So I’m not sympathetic towards certain Pentax users, who bleat that they can’t make adequate photographs—not until the company delivers a DSLR with a 24×36 mm sensor. (With pre-Photokina Pentax rumors swirling, the issue is once again a hot topic for discussion.)

Full-Frame Pentax?

Yes, many long-time Pentaxians do have investments in full-frame “FA” lenses, ones that could cover the format.

But keep in mind, over the history of photography there have many camera mounts which died out entirely—obsoleting numerous fine lenses. How about Contax/Yashica mount? Minolta MD, or Canon FD? (Or even earlier systems, such as Exakta or the superb Zeiss Contarex?)

By contrast, any Pentax lens made since 1983 can at least be used on a modern DSLR—metering automatically, and with the benefit of built-in shake reduction. So the handwringing that some great lenses have been “orphaned” by Pentax can get a bit exaggerated.

Some point out hopefully that not long ago, Pentax introduced new telephoto lenses bearing the FA designation. Ergo, full-frame is coming!

But truthfully, no decent lens design even breaks a sweat in covering an image circle much smaller than its focal length. With a telephoto, full-frame coverage is basically “free”—so why not go ahead and use the FA labeling? It allowed Pentax to finesse the format question for a while longer; and maybe even sold a couple of extra lenses to Pentax film-body users.

I do agree with the complaint that the viewfinders of most consumer DSLRs are puny and unsatisfying. The view through the eyepiece of a full-frame body is noticeably more generous.

However, the electronic viewfinder of an Olympus E-P2 neatly solves this problem too—at a price savings of about $1400 over today’s cheapest full-frame DSLR. And the EVIL wave is only gathering momentum.

The perpetual cry from full-frame enthusiasts is that Moore’s Law will eventually slash 24x36mm sensor pricing. To me, this seems like a wishful misunderstanding of the facts.

The shrinking of circuit sizes which permits faster processors and denser memory chips is irrelevant to sensor design—the overall chip dimensions are fixed, and the circuitry features are already as small as they need to be.

Also, figuring the costs of CMOS chip production is not entirely straightforward. It costs money to research and develop significant new improvements over prior sensor generations; it’s expensive to create the photolithography masks for all the circuit layers. Then, remember that all these overhead costs must be recouped over only two, perhaps three years of sales. After that, your sensor will be eclipsed by later and even whizzier designs.

Thus, there is more to a sensor price than just the production-line cost; it also depends on the chip quantities sold. And full-frame DSLRs have never been huge sellers.

If APS-C sensors prove entirely satisfactory for 95% of typical photographs (counting both enthusiasts and professionals), a vicious circle results. With no mass-market camera using a full-frame sensor, volumes stay low, prices high. But with sensor prices high, it’s hard to create an appealing camera at a mass-market price point.

Furthermore, let’s consider the few users who would actually benefit from greater sensor area.

For the few who find 12 megapixels inadequate, nudging the number up to 24 Mp is not the dramatic difference you might imagine. To double your linear resolution, you must quadruple the number of pixels. Resolution-hounds would do better to look to medium format cameras, with sensors 40 Mp and up—which conveniently, seem to be dropping towards the $10,000 mark with the release of Pentax’s 645D.

The last full-frame holdouts are those who need extreme high-ISO potential. There’s no doubt that the 12Mp full-frame Nikon D3s makes an astounding showing here, with high-ISO performance that’s a solid 2 f/stops better than any APS-C camera. This is a legitimate advantage of class-leading 24×36 mm sensors.

Yet aside from bragging rights, we have to ask how many great photographs truly require ISO 6,400 and above. ISO 1600 already lets us shoot f/1.7 at 1/30th of a second under some pretty gloomy lighting—like the illumination a computer screen casts in a darkened room.

A day may come when sensor technology has fully matured, and every possible sensitivity tweak has been found. At that point, a particular sensor generation might hang around for a decade or more. So might those long production runs permit much lower unit costs for a full-frame sensor? There will still be a full-frame surcharge simply from the greater surface area and lower chip yields.

But who knows, perhaps it’s possible? By then we may have 40 Mp, 24×36 mm chips that are our new new “medium format.”

Stout Little Fellow

As with any small-sensor compact, there’s still some image-quality compromises. The active area of the TL500′s sensor measures about 7.5 x 5.6 mm, so ISO 800 still shows obvious noise.

However this new Samsung is beginning to look like one of the better options in the “serious compact” segment. (Street prices will start out about $400, presumably to drift downwards from there—that’s higher than a Canon S90, but well below Ricoh and Leica levels.)

A TL500 feature Samsung is proudly trumpeting is the camera’s f/1.8 maximum aperture (some might even mistake the needlessly-large “F1.8″ on the front for the camera’s model number). But we do need to dial down the hype about this.

An f/1.8 aperture would have been considered boringly underwhelming on any 1980s standard lens—yes, even for zooms, if we consider Super-8 movie cameras (whose image area is comparable to today’s point & shoots). Also, f/1.8 is only one third of a stop improvement over the f/2.0 offered by Canon’s S90 or Panasonic’s LX3. This is a practically negligible difference.

Also, don’t imagine that this lens is going to offer some miraculous shallow depth of field. Small sensors mean short focal lengths, and 15.6mm is the long end of this camera’s zoom range. Wide open at f/2.4, the depth of field would equal that at f/11 on a 35mm camera (using the equivalent 72mm focal length). In other words, the TL500 might give a background that’s slightly fuzzy, but not so blurred that it disappears.

One mystery is the exact CCD chip used in the TL500. Of course, Samsung itself is a major sensor manufacturer; but the lack of 720p video and the exact 3648 x 2736 pixel dimensions look suspiciously like Sony’s ICX685CQZ.  I strongly suspect this is the same chip used by Canon in the S90 and G11; and by Ricoh in the GRD III.

As with many of us, the first time I ever laid eyes on the word Samsung, it was probably on the front of a microwave oven. So perhaps it’s unavoidable that in online photo forums you’ll sometimes hear people slam Samsung—vowing to never take them seriously as a manufacturer of cameras.

Well, it’s probably time to get over that idea. Between the TL500 and the interesting NX10, it’s clear that Samsung is learning fast.

You might be aware that Samsung is already a huge force in semiconductor manufacturing (for example flash memory), as well as in LCD displays (including one these words are being typed on). Samsung was also among the first to commercialize OLED displays, a much-praised aspect of both these recent cameras.

In fact, no lingering racial snobbery should ever cause anyone to underestimate Korean industry. Some may know that Korea’s shipbuilding industry (including Samsung Heavy Industries) builds more ships than the rest of the globe combined. Or think of Hyundai—now the world’s fastest-growing, most profitable carmaker.

Sometimes the mocking comments about Samsung remind me of a Popular Photography article from September, 1946. This gave an overview of the “Jap” (yes, I quote) camera industry after WWII—basically dismissing all its products as clunky, inferior imitations of better American and German cameras. (It did allow that one strangely-named lens, the “Nikkor,” was of decent quality.) Even if at the time, there was some small grain of truth to this smug assessment, we all know quite well what happened eventually.

Are we witnessing the arrival of Samsung (and Korea) into the top tier of the world’s photo manufacturers? Perhaps so. (And while you’re at it, keep an eye on Korean lensmaker Samyang… )

There’s certainly an enormous amount of new information there to chew over, although presented in a somewhat bewildering format.

Unfortunately one side effect of the change is that it has broken all of the links my articles included to their sensor comparisons.

Sorry about that. I may try to fix a few of them as I have time.

Has everyone forgotten that it’s more than three months away?

The pleading for 24×36 often comes from those who might have already invested heavily in earlier FA lenses—meaning, ones originally designed to cover the old 135 film format.

I understand the emotion here—it’s quite true that an APS-C sensor wastes some of the capabilities of those lenses. And let’s take it as given that Pentax’s engineers would be fully capable of producing an excellent 24×36 camera.

But does Pentax “owe” its existing customers such a model?

Remember that Pentax had a serious near-death experience over the past couple of years. And its new corporate owner, Hoya, has taken a tough line in pushing them back to profitability.

From a business point of view, it’s hard to see an argument for selling a niche, low-volume body that might reduce the number of lenses Pentax could sell.

Entry-level DSLRs are the industry’s volume, brand-awareness, and profit leaders; and APS-C sensors will always serve well in those models.

Now, functionally, the main advantage of 24×36 over APS-C is in high-ISO performance. All other things being equal, the larger sensor area potentially yields about one f/stop’s gain over APS-C.

But the current Pentax K-x already gives quite decent performance at ISO 1600. And rather than fetishize a particular sensor size, our question is really “what do I need to get the shot?”

An equally valid route is to use APS-C lenses that are one f/stop brighter. (Depth of field stays comparable, if you use the correspondingly shorter focal lengths.)

So Pentax could also expand its choices of “DA” (APS-C) lenses—offering brighter apertures—even if it meant duplicating some existing focal lengths. (People may forget that two-tier lens lineups were standard practice for most manufacturers’ film SLRs.)

It has often been noted that Pentax has several glaring holes in its DA lens lineup:

• Any lens f/2.0 or brighter (aside from the unimaginative and problematic 55/1.4 SDM)
• A “normal” lens for APS-C, meaning 28–33mm (no, the FA Limited 31mm costing $1000 does not count)
• Any prime costing less than $340

Pentax has no answer to Nikon’s hugely popular 35/1.8, which sells for $200. Pentax does not compete with Sigma’s 30/1.4, which costs $440.

Yet new lenses in this range would do much to negate the perceived “full-frame advantage.” These might be an appealing, reasonably-priced purchase for any Pentax DSLR owner. Would not this offer Pentax more sales potential than a new, costly 24x36mm body?

Quirky focal lengths are something of a Pentax tradition; so to start things out, I’ll suggest one:

How about a 29mm f/1.7?

Back-side illumination (BSI in the industry jargon) is a tricky and costly chip-fabrication technique, where after depositing all the wiring traces on a silicon wafer, the substrate is flipped over and almost entirely thinned away. This leaves the wiring on the underside of the light-sensitive photodiodes (as Sony describes here), so these unobstructed pixels will theoretically collect more light.

BSI is promoted as one of the technological breakthroughs which might help save image quality, even as manufacturers race to cram more megapixels into tiny sensor areas. In fact, the IMX050CQK actually scaled back its pixel count to 10 Mp, compared to the 12 and 14 that have been becoming increasingly common in the point & shoot market.

A Whizzy Small Sensor is Still A Small Sensor

Sony introduced the chip in its own first models in the fall of 2009, for example in the WX1. But clearly Sony found it advantageous to spread the sensor development costs over a larger production run, and apparently they’ve aggressively marketed the chip to other camera makers as well. Pretty much any 10 Mp camera sold this year advertising a backside-illuminated sensor uses it. It seems particularly popular in today’s nutty “Ultra Zoom” market segment.

So I was interested to read the review just posted by Jeff Keller of Nikon’s P100 ultrazoom camera, which uses this chip. See his conclusions here.

As reviews of these new BSI-based cameras filter out, the word seems to be that they do offer decent image quality—but hardly anything revolutionary. If their high-ISO images look smooth, it seems to be partly thanks to noise reduction processing, which can destroy detail and add unnatural, crayon-like artifacts.

The issue is simply that the IMX050CQK remains a very tiny chip: The image area used is about 6 x 4.5mm. That image circle, just 7.5mm, is about the same diameter as a pencil. Photon noise and diffraction blur using such a tiny sensor will always limit image quality.

It’s interesting that Sony must resort to some special hocus-pocus to overcome this, such as a low-light mode that quickly shoots 6 frames and then “stacks” them to smooth out noise.

In truth, it seems that the real development breakthrough for Sony’s new sensor was in readout speed. Not only is HD video a must-have bullet point for every camera today, the fast readout allows Sony to offer several gee-whiz features such as fast shooting bursts, “sweep panorama,” and the image-stacking functions I mentioned.

I had all this on my mind when looking at an interesting bit of chip-maker PR that Image Sensors World posted the other day. Aptina (who I’ve mentioned before) is a leading CMOS sensor producer—although most are sold in the exploding phone-cam market.

Their perspective is that BSI’s advantages must be weighed against its problems (especially see page 5 of this PDF).

“Crosstalk” means that the signal recorded by one pixel (which ought to correspond to just one color, e.g. red) might bleed into its neighbors (which ought to represent green and blue). So if BSI has a problem with increased crosstalk, its raw data will require some extra massaging by the camera’s processor to create the final image. Potentially, this could blur fine detail more, compared to a conventional front-side sensor.

Aptina claims that it is only when pixel sizes drop to 1.1 microns that BSI’s advantages become worthwhile. For reference, that would mean a 22-megapixel point & shoot, or an 8 MP iPhone camera.

On camera discussion boards, you’ll sometimes hear the wish expressed that back-side illumination could come to larger sensors (like APS-C), and somehow allow a great revolution in their sensitivity.

But this seems to be a vain hope. Aside from the high cost (and unproven technology) of thinning such large chip areas, there is simply much less advantage when pixel dimensions grow to a reasonable 5 or 6 microns across. The wiring traces obstruct a much smaller fraction of the light-sensitive area then.

This was noted by Sony representative Toru Katsumoto during the European NEX camera launch. (Attention Toru: No, Sony did not invent BSI, nor create the name.)

Sensor technology has indeed evolved rapidly over the past decade. But we’re reaching a point where technology tricks can no longer overcome physics. Tiny sensors have inherent limitations; ones we may choose to tolerate only when portability is the overriding design criteria.

But when it comes to image quality, larger sensors still win.

Even 20 Months after the introduction of the Panasonic G1, there is still no universally-agreed-upon term for this new class of cameras.

The defining aspects of the genre are a largish sensor size (as compared to typical compacts) plus interchangeable lenses. Yet by omitting any reflex viewfinder, and instead streaming a live digital image from the sensor, the body size can be reduced from the bulk of conventional DSLRs.

Name that Evil

Now, the most widely-known term (and the one I use) is “EVIL,”  meaning “electronic viewfinder, interchangeable lens.”

A few pedants object that the Olympus E-P1 does not have a “viewfinder” in the sense of something you hold up to your eyeball (nor do the Sony NEX models, so far). But you can slightly revise the phrase to be “electronic viewing” instead, if this bothers you.

Of course EVIL originally was just someone’s little joke—but one that caught on and went viral. (Mike Johnston at the Online Photographer puckishly coined “DEVIL” the other day, adding a D for digital.)

But most industry insiders consider EVIL undignified, if not outright repellent. So there’s been quite a lot of scrambling to come up with alternatives.

Imaging Resource (who I otherwise greatly respect) quixotically stands alone in using “SLD,” meaning “single lens direct-view.” That should probably be SLDV instead—although admittedly TLA’s (three letter acronyms) do seem to have an edge in becoming widely accepted.

Unfortunately the terminology “single lens” does not really clarify anything these days. “Multi-lensed” viewing systems (e.g. twin-lens reflexes or separate viewfinder optics) basically have disappeared from today’s marketplace. The distinction being made back in the 1950s with the term “SLR” is unnecessary today.

DP Review has tended to go with “mirrorless interchangeable lens camera,” or “interchangeable lens compact,” i.e. MILC and ILC.

Using the term “mirrorless” frankly bothers me. There’s little reason to doubt that electronic viewfinders will continue to improve; so eventually the term “mirrorless camera” may sound as silly as “horseless carriage.”  Also, including C for camera in the acronym seems like a recipe for redundancy. People will naturally be making comparisons between DSLR and MILC cameras, etc.

A problem with “ILC” is that among Panasonic’s entries, only the GF1 is especially compact. (Likewise if we consider the M in MILC to mean “mini” or “micro.”)

Now, I’m sure the camera marketeers of the world yank out their hair in fistfuls every time someone says “EVIL.” The Micro-Four Thirds consortium could have saved themselves a lot of grief by cooking up a more positive-sounding name for the category right from the get-go. (Instead, they blathered about “digital interchangeable-lens type cameras.” Riiight.)

Some have tried to push the name “system camera.”  That is both inaccurate (DSLR accessory systems are far more extensive) and completely forgettable. Meanwhile Samsung refers its NX10 as a “hybrid” camera, which is equally vague. Presumably it means hybridizing DSLRs and compact cameras. “Crossover” would have been just as unhelpful.

Realistically, EVIL is the term everyone recognizes today. The alternatives are too fragmented and unfamiliar; none have gained much ground. No one on camera discussion boards is talking about MILCs. Camera manufacturers might hate “EVIL,” but that’s not our fault.

An example from cosmology might be relevant here. The term “Big Bang” was actually coined by Fred Hoyle for a 1949 popular radio broadcast; it was a slightly mocking term, as Hoyle actually believed in a competing theory. Yet the term stuck; and eventually everyone just forgot about its original slightly negative connotations. A 1993 contest by Sky & Telescope magazine actually declared defeat in choosing a better replacement.

EVIL is accurate, catchy, and short.

I predict it’s here to stay.

Big Sensor, Small Camera

The fullest NEX-5 review available so far is at Imaging Resource (while DP Review has vented annoyance that they were given pre-production cameras, and so were forbidden to publish test images).

The clearest win for the NEX system is in body size—which (for an interchangeable-lens, APS-C camera) is quite impressive. I hope all the other EVIL brands are paying very close attention.

The tiny NEXes also make it apparent that zoom lenses become even more ridiculous in the EVIL segment: Compact primes are the obvious way to capitalize on the mirrorless size advantage.

So it’s a mystery that the sole pancake prime offered at launch is a very wide 16mm f/2.8 (24e). Sony clearly needs to fill out its lens lineup still—at least adding a fast “normal” lens (approximately 30mm for the APS-C sensor format).

But perhaps they needed more time to complete something competitive with Panasonic and Samsung’s well-regarded pancakes. Sony’s tiny ultrawide stands alone here, and will admittedly be intriguing for many shooters.

Sony could not resist stuffing in a few more superfluous megapixels: These models use a 14 Mp sensor. But the high-ISO results seem very decent. Not quite class-leading (even compared to Sony’s own 12 Mp sensors used in the Nikon D5000 or the Pentax K-x); but it’s clear that Samsung’s 14 Mp sensor from the NX10 is being left in the dust.

I can tell from Petavoxel’s site traffic that the Samsung NX-10 generated much curiosity wondering if it could accept an adapter for Leica-mount lenses, in M bayonet or 39mm thread. (The answer seems to be no, unless someone can show me otherwise.) But the story for Sony’s new “E” lens mount is much more interesting:

Shallowest Lens Mount EVAR?

Thanks to Sony for providing that handy sensor-plane mark on the NEX-3. Scaling from the published body dimensions, this is one incredibly shallow lens mount. (The throat diameter is pretty generous too, from what I can estimate). Even for the shallow lens register of Leica lenses, there’s about a centimeter of extra depth to allow a lens adapter.

So the good news is that mechanically, it would be possible to adapt just about any other lens to the E mount. What we don’t know yet is whether Sony will intentionally cripple this function, either by requiring a Sony-chipped lens to be attached or by having poor support for manual focusing.

The user interface of these new NEX cameras follows in the footsteps of the Olympus E-PL1, in being very “point & shoot” oriented. There is no tactile function wheel for manual control, a major downer for any serious user. It remains to be seen whether Sony expands the NEX lineup into “enthusiast” models with better manual controls.

Also mystifying is the lack of an optional eye-level electronic viewfinder. Sony has created a new proprietary accessory port, but currently it is only used for the (included) companion flash, an optical viewfinder for the superwide lens, and a video microphone. It seems impossible that Sony would deliberately handicap themselves by not including connectors for an EVF as well; so I guess we should assume that will come later.

I do respect Sony’s decision to keep the flash separate, given that ISO 1600 shooting seems quite acceptable with this NEX sensor. But again, it does make one wonder whether Sony has a brighter-than-f/2.0 lens option waiting in the wings somewhere.

Hiroshi Onoda made some comments to the Spanish Pentaxeros website, hinting at future “more professional” DSLRs (google translation). Then, Stephen Sanderson told the UK’s Amateur Photographer that Pentax “hadn’t ruled out” a mirrorless, EVIL model.

Both statements were vague and ambiguous enough to set off a storm of speculation on Pentaxian discussion boards. Many hearts fluttered, imagining Pentax might introduce a DSLR based on a 24 x 36mm “full frame” sensor; others wondered whether they might join Olympus and Panasonic in the Micro Four Thirds camp.

Pentax Surprises Ahead?

But there’s two key facts that must always be remembered about Pentax:

• They do not make their own sensor chips
• Last year, the company almost went under

At PMA 2010, Pentax USA president Ned Bunnell gave an interview with Imaging Resource. He said plainly that Hoya, Pentax’s new corporate owners, were demanding more focus on the bottom line, and a clearer marketing strategy.

So, Pentax simply cannot build every camera its fanbase thinks might be cool.

Both Onoda and Sanderson tried to explain Pentax needs “go down a different route,” rather than compete head-to-head with larger and stronger competitors like Canon and Nikon. (Pentax learned this the hard way back in 1980, with their pro-spec LX film SLR system.)

One key concept in marketing is to identify your brand’s particular differentiators, then emphasize those. The Pentax identity seems to be in well-engineered but “approachable” models, appealing to beginners and enthusiasts.

Ned Bunnell noted Pentax was very aware of shoppers making their first move up from point & shoots (e.g. colors help make the K-x appealing as a fun family camera). Pentax also gets better recognition among hobbyist amateurs, some of whom have a fond history with the brand’s film SLRs.

The entry-level K-x is doing well with both groups. Pentax might capitalize on that by offering a step-up DSLR model, priced more modestly than the K-7 (which is nearly $1000). And the K-7 itself is due for some enhancements: It’s an embarrassment that the entry-level K-x has a better sensor. Meanwhile, Bunnell said Pentax was listening to complaints that the K-x lacked illuminated AF points in the viewfinder, suggesting one obvious change for a future successor.

So there are a number of likely next steps Pentax might take to “complete its DSLR range” as Onoda said. It is quite clear that Pentax cannot, and will not, go after the pro-shooter markets already well-served by Canon and Nikon’s top DSLRs.

Instead, Pentax’s 645D essentially created a whole new market segment, entry-level medium format, where they have no direct competition. But even there, Pentax is moving very cautiously.

So what of an “enthusiast” full-frame DSLR? Armwaving assertions that “costs of semiconductors always fall” are easy; but there remain real reasons why a full 24×36 sensor will always carry a premium price. And there are large up-front costs in creating CMOS masks for a new chip. So who would manufacture one for Pentax?

Canon doesn’t sell its sensors to competitors. Kodak is committed to CCDs—not CMOS (note the high cost, lack of video, and comparatively poorer performance of the Kodak-chipped Leica M9). Samsung’s future NX cameras will all be APS-C; selling enough 24×36 chips outside the company to make it worthwhile seems doubtful.

That leaves Sony—known to build 24×36 chips for Nikon’s DSLRs, as well as for their own. And Pentax is already buying its K-x sensor from Sony.

Sony’s 24 Mp full-frame A850 sells for $2000. For the market segment, this counts as aggressive pricing. Even so, Sony has struggled to win market share—in part because gaps in their lens and accessory lineup make them uncompetitive with Canon & Nikon (issues that would apply to Pentax, too.)

Why would Sony ever sell 24×36 sensors to Pentax at a price letting them undercut Sony’s own DSLRs?

It’s also silly to think that APS-C sensors are really “holding back” enthusiast photographers. The sensor in the K-x demolishes the ones $4000 professional cameras used a few years ago—a time when photography still limped along somehow. As photographers get into (paid) work where 24 megapixels might conceivably matter, they’ll also need the full professional accessory lineup which Pentax cannot provide.

Now in some ways, an “EVIL” camera would be a better fit with the Pentax identity: A high-quality but approachable, small camera.

Currently, the only source for Micro Four Thirds sensor chips is Panasonic. Originally, the Olympus name brought the credibility of a traditional camera brand to µ4/3. But today, it’s not clear what Pentax joining too would offer Panasonic—besides competition. Pentax would also be starting from zero with a µ4/3 lens lineup.

If Pentax moved into EVIL, the most obvious choice would be using the same APS-C sensor as the K-x. Pentax would need to use a shorter lens register than K mount; and to have a few new compact lenses available at launch. But an adapter for their current DA lenses would buy Pentax time in establishing another lens system (although this does undercut the EVIL size advantage).

The wild card is whether Pentax’s former, and now frayed alliance with Samsung would lead them to use the NX lens mount. A rumor surfaced that Schneider-branded lenses would come soon to the NX system; so Samsung doesn’t necessarily need Pentax. (Though keep in mind that at least one of Samsung’s prior “Schneider” lenses was a straight-up relabeled Pentax.)

But I “wouldn’t rule it out.”

1.4   2   2.8   4   5.6   8   11   16   22

This sequence is admittedly peculiar-looking, and it always confuses beginners. Why do larger numbers mean smaller-diameter lens openings?

But as many of you are aware, these numbers are actually “f/ratios“—that is, they’re the ratio of a lens’s focal length to the aperture diameter. Setting a lens to f/4.0 means its aperture opening measures one-fourth of the focal length.

F/Ratio Definition

(I am ignoring a slight complication here. In camera lenses, those sizes are distorted by the refraction of light through multiple lens elements; so it’s the “effective” diameter and focal length which apply. Still, this simple ratio idea will help us understand some essential truths about lens design.)

Opening a lens up by one full f/stop doubles the light it collects. To do this, the area of the aperture opening must double.

Doubling the diameter of the aperture would actually quadruple its area. Instead, to open by one f/stop means widening the aperture by a factor of √2—which equals 1.414. So the √2 the origin of those odd-looking f/stop numbers, 1.4, 2.8, 5.6, etc.

When the aperture blades are fully retracted, it’s the clear diameter of the glass elements themselves which defines the f-ratio. Lenses are always labeled with this maximum opening, as this is what limits our ability to take photos under dim lighting:

f/1.7 Maximum Aperture—A Typical 1980s "Kit Lens"

To create a lens that is one stop “brighter” means that the area of its elements must double. This doubles the glass surface that a manufacturer must grind and polish to exacting tolerances. And the volume of glass required will roughly quadruple—thus so will its weight.

Wide-aperture lenses are challenging to design while still keeping image quality high. Actually, the designer of a fast lens might need to jam in even more additional elements—juggling their extra surfaces to help cancel lens aberrations.

So, lenses of the widest apertures are always prized, and are sometimes exotically expensive.

Creating a lens with a variable focal length—a zoom—also means extra chunks of glass are required. A wide-range zoom might require 10 or 14 lens elements; while you can create an excellent single-focal-length (“prime”) lens using only 5 to 7.

Camera makers know that bulk, weight and cost are major turn-offs for consumers. Zooms already have a problem, needing all those extra elements. So to keep things reasonable, typical “kit zooms” must compromise heavily on their maximum aperture.

The focal length/aperture ratio holds for zoom lenses too. The widest aperture of the diaphragm can’t be bigger than the glass itself; so as the zoom’s focal length increases, its wide-open f/ratio must decrease. That is, zooms are brightest at their short end; dimmest at the long one. Again, this is indicated on the lens barrel:

Apertures Dim When You Zoom

The only way to maintain a nice bright maximum f/stop while zooming in is by using oversized lens elements. You’ll see this in some f/2.8 “constant aperture” zooms. Such lenses tend to be the province of professionals or hard-core enthusiasts only, because the penalty in cost and weight is significant.

With an everyday vanilla zoom, in the middle of its range you forfeit about 2 f/stops compared to an equivalent prime.

What this means is that for available-light shooting, you might have to accept a 1/15th second shutter speed rather than 1/60th. Even on a camera with an anti-shake system, a subject in motion can appear blurred at such a slow speeds. Or alternatively, you might try cranking up the ISO setting—but at the cost of higher noise and poorer dynamic range.

An interesting aspect of optics is that a particular lens design can be scaled up or shrunk down in size, while still having all the computed ray paths remain the same. Hence you can take any design and reduce it to compact size—as long as you shrink down the sensor format by the same amount.

You can see this shrinkdown happening in today’s superzoom cameras.

Consumer-level models now battle to outdo each other in “zoomsmanship,” boasting ranges like 15x, 20x—even up to 30x. But there is no physical way to achieve those ultra-telephoto settings except by shrinking the entire system. This implies using a teensy-weensy sensor chip too. And those tiny sensors (with tiny pixels) are the source of the many ills which this blog regularly protests: Excess noise, and resolution degraded by diffraction.

A more promising direction would be to take wide-aperture primes from the film era (and many excellent designs exist), then shrink those down—presumably with a nice savings in weight and cost. For APS-C, their dimensions should reduce by 2/3rds; on Micro Four Thirds, by half. Yet those sensor sizes still permit large enough pixels for high image quality.

This brightness potential becomes obvious when you look at C-mount lenses designed for the 1″ video format (a 16mm image circle, slightly smaller than µ4/3). These are physically small lenses, and apertures of f/1.4 are routine.

One complication is that legacy lens mounts derived from film SLRs all have long-ish flangeback distances, of about 45mm, to leave room for the reflex mirror box. You can’t just shrink down a film-camera 50mm f/1.4 into a svelte new 30mm version—it would sit too close to the sensor, and get whacked by the mirror.

But with “EVIL” cameras, there are no such restrictions. So, shouldn’t we be seeing an explosion of exciting new lenses today—fast yet compact? Yet to date, only Panasonic’s 20mm f/1.7 pancake for µ4/3 has delivered on this promise. Mostly, manufacturers are still wallowing in zoom-think.

EVIL Cameras Could Still Be Brighter

If our ideal is to create cameras with great image quality, but at a convenient size to carry, it is clear that we need to reconsider the zoom.

Reviewers become giddy at small improvements in sensor quality—ones amounting to a fraction of an f/stop. Why does everyone forget that any zoom sacrifices two or more stops right off the bat?

Zooms have become ubiquitous in today’s camera market. Yes, they give us convenience in framing. But shoppers have somehow forgotten what their true compromises are.

They steal light from us.