I would just point out that the Passport is used to make DNG profiles, whilst the new Datacolor product makes adjustment sets (based on the full range of coloured patches), so if anything, it is much more akin to Thomas Fors scripts or the adjustment techniques publicised by Bruce Fraser.

I'm out of the UK from this weekend for a few weeks, so won't be able to give the production version of the device and software a fair test until later next month, in the similar way I looked at the Passport before it was launched.
http://www.northlight-images.co.uk/reviews/photography/colorchecker-passport_1.html

I'm minded to think you are being somewhat unfair and unnecessarily personal with your response. Personally, I make a deliberate point of not selling any hardware or software, and quite deliberately not to set one product off against another - I've found that all too many comparisons you see in 'reviews' tell you more about the personal choices, preferences and biases of the reviewer than represent a methodical (and useful) comparison. I generally trust my readers to make their own informed decisions and am always prepared to discuss anything I've written.

Out of completeness, I should also mention that I do pre release evaluation and testing for a number of competing companies in the printing and colour management fields. I take it as a personal 'pat on the back' that such companies are happy to know that I work with their competitors and trust my integrity in saying what I really think of their products.

[digitaldog]

As someone who has looked at pre-production software for the SpyderCheckr I would just point out that the Passport is used to make DNG profiles, whilst the new Datacolor product makes adjustment sets (based on the full range of coloured patches), so if anything, it is much more akin to Thomas Fors scripts or the adjustment techniques publicised by Bruce Fraser.

A DNG profile is clearly a better route if you understand the differences in the processing paths using it versus updating HLS sliders (which WAS NOT) what Bruce and Thomas were doing. The were updating the calibration tab which like a DNG profile takes place at a different and arguably better area within the raw processing pipeline.

I'm minded to think you are being somewhat unfair and unnecessarily personal with your response.

Granted. It would be useful to see DataColor actually come up with something new and from their own brain trust rather than (and this is a personal opinion) rip off others intellectual ideas and perhaps property. I think no less of this when a photographer uses another idea for an image. It rubs me the wrong way. Where’s the innovation?

Personally, I make a deliberate point of not selling any hardware or software, and quite deliberately not to set one product off against another

I fall into the first part of that camp but because of it, not the second. I don’t sell anything so I can pretty much say whatever I want. Its a refreshing lifestyle, you might want to give it a shot <g>

[keith_cooper]

"...you might want to give it a shot <g>"

No thanks - I know more than enough about how much work it takes to provide rigorous, accurate -and- relevant comparative reviews. I quite like not going out of my way to be mean to people :-) :-)

I'd not be happy with the classic "we gave this 87% for X" approach and since my main job is (by choice) as a photographer, I just don't have the time (or temperament) to devote to such quantitative analysis :-)

[keith_cooper]

Since we're on the topic, here's a tutorial for a DIY clone I did a while ago...

Excellent suggestion!

Interestingly enough, I received a brand new 24-70 from Canon this morning (lack of parts to fix the old one it seems)

Since I'm just off on a trip and had the Datacolor device around, I tested the lens by a number of means.

At 50x FL distance I couldn't be sure of much other than it was sharper than the CPC loan copy (Yay!). I suspected that a slight adjustment might help, but the difference was pretty minimal.

At 20x FL there was a peak in sharpness at -2 ...not at all easy to see and the ruler is of no great help.

back at x50 I could see that -2 was indeed slightly sharper than 0, so that's what it's set to on my 1Ds3.

Then I tried the procedure again with a deliberate target misalignment of a few degrees - no difference whatsoever in the results - just as imprecise :-)

My suspicion is that a lot of people expect the differences to be a lot more visible than they actually are, and since they are expecting precision, are somewhat easily influenced by people telling them how precise their methods and setup need to be. I can see people trying it out, not getting a pleasing result (i.e. everything looks better and their photography improves) and deciding that it must be that they need better quality camera/lens/accessories etc

Having helped set up precision optical testing equipment in the past (mirror testing) I just couldn't equate the accuracy needed there with the somewhat imprecise methods we're using with various equipment here.

Let us hope that there is a way of auto calibrating AF with a target (using phase and contrast AF) available in cameras before too long - it was suggested to me that this function was already in the service menus of a number of current cameras, although I've no evidence for this.

The more testing I do, the less I see current AF adjustment as a mass market feature. I know a some people are annoyed over lack of the feature in the Canon 60D, but I suspect Canon knows how much real use is made of such features in their target audience.

Of course none of this makes much difference when marketing stuff, but the business of selling photography equipment is something I make a point of avoiding ;-)

[Michael Tapes]

here's a tutorial for a DIY clone I did a while ago.

It is amazing to me that that photographers that are so very concerned with the copyright and copying of their work (as they should be), see no problem in copying and disseminating the intellectual property of others. Not even a mention of the designer, which at least would make it polite. Methinks a double standard.

Michael Tapes
Designer: LensAlign

[digitaldog]

It is amazing to me that that photographers that are so very concerned with the copyright and copying of their work (as they should be), see no problem in copying and disseminating the intellectual property of others. Not even a mention of the designer, which at least would make it polite. Methinks a double standard.

Agreed! And even if I could build the product for $20, there’s MY time! By the time I put this together, I’d have lost money compared to just purchasing your product, supporting your endeavors and having someone to contact (praise or blame) if there were issues or support needs. But you are right, the idea of “don’t do as I do, do as I say” in terms of intellectual properly is most egregious.

[keith_cooper]

Agreed! And even if I could build the product for $20, there’s MY time! By the time I put this together, I’d have lost money compared to just purchasing your product and the hours of enjoyment I'd get

Perhaps not everyone wishes to cost their time at your rates :-)

Personally I'd be be happy with a pattern printed onto paper and nailed to a wall - If you were very careful you could use the nails to ensure optimum alignment - not that I actually believe it to be nearly as essential as some might suggest...

[digitaldog]

And Andrew, though I appreciate your photographic expertise and have found it very helpful in the past, I do not appreciate your piling on here.  Let's first establish whether there is any actual IP infringement going on before pointing fingers, shall we?

I never said a word about IP infringements. It probably is all very legal. That doesn’t mean it doesn’t leaves a bad taste in my mouth that this one company in something like 60 days released two products that appear to be based upon existing products and in the case of the LensAlign, output a high level of marketing BS and who’s mouthpiece when asked about the competitive advantages here (which he brought up) snuck out of here without a peep on the subject.

My comment on intellectual properly was aimed not at DataColor but at photographers who complain about copyright or others who use existing ideas then buy products from a company that basically practices the same game. It may all be perfectly legal and business as usual. Doesn’t mean its kind of distasteful.

[Mike Arst]

I have found this message thread extremely inspiring. After reading through it a couple of times, I've realized that this whole design-and-engineering-and-manufacturing thing is just a ruse -- a ploy by The Man to take our money. It's just evil.

Therefore I have decided to stop buying all manufactured photographic products. Henceforth I will be making my own cameras and lenses from spare parts obtained easily at the Dollar Store. According to The Internet, you could have an entire professional photographic system, possibly even including a 2000 w/s studio strobe and a print signed by Phil Marco or Janet Reno or Chopin or some other famous person, for maybe $75 or $100 (not including the calipers), which would probably include having to buy some of those unrecognizeable little striped tubular thingees with wires coming out of them hanging in plastic bags on pegs at the back of the Radio Shack. Thyristors or whatever they're called. Plus the calipers, of course.

[keith_cooper]

Just me or does this thread seem to vary every time I look at it, so I have responses to messages that aren't there?

I see that this vanished, for example  DIY focus

Perhaps I'll now get a crowd outside my own house with pitchforks and burning torches...

[pegelli]

>>I guess DataColor was unaware that LensAlign has been in the market for almost 2 years.

You did say Marketing Materials. Let's just say our marketing team are not all aware of it. Oversimplified marketing is a fact of life. Those who have heard of LensAlign and other solutions will take this with a grain of salt. Those who have not are new to the field anyways.
 

I must say that I have no respect for a company that hides behind their incompetent marketing team rather than apologises for it.
My money goes elsewhere

[AFairley]

Because of the negative reaction of some forum members, I have deleted my post with the link to the DIY LensAlign clone, along with my subsequent post explaning why the clone does not infringe on Mike's or LensAlign's legal intellectual property rights in any way.  I find the occasional pissing matches that flare up on the forum from time to time tiresome, and I have no wish to be part of one.