dante stella stories photographs technical guestbook

 You were the Chosen One! You were supposed to destroy the Sith, not join them...Bring balance to the Force, not leave it in darkness!

—Obi-Wan Kenobi to Darth Vader
Whither Micro Four-Thirds?

This site originally panned the Olympus E-PL, concluding that it seemed (right off the bat) like a first effort that could have been improved.  My complaints about the E-P1 were that for all of the claims that Micro Four-Thirds (43) would allow tiny cameras, the E-P1 was large, heavy, slow and somewhat unwieldy to use - all to an extent that was difficult to justify in light of the sensor it was carrying.  Maybe my reaction was over the top - and it certainly set off the E-P1 fans on dpreview.com – but history does demonstrate that the E-P1 went down in flames within five months of its release.

One striking thing about product offerings in Four-Thirds and 43 is the  proliferation of camera models with almost no low-level specification upgrades.  This has been true across manufacturers - and even within some.  For example, the Olympus E-P series (43) has included four models in two years with very little differentiation in capabilities:

  • The EP-1 was introduced in June, 2009, and its first successor model, the EP-2, came out that same November - with EVF capability.
  • The second successor, the E-PL1, came out in February, 2010, with a built-in flash, a lower cost, and more plastic (that's not necessarily bad). 
  • The E-PL1s was announced in November, 2010.
  • The E-PL2 is slated for announcement in January 2011 (though this has not been confirmed).
  • This short attention span in 43 is much more reminiscent of old Sony (i.e., pre-Konica-Minolta) than Olympus at any time in the past.  For the digital Pen line, for all of the model variation, there has not been any significant upgrade of the 43 sensor - the most fundamental determinant of image quality. Panasonic, for its part, has also been proliferating models at a fairly rapid pace, running through the G1 (9/2008), G2 (3/2010), GF1 (9/2009), GF2 (11/2010), GH1  (3/2009) and GH2 (9/2010). At least with the GH2 (the sixth of six models), Panasonic increased the pixel count to 16 million from 12.1.  Olympus seemingly has been stuck at 12.3.

    For comparison of Olympus product cycles, the Pen F from which the E-P1 takes its design cues (and its marketing) was produced from 1963-1966, and the metered Pen FT was produced from 1966-1970.  Even the original OM-1 was produced for six years -  1973 to 1979.

    All of this running in place is beginning to beg the question of whether (1) these two remaining 43 manufacturers are stuck in a cycle of churning product (with continual and minor improvements) to maintain their own relevance or (2) whether desperation is setting in for the 43 (if not also the greater Four-Thirds market).

    Even before 43 hit the market, there was a huge number of Four-Thirds products from Leica, Panasonic, and Olympus that all revolved around (1) a common, "open" standard* and lens protocol; (2) relatively undifferentiated sensors (Kodak and Panasonic); and (3) relatively undifferentiated camera body specifications.  If you're a student of economics, you see where this is going: an open standard coupled with undifferentiated products makes you a price-taker in a perfectly competitive market.  That is seldom a recipe for massive profitability.

    *Per the Four-Thirds consortium ("Consortium") web site, the basic standard (whether Four-Thirds or 43) is available to camera manufacturers on a nondisclosure-agreement basis and cannot be provided to individuals or researchers.  That sounds like an amazing (if not intentional) barrier to entry. It is also a little bit strange to advertise a standard as open, yet patent it.

    Although there was much ballyhoo about Four-Thirds (and 43) being "open standards," the reality is that only two companies actually manufactured cameras (Leicas were reworked Panasonics), and Olympus made and sold 14 out of 17 Four-Thirds bodies (by model count).  Four-Thirds is now identifiable as an Olympus drum solo, and 43 is an Olympus/Panasonic buddy picture.  Sigma is a fair-weather friend, mostly adapting APS-C lenses but making a couple special Four-Thirds models.

    It is this writer's perception that Four-Thirds failed to set the world on fire, and 43 has managed to prolong the agony.  The original Four-Thirds format had its challenges: a 4:3 aspect ratio that (1) didn't really conform to the norms of 3:2 for professional SLR systems (be they APS-C or 36x24mm), (2) was designed to match TV screens that were being phased out in favor of 16:9, and (3) required (and requires) the sacrifice of megapixel real estate to get to a conventional aspect ratio (for stills or for film). 

    The one thing Four-Thirds did have at the beginning was a range of relatively compact bodies (like the E-Volts with their porroprisms) and lenses that were scaled appropriately to a sub-35mm sensor.  This was better than what initially happened with Nikon and Canon: huge lenses throwing excessive image circles for APS-C sensors housed in huge bodies (for example, the Nikon D1).  But today, Four-thirds seems to be stagnating (you know it's bad when the gladhanding dpreview.com calls the E-5 a "warmed over" E-3).  And even for 43,  where there is arguably more action, only niche players are involved.  Olympus is the clever but marginally important manufacturer that tried this before with the half-frame Pen, and Panasonic has no history in system digital cameras outside Four-Thirds. If Panasonic is making the sensors, every system will be limited to what Panasonic wants to make or what Olympus will pay to develop.  Beyond that, there is only so much you can do with processing, AA filters, and tweaking the way contrast detection autofocus works.  Look at the Nikon D3x and the Sony A850, which share common sensors - and there is not as much a difference in image quality as the $6,000 price difference might suggest. 

    But product offerings might be a completely superficial way to tell where the format is headed. Perhaps the promises and delivery are a better measure of whether the format has any staying power?

    Micro Four-Thirds
    The smaller sensor size makes possible smaller and lighter camera bodies and lenses. In particular, the Four-Thirds system allows for the development of impressive f/2.0 zoom lenses, which would be prohibitively heavy, expensive, and difficult to design for larger sensor formats.

    — Wikipedia (probably cribbed from Consortium literature)
    Camera bodies were still comparable to small 35mm SLRs. 

    The f/2.0 zoom lenses came to pass, but they were just as big as the f/2.8 zooms sold for FF cameras.

    No f/2.0 zoom lens is available for less than $2,000 new - making them more expensive than their Nikon and Canon f/2.8 equivalents.

    The huge high-ISO advantages of FX cameras like the D700 wipe out any advantage of having an f/2.0 maximum aperture on a smaller-sensor camera.
    Camera bodies are still fairly comparable to  compact 35mm film cameras of the 1970s and 1980s.  An E-P camera, by way of example, is comparable in size to a Leica M.

    Lens issues are the same.
    Telecentric optical path means that light hitting the sensor is traveling perpendicular to the sensor, resulting in brighter corners, and most importantly improved off-center resolution, particularly on wide angle lenses.

    — Same
    This makes both cameras and lenses thicker - easily comparable to full-size APS-C or FX cameras.
    No change, though to maintain the same telecentricity, lens size and exit pupil distance must remain the same as the larger Fourt-Thirds cameras.

    Telecentricity was overcome by Leica with the M8/M9 and the upcoming Fuji X100: it is a question of making offset microlenses.
    Because the flange focal distance is significantly shorter than most mounts (such as Canon FD, Canon EF, Nikon F and Pentax K), lenses for many other SLR types, including the old Olympus OM System, can be fitted to Four Thirds cameras with simple mechanical adapter rings. (Such mechanical adapter rings typically require manual setting of focus and aperture).

    — Same
    Not a unique attribute of Four-Thirds; any thin body theoretically could do the same.  Many legacy lenses not worth using due to difficulty of focusing, poor optical performance on a small, dense sensor.
    Same (e.f. Sony NEX)
    A smaller sensor makes it easier to achieve a deeper depth-of-field, when needed, reducing the risk of out-of-focus photos.

    — Same
    A somewhat perverse marketing claim, since shallower depth of field long has been the holy grail for the Japanese industry; if better depth of field is the goal, even more compact cameras are preferable.

    "A reduction of about 50% in flange back length without compromising the high picture quality of Four Thirds System"

    — Consortium
    The flange-back distance going from 38.67mm (almost as much as a Konica 35mm SLR)  to 20.00 mm does not make much of a  difference if telecentricity must be maintained.

    Flange-back distances are only one dimension of camera size and are only really meaningful when using a "pocketable" lens, of which three exist for 43.

    "Approx 6mm reduction in lens mount diameter and additional of two signal contacts for compatibility between mobility and reliability."

    — Consortium
    This has some influence on lens size, particularly if the barrel size tracks the mount size.  Ok, as long as you don't try to adapt standard Four-Thirds lenses.
    "Soon users will be able to switch easily between shooting still images and movies using natural, intuitive operations while keeping the useful Live View on the monitor screen."

    — Consortium
    This will be intuitive because 43 is intended to deprive users of a separate viewfinder (see the Four-Thirds site for the statement on "shooting freedoms beyond the viewfinder."

    First, someone had better tell Panasonic that 43 cameras shouldn't have viewfinders, since both the GH1 and GH2 have electronic, eye-level ones.

    Second, since many Four-Thirds cameras have both optical viewfinders and live view, exactly what freedom does 43 confer?

    "A system that allows continued use of existing Four Thirds System Lenses."

    — Consortium
    With the required adapter, you are now up to almost exactly the same overall size as a regular  Four-Thirds camera.

    Existing Four-Thirds lenses would have to be fairly compelling, particularly after the cost of an adapter that maintains automation.*

    *When a Four Thirds lens is mounted on a Micro Four Thirds body using an adapter, all of the functions of the Micro Four Thirds System may not always be available." (Consortium)

    There is nothing blatantly untrue about the claims of the Consortium, but nothing inherent to the Four-Thirds or 43 systems would cause someone to choose one of these formats over APS-C or FX. And the deployment of faster and smaller zoom lenses, the really interesting thing at the beginning of Four-Thirds, is far less compelling now given their cost and the massive progress in everyone else's sensors. Canon, Sony and Nikon have sunk gigantic amounts of money into improved high-ISO performance - and it shows.

    At the end of the day, we have a number of reasonably capable Four-Thirds and 43 cameras.  We have all seen some fantastic work done with them.  But query whether any variant of the Four-Thirds system (regular or micro) has the qualities that will make it a long-term contender in the marketplace. Several things agitate against it:

  • Only one company is designing imagers for it, putting all of the eggs in one basket.
  • There is no completely competitive low-light sensor available in the format.
  • The only way to amortize the cost of a new sensor is in proliferating models based on the same imagers.
  • Designers of third-party lenses are much more incentivized to "port" APS-C designs than make new formulae for Four-Thirds, leading to some strange and interesting fields of view.
  • No volume producer of professional equipment has ever signed on to the standard. Fuji and Kodak are relatively marginal players even in the point-and shoot market; Leica quit making Four-Thirds equipment; Sanyo has yet to produce a successful mass-market camera; and Sigma likes APS-C sensors for the DP-1 (and upcoming SD-1).
  • Even the secondary use of 43 cameras (mounting legacy lenses from other systems) is being taken over by APS-C platforms like the Sony NEX.
  • It may not be time for a high-low on the survival of Four-Thirds and 43, but things are not looking too good right now. It is possible, however, that 2011 will bring some quantum leap that will make 43 an attractive and viable format.