Life is a Wild Ride

Technical Knowledge: Why Get So Wrapped Around the Axle?

If you've ever been frustrated by someone who tends to "get technical," you've probably dismissed much if not all of any knowledge that exceeds yours as unnecessary. After all, your images look fine...to you. No complaints.

Here's a hard fact of life: You only know what you know. If you were to sit down with one of your prints and a true expert started criticizing it, you'd suddenly find yourself sensitized to issues you didn't even know existed a few minutes prior. In a very short space of time your known universe will have expanded significantly, and you will have encountered the realization that your preexisting knowledge wasn't nearly as comprehensive or sufficient as you had thought.

I'm speaking from experience, and can testify that it's the satisfaction of conquering ego and formerly intimidating issues that has generated my voracious appetite for learning. I know of many people who are more advanced than I am, and their appetites for learning don't seem to have abated. I admire them for that, but I'm intimidated that it may be an eternal process.

Some may argue that you don't have to know how an engine works to drive. That is true, but argumentative and shallow. A professional driver most certainly should know how his engine works in order to maximize its performance and reliability. Arguments against knowing something tangibly and directly effectual are generally pretty lazy...and I don't feel particularly obligated to apologize for saying so.

I'd like to share with you a short description about a fundamental of digital imaging. This isn't something deep inside the science, something esoteric that you'd never be asked to engage directly or never employ directly-it's gamma. You probably know gamma as a setting for your Mac ("1.8") or your PC ("2.2"), but I'm willing to bet that you couldn't explain it.

Gamma is the rate of change in the luminosity of a pixel relative to the change in that pixel's input value. At a gamma of 1, a perfectly black pixel will be perfectly black and a fully energized pixel will be perfectly white, with all intermediate steps produced in equal increments.  This is the way that most common digital camera sensors capture images, and it is termed a "linear" format.

However, human visual perception is not linear, and digital captures must be adjusted to appear natural to us. This adjustment is nonlinear, and as such has a variable rate of change throughout the range of the continuum from black to white. Think of your Curves Dialogue Box in Photoshop. When you pull the Curve in the middle, one portion of the resulting Curve will have a steeper slope and one will have a more gradual slope than before the adjustment (it's also no longer a "linear" line, it's a "nonlinear" curve!). Have you ever considered that before? Do you know how that affects saturation? Please reread the second paragraph of this post if you're feeling haughty, because there's more...

At a gamma of 1.8, the rate of change in pixels as the input is adjusted is adjusted itself, but not as much as gamma 2.2. At gamma 1.8 pixel luminance graduations in the dark portions of the continuum have less difference than with a gamma of 1, but there is still some potential for posterization. Yet, a gamma of 1.8 is most perceptually hospitable to humans. This round goes to Apple.

At a gamma of 2.2, the rate of change is lessened so that graduations in the dark portions of the continuum have even less difference than they do at gamma 1.8. The differences are palpable. This round goes to PC.

Gamma adjustments are made during RAW conversion, during color management processes when displaying your image on your monitor/s, during print preparation...and yes, when you apply Curves (or Levels, for that matter). You use gamma adjustments of varous sorts during virtually every step of the image production process, and the effects are fundamental and profound.

How fundamental? Do you "expose to the right" to ensure that your capture has as much tonal virtue as possible? If so, you are using a simple, but very effective manipulation of gamma. Congratulations!

So, if I were to ask you to explain why the sky is blue when the color temperature outside is 3,000K, or why 18% gray is considered neutral/average, what might you say? Don't you think this stuff is intrigueing? I sure do. Forgive me.

Knowledge can effect your income (your income could be considered your "Report Card" from a commercial point of view). Aside from competing more effectively because your sales dialogue is more constructive and persuasive, knowledge can save you a great deal of money. Suppose I were to make the statement that a Canon 1DsMII provides no more resolution than a Canon 5D in an overwhelming majority of imaging applications, or that your "1440 dpi" printer can actually only print a photograph at 180 dpi, would you believe me?

If not, stay tuned...

Note: If there are any readers who can explain the math behind gamma (using it as an exponent), please do. This too is a bit distant from my focus of expertise, and I'll appreciatively defer to any real experts on the matter.

L

Blending Modes. Waring Never Had It So Good.

From Dallis Fontenot of Plano, Texas, on 4/25/07:

Hey Leo!  I would love to know how you did the blending on the Marcon image!

Dallis Fontenot
Engineered Structures, Inc.
Plano, Texas

Here's the image Dallis is referring to:

Click on the graphic to see it enlarged.

On this shoot I employed a technique called "Bracketing"-making a series of underexposed, correctly exposed, and overexposed images over time. The intent is to capture details unavialable in a single exposure. For example, an underexposed image will provide details in the lights and sky, and the overexposed image will provide details in the shadows, both of which might be severely compromised if available at all in the correctly exposed frame. Shooting over time (in this case, about half an hour, but it can range into the hours) provides a selection of lighting effects produced by everything from the sun to the architectural lighting...even my own additions to the luminance of the subject ("Painting with Light," which will need to be its own post).

The camera was tripod mounted, of course, and a Canon 24mm L TSE lens was used. Perspective correction is best accomplished on the camera unless the persepective issues are very small-that too has been a subject in PAPAChat recently but will have to wait for its own post.

The fundamental tools used to mate the various exposures are:

• Layers
• Blending Modes

I've been struggling with which topics to breach first in this series, and both Layers and Blending Modes are among my choices. Dallis has presented an opportunity to introduce both in a single application, but because of their depth, multiple posts will be necessary to explore either in usable detail.

The basic strategy is to use a couple (I've used as many as 5, but try to keep things down to 2 or 3) exposures as separate Layers (Window>Layers) in a single document ("image"). The order of stacking will depend on the Blending Modes and Layer Sets you employ, so let's set ordering aside for the moment except to note that Layers at the top of the Layer Stack are in fact...at the top.

You add an image to another image as a Layer by simply selecting all of the first image (Ctl-A), then copying it to your "Clipboard" (Ctl-C), navigating back to the destination image by clicking on it, and pasting the Clipboard contents (Ctl-V). Of course, in this type of application the images must share exactly the same pixel dimensions and should bear the same .icm/.icc profile (yet another topic for another post-Color Management). It's best to set up your layered working files as 16 bit .tif files straight out of RAW conversion.

As long as your camera was perfectly static during the shoot (requiring a good tripod) and you didn't vary your aperture much if at all (lens performance varies significantly with aperture variance), the images should "register" or align perfectly with a minimum of Seidel and Chromatic issues (don't make the assumption that you don't really need to understand Geometric Optics in order to produce better images!).

Save your image frequently. Since I work in 16 bits I save my images as either .tif or .psd, both formats that are very friendly to Layers, Alpha Channels, Layer Transparencies, 16 bits per Channel, and other characteristics that are common in advanced image preparation techniques.

Assuming the above steps were successful, we have a multilayered image in which we will be able to rearrange the order of the Layers, Mask out portions of Layers we don't want to be used by our Blending Modes, apply Adjustment Layers and Fill Layers, and even dive deep into Photoshop if we need to employ Channel Operations ("ChOps") for extreme selection, color, or contrast manipulations.

Before we explore Layers any further, please take the time to review the  basic Photoshop Help (F1) on both topics. Don't worry about Blending Modes for the moment-they're deep, and we'll not only be applying them in Layers but with Brushes!

L

Questions Here!

If you have questions, or topics relating to Photoshop or Flash that you'd like to discuss, and which are not related to any existing postings, please leave them here.

I'll try to use them as topics for a new post.

L

 

Professionalism. The Secret Ingredient.

My graduate work focused on business administration and law. Those cardinal disciplines, at least academically, involve ethics. Ethics is a "soft" topic in that it is relatively formless (compared, for example, to chemistry) and extremely contentious.  I'd really love to use the word "invertebrate" here, but I just can't put myself up to it.

Statutes, in spite of their objectivity and the ambiguities of language, are only an ethical framework because compliance is generally recognized to be a social plus-punitive actions are inconvenient and tailored to discourage recidivism, so they stink by most accounts.

Bona fide religious principles lack the intimidation of sanctioned legal enforcement, but certainly are fundamental to most ethical postures.

So, where am I going with this? Is professionalism simply a matter of subscribing to some ethics, mores...or must it include compliant behavior? Is professionalism a matter of pure function-social competencies such as dressing properly and using submissive or even obsequious language? Does professionalism require technical competencies?

In the military, professional relationships and behaviors are strictly prescribed. Sanctions may be severe and immediate (not even requiring court action, ref. Article 15 of the UCMJ). The business world has its own severe and immediate sanction-that is, the refusal of business, but there is no truly formal, comprehensive establishment of "professional" behavior.

We're left with a rather inchoate classification that can mean many different things to many different people, all of whom express approval primarily by trading dollars (yeh, I'm an unapologetic Capitalist). Lots of dollars falling into your case doesn't necessarily mean that you're "professional," but a paucity certainly can mean that you're not.

This is far too broad a topic to completely examine in a single blog posting. Please allow me to share with you some of my ideals for professionalism, then comment on what you believe my logical phallacies might be.

1. Professionalism  includes specialized competencies and knowledge, and must include the ability to effectively communicate about those abilities and knowledge. This fits perfectly with the definition of a merchant, which is a fundamental  legal philosopy dealing with the commercial provisions of goods and/or services.
2. Professionalism includes socially acceptable behavior and presentation. Not speaking perjoratively about your competition, showing respectful deference when appropriate, and contributing selflessly to a common goal are good examples of behavior that will reliably be considered "professional."
3. Professionalism includes the effective fulfillment of the needs of others. Punctuality. On-time delivery. Orderliness. These not only facilitate eveyone else's efforts, they are implied, if not express, in contractual relationships.

There is a laundry list of things that are not "professional" by most reasonable criticisms-enumerating them is pointless. It is important to realize that the elements of professionalism are not topical-they are an inherent part of a personality. A professional facade is not a professional essence at all-it is counterfeit and doomed to collapse under duress.

"Professionalism" is then a moving, morphing target but not without indentity or mass. We can probably all agree on a great majority of the elements of professionalism without even sharing a majority of common cores of experience.

My personal ethics and professionalism are driven by what I consider to be the best business manual ever written-the Biblical book of Proverbs. I look forward to the day when the Book of Proverbs, and its context, are no longer a "Secret Ingredient."

L

JPEG. The "E" Means "Experts."

One of the most contentious issues in digital asset management is file compression. The misuse of what are extraordinarily intelligent compression routines has produced a conventional wisdom that .jpg is more of an obstacle to be overcome than a powerful tool offering tremendous benefits.

First of all, JPEG stands for "Joint Photographic Experts Group." The term "Experts" is not hyperbole.

The classic .jpg compression with which most practitioners are familiar is in fact a "lossy" compression algorithm, and the derogatory effects of the compression increase as the compression is more aggressively applied. You are not obligated to use any particular compression level, and in fact, each image will react differently to various compression levels.

In .jpg compression, the raster image is first converted to a special color space ("gamut") friendly to the algorithm, and then divided into 64-pixel squares (8 pixels by 8 pixels) on a per-channel basis. If the division isn't perfect (there may be some pixels left over), the algorithm compensates. Next, a formula is written for each 8x8 block-that formula takes into account not only the native image character, but issues of human visual acuity and color management.  Then, everything is packaged up back into your specified color space, tagged, and saved. The names of various process elements are intimidating: Discreet Cosine Transform, Quantization, Chroma Subsampling, et. al.

Again, the more compression you invoke, the more derogatory the effects on the image.

Those derogatory effects are additive if you recompress an image with a different compression setting. In other words, opening an image in PS at a compression level of 4/12, then saving it again at 4/12 will not further degrade the image. Saving it at 3/12 certainly will.

At various compression levels, the 8x8 encoding blocks of .jpg become obvious, and are generally considered objectionable. .jpg corruption usually manifests as apparent waves, particularly along high contrast edges-those "waves" are referred to as "artifacts."

I'm a strong believer in hands-on experimenting to reinforce a point. Here are a couple of experiments you should run-the results will speak for themselves.

1. Open an image you consider very sharp.  Make a duplicate, and resize that duplicate down to 1/4 it's original dimension to ensure that native pixel-to-pixel contrast is maximized.
2. Duplicate your duplicate several times so that you have perhaps 6 iterations.
3. "Save As" each iteration at a different .jpg compression level, e.g. 2, 4, 6, 8, 10, and 12/12, Specify that amount in each file name or you'll get confused. I do.
   
4. Reopen all the images and zoom in (for example, Ctl-+ to 200% or 400% on each). Take note of at which compression level the 8x8 blocks appear to be obvious, then obnoxious.
5. Using your Info Palette (Window>Info, or just press F8) you will be able to locate individual pixels and check their color constituents (zoom in to 1600%, and reference which pixel you are hovering over by checking the X and Y coordinates of the pixel in your Info Palette). Compare individual pixel color constituencies from image to image, and note how much corruption seems to occur in areas of smooth tonal change (e.g. the sky) vice around sharp, high contrast edges (e.g. the center aisle in the US Senate).
6. Quantify the amount of corruption across what you might consider a statistically significant sample of pixels at compression level 10/12. If you have no interest in statistics or don't have SPSS at home (after all...who would? Ahem....), try this site.
7. Sit back in amazement as your realize that (my experience) uncorrected (.tif's from RAW) aerial images scanned from film, even in the aRGB color space, endure tonal corruption of as little as 1/3 of 1% (.003 of the total range) at a .jpg 10/12, in the most hostile areas (around high contrast edges). Particularly if you have a calibrated monitor/s, test yourself to see if you can sense the difference.
8. Resolve never to make disparaging remarks about .jpg again.

Here's another instructive experiment.

1. Open an image representative of most of your work.
2. Duplicate it 3 times.
3. Apply Gaussian Blur to Copy 1 (at Radius 1) and Copy 2 (Radius 100).
4. "Save As" each iteration at the same compression. Try 10/12, but be sure to label each file.
5. Look at the file sizes on your hard drive. Isn't that simply amazing? Consider the implications for low bandwidth raster transfer.

To be sure, classic .jpg has it's limitations: No Layers or Alpha Channels, 8-bit only, and no truly lossless compression. However, JPEG 2000 (".jp2," a different animal than .jpg) does have a lossless option (it is very processor intensive) and even allows the practitioner to specify a final file size!

Advanced Note: Although you can't save Layers in .jpg or .jp2, there's nothing preventing you from saving those Layers as separate files...but no, it doesn't work on Adjustment Layers. .jp2 does save Alpha Channels!

L

Flash CS3 Pro. The Floodgate is Open.

My Flash machine is a triple-header, with monitors 1 & 2 drive by an FX3000 (Quadro) and monitor 3 driven by a PCI GeForce FX5500.  It's been running perfectly with other Adobe products (Photoshop, Captivate, Acrobat), even when they are spanning all three monitors. Flash CS3 Pro doesn't seem to play well...moving open Panels or mousing over the menu bar is very tentative and choppy. I've tried cutting to 2 monitors and done some ankle-deep troubleshooting, but nothing seems to have any therapeutic effect. My Norton suite is now the prime suspect. Please let me know if you've suffered the same...

Flash CS3 is a substantial upgrade over 8. It comes bundled with Bridge CS3 and Device Central of course, but the new heart and soul is ActionScript 3 ("AS3").  Adobe claims that AS3 is substantially-up to 10 times-faster than legacy AS. With the new emphasis on modestly athletic portable devices, I'm praying the efficiencies bleed over into Flash Lite 3 when it surfaces.

Even if you only operate in the Flash IDE, you'll still benefit profoundly from upgrading to CS3.

There are some particularly promising aspects to AS3 for aerial & technical photographers, but there are still some gaping holes as well. For example, there has been no development of color management capabilities.  I was in a conference with Adobe a couple of weeks ago and brought the matter up...they did mention that it might be worth looking into at some point. For now, critical forensic and medical use of uncalibrated devices and hostile Delta-E metrics stays the course.

One of my sharpest criticisms of earlier Flash versions was that the Help documents, while compliant with tech writing standards, didn't serve the development of strategy. To my way of thinking, this may have been an insurmountable problem for many non-programmers since they were forced to search aggressively outside of the software package and most third party publications for meaty strategic and logistic guidance. The new Flash Help (F1) document has expansive and easily assimilated tips on coding strategy, as well as optimizing, and deploying your applications. Snaps to Adobe for their coverage of alpha transparencies, bitmap caching, minimizing redraw regions, and text anti-aliasing.

CS3  has graduated to the On2 v6 codec for Flash Video (.flv), which is accepted to provide a superior quality at equal datarates to previous codecs including MPEG-4. Unfortunately, CS3 hit the market just in time to witness the release of v7. I can't wait to see how Sorenson will respond.

AS3 does not allow code attached to Objects. I realize that there is some deep technical matter forcing Adobe's hand, but isn't that a bit like telling the homeowner he can't put his feet on the sofa? This obviates Behaviors and Components, of course...no great loss for the purists.

AS3 .swf's cannot access the Functions and Variables in an AS1 or AS2 .swf.

 

We'll start looking at Flash CS3 specifically for aerial and technical photography very soon. For now, I'd recommend that if you qualify for an upgrade that you do so (the download is only 404MB, $199). If you're new to the game, you'll have to wait for the free 30 day trial...or just go ahead and dive in-newegg.com. Whatever you do, don't be intimidated! We'll walk through it slowly and methodically.

L

 

Welcome to the Hurricane!

Idaho Airships, Inc. is a technical and aerial imaging company. Our customers and subjects include Walt Disney World, The NFL Super Bowl, the PGA, Sea World, Major League Baseball, The Olympics, Turner Broadcasting, NFL Films, ABC's Wide World of Sports, and some of the most notable Casinos, Theme Parks, Resorts and residences in the nation. Perhaps our most interesting work-night aerial video of Shamu's performance, with our blimp. Shamu might have thought it a relative-she didn't complain a bit. Idaho Airships, Inc. was a top 100 finalist as The Most Entrepreneurial Company in America, as identified by MailBoxesEtc. in 1998, our first full year in business. Idaho Airships, Inc.'s national exposure includes the December, 2004 issue of Entrepreneur Magazine.

Idaho Airships, Inc. has performed digital work ranging from the scanning of archival 4x5's of the original Lewis & Clark Expedition's Journal to restoring an image of Abraham Lincoln's inauguration. In 2006 we restored a portrait of Queen Elizabeth II for our rightly respected friend Paul Hosefros.

Idaho Airships, Inc.'s compositional predispositions are objective. Our equipment and processes are calibrated. Our products are technically correct and forensically defensible.

Few photographers engage the science underlying their craft, and most of the deep and weighty Photoshop tutorials I've seen don't directly address issues necessary for the effective preparation of technical, forensic, and aerial images. I recently used Adobe's Captivate to prepare a series of tutorials for the Professional Aerial Photographer's Association, but even with an Athlon 64 x2 running at 2.6GHz (an overclocked Toledo core with the 1MEG L2's, ~"5000+"), recording operations on files that can easily range to the dark side of 500MEG became clumsy...and the presentations far too large for FTP distribution. So, let's try it in a blog format...

The days of a static image doing the job (rising above the noise to communicate persuasively) are coming to an end. Contemporary graphics need to sing and dance...literally. Interactive animation and audio far beyond the capability of classic video are rapidly developing to be the new standard. Photoshop, Flash, and CG-they're the future, at least for the present, of visual communication. They're the disciplines we'll be exploring on this forum.

L