info@grayfoximages.com

§

Lately, Canonís leadership in the field of imaging sensor technology has undergone a bit of erosion.  Canonís early adoption of CMOS technology conferred an advantage that held for nearly a decade.  However, with the advent of Nikonís Sony-manufactured Exmor technology, featuring greater circuit density and allowing on-chip A/D conversion and noise reduction, Canon has fallen behind in at least one metric, namely low ISO dynamic range.  This has led to dissatisfaction among some Canon users, particularly landscape shooters, who perceive that high DR, high-megapixel offering from Nikon and Sony provide a capability that they would like to have from Canon.  This discontent has spilled over into Canon-specific online forums, resulting in some rather acrimonious sparring matches as high DR proponents voice their displeasure and doubters defend the status quo.

(Jump to 1D X Mark II update at end.  Our 1D X Mark II first impressions here.)

Internet Angst

Coming from an engineering background, in a domain that was at times inherently adversarial in nature, I learned rather quickly from experience that I could not get away with going to the podium with a brief that failed to recognize and deal objectively with the tradeoffs inherent in a contentious issue.  Of course, no such rules apply in Internet forums.  The Canon sensor DR question is a prime example.  One suspects that most photographers place sensor DR in the context of cameras as tools, with acknowledged strengths and weaknesses.  One can either acquire the brand of camera best suited to achieve one's goals or work within the limitations of what one has.  However, inevitably some feel compelled to lament the DR gap online -- despite the fact that no one within earshot can do anything about it -- and a vocal few respond to such criticism as if their camera brand was a sports team and they were its fans, obsessively cheering it on and shouting down any opposition -- at times irrationally, and even rudely, contrary to facts in evidence.  As a result, much of what appears on Internet tends to reside at one or the other of the extremes.  Finding balance amidst strongly held, polarized opinions is difficult, and forums are littered with angry posts that argue past each other. 

Briefly stated, advocates claim that there are situations where greater DR makes possible capture of scenes with a wider than usual range of light levels without resort to special techniques such as selective filtration or bracketing and HDR processing.  And, it also increases the degree to which shadows can be raised without exposing ugly noise.  In very round numbers, today's Canon's usable sensor DR measures near 12 stops whereas the latest Sony offerings provide 14 or more at base ISO, an advantage of two or more stops.  Plenty of objective measurements on Internet verify this, as well as the experience of respected photographers who have switched brands based on direct comparisons.  A few vocal Canon users vehemently condemn Canon for not responding competitively, and some suggest that Canon has lost its way with regard to innovation or is simply exploiting long-time customers with stale technology.

A great many naysayers resist the anguished cries of Canon DR advocates, for a variety of reasons, some of which are frankly, specious.  A frequent refrain is that any deficiency in results is simply due to a lack of skill on the photographer's part -- as if basic physics could be overcome by sufficient skill.  And then, there are the photo Luddites who claim that differences do not exist.  The latter may be safely dismissed, given the evidence.  Others respond, quite accurately, that the number of situations where increased DR is useful is relatively small.  Unfortunately, some go on to imply that, therefore, no one should care about it.  (This begs the question: what gives one photographer license to arbitrate the needs of another?)  Then, there is the group that responds that while Exmor technology provides an advantage at low ISOs, the latest Canon sensors still perform well at high ISO levels.  (True as far as it goes, but irrelevant to those who can make use of higher DR at low ISOs.)

Finally, many naysayers respond, eagerly taking on the role of unpaid Canon sycophants, that since Canon occupies the number one camera sales spot despite any technological lead Nikon and Sony may have, there is no need for Canon to respond with an Exmor-like sensor.  If the differences were critical, there would have been a mass migration away from Canon with the advent of Exmor technology. Even granting that high DR comes into play in only a set of limited cases, this argument ignores the large investment that many have in Canon gear.

Perhaps a good summary is that, to a first approximation, and depending on the responder's objectivity, viewpoints expressed can all too often depend on the type of photography the responder performs.  High DR does confer advantages -- it's a simple matter of physics.  But, it only applies in certain, wide light range situations.  After all, people lived for generations with slide film, which had a tiny range compared to current digital sensors.  Everyoneís requirements are different; hence people's perceptions, value judgments and reactions can differ significantly, especially if they are emotionally invested in the gear they own.  As for the Canon market leadership argument, sales have probably been impacted to some degree, but so far not enough to cause a change of corporate direction, at least not in a way that can be seen at present.  However, providing Canon with excuses from the customer base to delay transition to the latest sensor technology is hardly a sign of buyer acumen, is it?

Cameras as Systems

For my part, Iíve never seen the logic in turning down additional gear capability, features that would help me capture images in a wider variety of circumstances.  (Note that no gear on earth will make me, or anyone else, a better photographer in the artistic sense.)  Nor would I ever presume to tell anyone else what their gear or photographic requirements are.  Thatís their business.  Conversely, if someone tries to tell me what my needs are, well, weíve just parted company.  And if anyone starts tossing around know-it-all generalizations about the photographic community as a whole, they better quote a credible survey or panel of experts of some repute as evidence, else theyíre blowing you-know-what out of you-know-where.

To be more specific RE Canon sensors and the dynamic range question.  First, it seems certain that at low ISOs Canon is lagging behind.  To deny this is simply a sign of self-absorbed obduracy.  So, what then?  Why not abandon Canon and go with one of the alternatives?  Well, for a couple of reasons -- at least.  First, for most users invested in Canon -- or any brand -- a camera is much, much more than a sensor.  It is, in fact, a system, and not only a system but often a quite expensive investment, likely composed of multiple bodies, lenses, flashes and accessories.  There is a cost, in funds and learning time to change systems or to add another system.

Second, no system is better than all others with respect to all metrics.  Each has advantages and drawbacks compared to others.  Because of this, choosing a system is a balancing act, a matter of tradeoffs, and that is a very personal equation involving one's own anticipated uses, working approach and financial situation.  Back in 1999, when I decided to get serious about bird photography, and as a consequence to sell out my modest investment in Minolta, I made a detailed analysis of Nikon vs. Canon as I thought each brand might best serve my anticipated needs.  I looked at bodies, lenses and accessories; and I weighted how important each was compared to my personal use profile.  The result was that I chose Canon, a significant factor being their image stabilized long glass, although the 100-400 zoom was a factor, as were the then three TS-E lenses.  Since then, Iíve added professional tennis as another significant interest.  Had my interests been different, I might have chosen differently.

As it happens, I still have the same interests today, and if I were starting over again I would choose Canon once more.  The 1DX and Canonís latest generation super-telephotos are simply outstanding for both bird photography and fast-paced sports action.  With this focus, Iím usually far more interested in relatively low noise and decent DR at high ISOs than I am in higher DR at low ISOs, and the 1DX happens to be quite good in that regard. 

However, I do photograph other subjects, and on those occasions, far fewer in number, when I do travel or landscape photography, and when the lighting range of the scene exceeds the DR of my Canon gear, I occasionally find myself resorting to bracketing and post processing with HDR software. The limitations of this approach are obvious, and for this reason Iíd prefer a sensor with greater DR.  Unlike some, I have no problem acknowledging the weaknesses of the system Iíve chosen, and I can certainly wish those limits were removed without feeling defensive about my choice of gear.  Until then I'll continue to use that system to the best of my ability -- because the strengths of the system best meet most (not all) of my overall requirements, compared to the alternatives.  Naturally, in an ideal world, I'd like it all!  But assessing tradeoffs is necessary in dealing with what the marketplace has to offer.

If my interests were different, my gear preferences might be different, and I realize that others have different requirements, needs that are best served by other gear or other brands.  If my current (few) needs that are less well met with present gear formed a greater percentage of my overall work, I would simply purchase gear of a different brand as a supplement to what I already have.  (Fortunately, I can afford to do so.)  I neither condemn Canon for failing to steer their ship in the direction that I personally want, nor defend them sycophantically and irrationally, assuming that since the deficiency does not bother me it should not bother anyone else.

What Next for Canon?

This brings us to Canonís current product lineup and possible future innovation.  When Canon first adopted the EOS system, they innovated at a ferocious pace, and as a result they achieved a (I think, well-earned) position of sales leadership within their field.  At some point thereafter, in my opinion Canon became a bit more conservative.  This happens to many successful companies, and there's no need to second guess the reasons why.  But during this span, Nikon, which had played second fiddle for some time, began its own run of innovation, thereby regaining some lost ground in terms of technical features and capabilities relative to Canon.  Sales followed, at least in the prestigious high end market, and soon big black lenses returned in greater numbers to sporting events and birding and wildlife hot spots.  Exmor technology continues that trend.

Further, for a long while some accused Canon of crippling camera body releases in order to protect sales of higher end models.  Nikon, meanwhile, often pushed ďproĒ features further down in their product line, presumably pleasing a broader slice of buyers on the ďdarkĒ side.  And Nikon, in the view of many, exhibited superior AF tracking in the all-important flagship models.  But, with the release of the 1DX and the suprisingly well equipped 5DIII, Canon seems to have hit a sweet spot in terms of features and image quality that appeals to buyers, not counting low ISO DR.  In particular, AF tracking is no longer a thorn in Canonís side on these models.  The 6D and the innovative 70D followed, and the 7D MkII continues that trend, incorporating high end features such as a 65-point all cross-type AF system that includes center point AF with f8 lenses, 10 fps continuous shooting, and a 150K-pixel RBG+IR exposure metering system with scene detection.  Except for sensor DR, Canon appears to have cast aside the slow approach. 

One could contend that on the lens front Canon has been in the lead ever since they adopted the EOS mount and image stabilization technology -- except perhaps at the wide end -- leadership that has helped maintain Canon's position at the top of the sales charts.  Even with the wides their TS-E lenses are outstanding; and recent wide releases have been much improved.  In fact, one could suggest that heavy owner investment in Canon lenses dating from the time when Canon was superior in many areas of camera technology, along with continued Canon optics excellence, helps keep Canon owners locked in.  A September 2014 C/NET article dealing with the future of the Canon telephoto line suggests as much, quoting an industry analyst thus, "I think we're starting to shift away from camera body vs. camera body to optics vs. optics."  The advent of new generation AF solutions in the 1DX and 5DIII, and now the 7DII, hasn't hurt either.  At any rate, regardless of any perceptions regarding the pace of innovation, sales leadership has remained a constant throughout.

Finally, then, there is the matter of Canonís position in sensors vis-ŗ-vis Sony and Nikon.  Canon adopted CMOS technology early on, and in doing so gained an initial advantage over its rivals.  However, Canon has continued to polish the same apple endlessly while others have chosen to innovate rather than refine, and as a result Canon finds itself in a trailing position.  The reasons have been discussed endlessly and need not be examined here.  To catch up, Canon needs to, at a minimum, shrink their on-chip circuitry and place A/D conversion on-chip.  Why they havenít already done so is not something I care to speculate on, or, for that matter, am qualified to do so.

Open-ended questions include, when, if ever, will Canon do this?  And, how much will their failure to do so in the interim harm them in the market place?  Although there are voices of reason, Internet dialog, which is worth exactly the electrons it is propagated on, is filled with extremes regarding these questions.  At one end, there are those who say that there is no need for Canon to change because Canon cameras take great pictures despite the DR gap, which doesn't actually exist; and on top of that Canon is the world's leading camera seller even with the DR gap, which doesn't exist anyway.  So shut up and go take pictures.

At the other extreme, gloom looms like a baleful black shroud over DR enthusiasts, because, they contend, Canon is shamelessly exploiting its customers by continuing to foist technology on consumers that is no longer state of the practice.  Failure to close the DR gap is a sure sign that Canon is about to follow Kodak into corporate photography oblivion.  Their  position is grounded in fact, but the equating of sensor merit exclusively to high DR ignores other aspects of sensor performance.

As usual, the truth must be elsewhere.  Innovation in sensor technology has not yet leveled out, and one suspects that Canon will eventually be forced to field a solution that maintains parity with its competitors, just as it did in the area of flagship autofocus tracking.  There are no guarantees, and it may not happen until loss of sales to competitors becomes noticeable to corporate bean counters and/or feedback from professionals and respected experts reaches a threshold level that is, at present, opaque to outsiders -- a process that is iceberg-melting slow compared to Internet forum dialogs.  Or it may be that the process is paced entirely by the economics of transitioning from one fabrication technology to another.  The viability of patents -- Canon's own and the relationship of theirs to those of competitors -- may be a controlling factor.

Until a solution appears, Canon users who require higher DR solutions at low ISOs will have to either bracket and use HDR software or go with a dual brand solution -- or just change brands.  Not fun, but the reality one must deal with -- the sad reality of large corporations that achieve leadership through innovation and then all too often exploit their lead by sitting on it.  Gripping about it on Internet forums, or belittling those who do, certainly wonít solve anything!

Update:  May 2016 -- Canon delivered the the latest iteration of its flagship pro camera, the 1D X Mark II, and it is reported that the camera's 20.2MP sensor incorporates A/D conversion on chip, thus finally addressing one of the technical gateways to increased low ISO dynamic range and better overall shadow noise performance.  Early reports are that while low ISO DR is improved Canon has not quite caught up with the best of its competitors.  At the other end of the scale, high ISO noise seems not at all improved over that delivered by the 1D X.  Shadow banding seems finally to have disappeared, but clearly more work remains to be done.  However, Canon has finally taken a critical step forward on the path to modern sensor development.


§

If you would like to express thoughts on this subject use the link to send an email. © 2014 Michael W. Masters
Notice: All images and written material on this web site are © 1999-2014 Michael W. Masters. All rights are reserved under US copyright laws. Images may not be downloaded or otherwise used without written permission of the artist. Written material may be quoted under fair use so long as attribution is given.