Throw Bits At It? Not Always.

February 16, 2010

Today, some musings that are a bit more (har har) abstract.

Sometimes we become numbed with the perpetual escalation of tech specs. The mindset of the computer industry, where each new generation promises more, faster & bigger, seems to be the new normal.

And it can become a self-fulfilling prophecy. If it is technically possible to ratchet up some specs number, inevitably we’ll choose to. That’s what keeps people buying!

But there’s one nice example of an industry that dangled ever-increasing numbers in front of consumers—who then yawned and said “no thanks.”

How many DVD-Audio or “Super Audio CD” titles do you own? (Okay, I’m sure someone out there is an enthusiastic adopter. But I mean, the average person.)

SACD Player

Did You Buy SACD? Me Either.

The original standard for music CDs uses 44,100 samples per second. Each sound sample has 16 bits; meaning it encodes the full range from soft to loud with about 65,000 discrete levels.

The CD standard was adopted around 1980; so at that time there was pressure to keep the bit rate low enough so that the player electronics would not be prohibitively expensive.

No one knew how data-handling ability would explode over the following decades. When you see a speed “60x” or “133x” on a camera memory card, those are referenced against the original CD-player (or, 1x CD-ROM) bit rate.

So in the audiophile world there were grumbles from the start that the CD bit rate was insufficient.

With only 16 bits, the quietest elements of music (like note decays and room ambience) must be recorded with somewhat coarse resolution. It’s similar to how shadow areas in digital photos can look noisier than highlights. A standard that used 20 bits or so would have preserved more fine “texture” in low-amplitude sounds.

And before digitizing sound for CD, any frequency of 22,050 cycles/second or higher must get chopped off. A high-pitched rising chirp that went past that frequency would make the A/D converter miss the true peaks and troughs of the wave; it would misrepresent it as a falling note, back down in the audible spectrum.

For CD audio, 22 kHz is the “Nyquist Limit,” and you need an “antialias filter” to block higher frequencies (yes, these exact same principles apply to digital-camera sensors). But there were complaints that the antialias filters degraded sound quality.

And while most adults can’t hear steady pitches much above 16 kHz, very brief clicks at a higher theoretical frequency might contribute some “edge” to percussion and note attacks. A lot of professionals preferred to record at 48 kHz instead. (You can go higher today, though bandwidth limitations in the analog realm become significant.)

Folks who produce music may have reasons for 24-bit sampling (tracks are often put through computation-intensive effects; you don’t want rounding errors), but 20-bit delivery covers an excellent dynamic range. Even taking a 20-bit master and dithering it down to a 16-bit release version can work well.

I apologize to all of you whose eyes glazed over during those past few paragraphs. The truth is, most of us found CD sound quality perfectly adequate.

If you do the math, the bit rate used for CD sound is about 1.4 Mbit/sec (in stereo). A reasonable standard that would have handled any outstanding quality issues might have been 20 bits at 48 kHz. That works out to 1.9 Mbit/sec.

But the arrival of DVD technology offered a huge increase in disc capacity. It was a great opportunity to sell a newer, zingier, whizzier-spec music format too.

So a “format war” broke out, but based on bit rates that were sort of crazy. Stereo in Sony’s SACD standard burns up 5.6 Mbit/sec—quadrupling the CD rate. DVD-A is a “family” of standards (a standard that doesn’t standardize); but its highest supported stereo rate is 9.2 Mbits/sec!

A leading writer in the digital-audio field once told me in an email that the reasons for these bit rates had more to do with “quieting the lunatic fringe” than with any technical justification.

But the public treated these new audio formats with indifference. SACD has found a niche in classical music, but most folks are completely unaware of it. (Even though, soon enough, millions did go out to buy a new disc format: Blu-ray.)

You all know what actually happened with music: Buying it online, and being able to take it everywhere in your pocket, totally changed the game.

iPod Mini

What I, And Probably You, Really Bought

So with downloadable music, the bit rate actually plunged instead. Today we’re buying music that uses 1/5th or even 1/8th the old CD standard! (Psychoacoustically intelligent data compression makes it possible.)

So… does this have anything to do with photography?

Camera manufacturers today (even those making enthusiast models) continue to use megapixels as the spec that defines “improvement.” Every year, more bits!

This shows a depressing lack of creativity. Past the point where this offers any real value, it’s just mindlessly chasing a number.

What we need is a serious rethink—to create something so novel and desirable that any talk about pixel count become irrelevant.

My feeling is that a game-changer for digital cameras is radical improvements in low-light capability. (This parallels the opinions of that recent Gizmodo article.)

We’ve suffered through decades of terrible point & shoots—whose slow zooms and limited sensitivity demanded nuclear-blast electronic flash for every indoor shot.

Flash is blinding, conspicuous, annoying to bystanders, and quite rightly prohibited at most museums and concerts. It drags out the lag time before shots.

It’s also a form of lighting which makes people look like shit.

Consider the fraction of our days we spend indoors, often under marginal illumination. But living rooms, restaurants, etc.—isn’t this where our real lives happen? Wouldn’t it be amazing to record those moments realistically, accurately, but without blinding and ugly flash?

What if you had a camera that could shoot at ISO 1600, cleanly? What about a camera where anti-shake let you trust shooting at 1/15th sec.? Plus a lens of f/1.7 or f/1.4—scooping up four times as much light?

Then, people could take photos freaking anywhere. Without flash. The light of a single candle is enough! (That’s LV 2, if you’re wondering.)

Now, to get a lens that fast, we’d probably need to lose the zoom.

Oh noes! Cameramakers’ second most-flogged spec number is zoom range. Our 12x is better than their 10x! I can hear the howls already: “Who would buy a camera without a zoom?”

Well, how many people use cell phones as their main camera today? Those have no optical zoom.

Some used to ask, “who would pay for a compressed MP3 when you can own the real physical disc?”

People who appreciate convenience. People who want technology to fit into their real lives. Us.


19 Responses to “Throw Bits At It? Not Always.”

  1. petavoxel Says:

    This post was too long already. But as a footnote about “low light monster” cameras, I wanted to mention the persistent rumors (based on patent applications) that Nikon may introduce some miniaturized interchangeable-lens system.

    The patents include an f/1.4 lens at what would be the “normal” focal length for the format; there’s even an f/1.2 portrait lens!

    A 17mm image diagonal is still small. However if Nikon were to show restraint and keep the individual pixel sizes at 5 microns, you could probably still get an acceptable ISO 1600 out of it. But that equates to about 5.5 megapixels.

    Personally, I think they should go for it.

  2. bshor Says:

    It’s true that compressed audio formats won out. But there’s another analogy to be made, which runs counter to your point. Point-n-shoots take not-so-great shots, but people choose them for their convenience. It was the audiophiles who insisted that the sound has been compromised by compression. Aren’t they in the same position as photographers who insist P&S pictures suck if only you look at the shadow areas, blah blah blah?

    Of course, from what I’ve read, there are diminishing marginal costs to higher bit rates in MP3 compression. Eg, it’s hard to tell the source from the compressed file at 320kbps or higher in A/B tests. But for a long time, MP3/AAC files were standardized at CBR 128kbps (eg, iTunes). Only recently have the specs been bumped up.

    • petavoxel Says:

      I’m just saying that for many people results matter, rather than getting hung up on numbers.

      An industry looking to promote their products can get hung up on one particular spec, even after it becomes irrelevant to what people might actually need or appreciate.

      Pixel counts have reached a point of “sufficiency” for most people (just as AAC or MP3 sound quality has). But bad indoor photos still afflict everyone.

      I think some are actually entertained by how grainy and lo-fi their phonecam shots are. But if a reasonably-sized compact which did dramatically better became available, I think people would go for it.

  3. rj74210 Says:

    The question that arises is whether there are enough people out there who would be willing to buy a convenient, pocket camera with “only” 5.5MP on the sensor, in exchange for really clean high-ISO performance.
    I personally use a Nikon APS-C DSLR for “serious” shooting, but always have a pocket camera for everyday use. I take far more pictures with my pocket camera than with the DSLR, convenience and size of the pocket camera allowing me to have it with me at all times. I’ve been using a Casio Z750 for five years, and it continues to be my favorite in good lighting, though I also use a Fuji F200 for low-light shooting and a Panasonic ZS3 for its long zoom, although both of the latter are ultimately disappointing for the noise and detail smearing that their images contain.
    The ideal for me would not be an interchangeable-lens camera, but rather an interchangeable-sensor camera! If I could slip a 5MP sensor into the ZS3, I’d be perfectly happy ! It seems that Ricoh is looking into this kind of thinking, planning to sell interchangeable lenses that have a sensor attached … the whole assembly fitting into a camera body that itself has no sensor. Seems a bit redundant to me, but what do I know !
    If I could slip a “low-light” sensor into a camera body, or a “high-illumination extreme detail” sensor for use on a bright day, that would be a really good solution for me.
    What do you think ?

    • petavoxel Says:

      Unfortunately, I think a bare sensor is probably too fragile for it to be a user-swappable part…

      I do appreciate that Ricoh was trying to do something a little different with the GXR system; and they may yet introduce the “killer low-light module.” But at the moment I’m not sure what it offers compared to just owning a couple of different cameras.

      I think it’s a certainty that we’ll see more models introduced this year which take the “small body, larger sensor” concept in new directions. I think that’s a great development. Really, I doubt I’d be writing a blog about digital cameras otherwise.

      • alex-virt Says:

        >Unfortunately, I think a bare sensor is probably too fragile for it to be a user-swappable part…

        It could be enclosed in a cartridge that is opened inside the camera.

  4. alex-virt Says:

    There is no direct analogy between sound and picture. You cannot freeze a piece of sound, magnify it and thoroughly listen every little detail as you can do with a picture. That is why sound quality is much more subjective than picture quality.

    There is a difference between CD and HD sound, and I can hear it even with my mediocre sound system and not so young ears. The difference is subtle, not as dramatic as between audio cassette and CD, and you may not notice it at first. However, it is much more pleasing for a long listening. You can enjoy hours of HD music without fatigue, something you can’t do with CD, let alone MP3. The most interesting thing happens when after long time listening to 24/96 you suddenly switch back to 16/44. Something feels wrong, but you can’t tell what it is. The music is kind of… dead.

    • petavoxel Says:

      You might ponder the remarks by The Online Photographer‘s contributor Ctein that 80 megapixel cameras may not be enough.

      People certainly exist who perceive minute quality differences, appreciate them, and are willing to pay for them. In the case of photography, many pay the price to buy a 24 megapixel full-frame DSLR; or even a 40 megapixel medium-format one.

      But most folks reach their personal level of “sufficiency” well before that. Most people simply want sharp-enough photos, rather than a particular number. They’ve been oversold on “megapixels” as the way to assess that, when that’s clearly misleading.

      Given a chance to take good-looking photos under candlelight, I think that would offer much more genuine value to people.

      • petavoxel Says:

        Speaking of the Online Photographer, I noticed their poll today asking how many megapixels would be sufficient for someone’s “dream camera.”

        It’s interesting that even among photo enthusiasts who read websites like this, half the respondents saw no need to go above 14 Mp. That’s for a “money is no object” camera.

        If anyone feels like buying me my dream camera, a D700 and three fast primes would pretty much cover it (aside from its bulk).

      • alex-virt Says:

        >You might ponder the remarks by The Online Photographer’s contributor Ctein that 80 megapixel cameras may not be enough.

        I’ve read it. 400 MP is crazy, 80 MP (about an equivalent of 4×5″ film) is nice to have for some occasions like a very good landscape or an art reproduction, but way too much for everyday use. 6 mp under candlelight is much more practical. Hope the idea of interchangeable sensors will be realized some day.

        BTW, Kubrick shot some scenes of “Barry Lyndon” under candlelight with a specially designed F0.7 lens. They look a bit soft but very nice.

      • alex-virt Says:

        >Speaking of the Online Photographer, I noticed their poll today asking how many megapixels would be sufficient for someone’s “dream camera.”

        This poll would make more sense if they gave a choice of sensor sizes too. My “dream camera” would have square sensor (10 MP Foveon or 25 MP Bayer) and ability to select aspect ratios while the whole square is stored in RAW.

  5. Huggs Says:

    I would buy a “fixed” f1.8 lens, APS-C sensor camera. For me it’s more practical and affordable than carrying around separate lenses.

    Plus, get rid of the useless “bells and whistles” that can be easily reproduced in Photoshop. Gimmicks do not produce better pics.

    • petavoxel Says:

      And such a lens would resolve much more optical detail before hitting the limits of diffraction, compared to any lens on a 1/2.3″ sensor.

      In the event that you really needed to crop hard to isolate a small part of the frame, the image could stand up to it.

  6. Patrick Says:

    Have you ever read the Meyer-Moran test on SACD vs CD? For stereo music, even trained listeners couldn’t tell the difference on a high-quality system. Sort of like the Pogue article where people couldn’t tell prints apart based on MP. Luminous Landscape did a similar, results-based, blinded test with the Canon G11.

    • petavoxel Says:

      No need to stir up the audiophiles here—it really doesn’t matter. Whether the improvement is 1% or 0%, it eventually becomes unimportant to most people.

      The difference between on-camera flash vs. candlelight is not nearly that ambiguous 😉

  7. robert e Says:

    “Psychoacoustically intelligent data compression makes [low bit rate music] possible.”

    This provoked a lot of thought. Technologically, we’re talking about smart downsampling, and we already have that in JPEG. But it does lead me to understand your thesis as a matter of “psycho-optics”, better known as marketing. Low-light performance is a good, simple idea, but I think we have something well-proven right under our noses, which is size.

    “Bigger is better” is a tried-and-true formula, and in a sense isn’t that what petavoxel’s crusade is about? Bigger sensors, bigger pixels? “Our new ZX-line’s sensors [sic] are 33% bigger than the ones in Nikon’s top compact, letting you take photos you never dreamed possible.” How simple is that? Yes, this is a matter of marketing pure and simple. More dubious, perhaps, than something like low light performance, but that’s got nothing to do with marketing effectiveness.

    Also a matter of marketing leadership. This kind of thing has to start a the top, with the pro cameras being marketed and compared in terms of sensor [sic] size (I don’t think you can sell bigger “pixels”, we need a better word.), and the “savvy” consumer encouraged to pick up on that.

    • petavoxel Says:

      It has always interested me that JPEG image compression and MP3 music compression are both able to achieve a bit reduction of about 1:5 before artifacts become perceivable. I wonder if this says something significant about our brains.

      I was interested to read once that a substantial amount of pattern extraction happens right within the layers of our retinas; the number of neurons in the optic nerve is only a small fraction of the total count of rod and cone cells.

      There is a useful term I learned recently, “spatial frequency response.” This is very analogous to audio frequency response. Adding finer pixel pitches enables higher spatial frequencies; but it’s really at somewhat lower spatial frequencies where we perceive the detail and “snap” in an image.

      I completely agree that an easy-to-remember “bigger is better” number was why everyone fixated on megapixels to begin with. Fixating on zoom range can be equally damaging, since there is no way to miniaturize a 12x, etc., zoom without using a tiny sensor.

      I’ve thought a bit about which spec might be a more useful, less damaging one to push as the next marketing bullet point. I do think overall sensor size and the size of individual pixels are two useful “bigger is better” numbers, which at least would start driving competition in a more healthy direction.

      • robert e Says:

        Damn, that’s great stuff about spatial frequency response (didn’t we used to call it “acutance”?). I’m not sure jpeg is operating in the same psychological realm that mp3 is, but that 1:5 ratio jives neatly with cognitive psychiatry’s current wisdom that 80% of what we “see” is constructed in our mind (at least that’s my lay understanding of the current wisdom).

        But re marketable numbers, it just hit me that we desperately need a replacement for ISO ratings–and that a more relevant metric could be a breakthrough both technically and market-wise.

        The fact that the best SLR’s these days are into six-figure ISO’s is a sign. As is the fact that compacts can be gained up to ISO’s that output pointilist doodles (nothing against pointilist doodles per se).

        Let’s just admit that ISO is an arbitrary and narrow measure of film performance, and almost irrelevant as an indicator of digital sensor performance. ISO is not concerned with noise, signal-to-noise ratio, or dynamic range. But we need some metric that indicates these important aspects of sensor performance–some number to indicate SNR and/or DR at nominal operating conditions.

        Yes, DxO is trying just this. But I’ve not heard one camera salesman mention a DxOMark, or seen a camera ad bragging about it. I love what they’re doing, but what they’re doing is not translating into a shoppable number.

        We can’t give into our geek impulses here. No “dB” at so-and-so RMS, etc. No synthesis of fifty different tests. We need a nice meaningful number that tells people how clean an image they’re going to get in room lighting at hand-hold-able settings.

        Maybe something like SNR/ISO X 100, averaged over the *advertised* ISO range of the camera (heh heh), measured under standardized conditions. Call it “SNISO”, or something equally too-ugly-to-not-catch-on.

        It’s a much more meaningful number than ISO alone, or megapixels; it naturally discourages useless and deceptive inflation of either; and it rewards cameras with useful high-ISO performance (we could weight it in that direction, too). It’s also a result-oriented number, rather than some physical threshold that may someday be obsoleted by technology.

  8. […] 23, 2010 Last week’s post, comparing excessive bits in music recording with megapixel overkill in cameras, drew more comments than […]

Comments are closed.

%d bloggers like this: