Converting digital color images into Black and White ~ back to basics with filters

After covering the color to black and white conversion process in a simplified manner I think it might be helpful to take a look at the very foundation of image conversions.  That is to say, how we as humans "see" or "experience" the relative color intensity as it relates to gradations of gray.

Setup -

• Three color wheels illustrating different color principals organized into a single image
• Rawtherapee used to process images through the "Black and White" tool
• Gimp used to reorganize the converted images so the original color image is show side by side - so viewers can gauge the relative color "intensities"

Black and White Conversions -

Color image simply de-saturated -


As a first pass this isn't too bad.  If you like the results you can stop here and "call it good."

But as you can likely see, the dark blue areas, the reds, and greens in the upper-most wheel don't "feel" like the tones are correctly expressed in grays.  They are either too "light" to our eyes or too "dark."

So if the goal is to closely match how a color "feels" in gray relative to other colors then simply de-saturating an image might not "feel" right.

Conversion using luminosity values -


The subtle blue values seem to convert rather better in this technique than with simple de-saturation.  However, the greens, yellows, and reds still don't "feel" right.


Conversion using the Channel Mixer (no filter) -


To my eyes this is an improvement over the luminosity and de-saturation processes.  There is more subtle tonal variations between the colors.  Yet, the conversion still feels like it's missing a bit of "pop", which is to say it "feels" as if there might not be enough separation between the subtle shades of color as they are expressed in gray.


Conversion using a Channel Mixer Yellow Filter -


The classic approach in silver halide films was to take a panchromatic black and white emulsion and to shoot with a yellow filter.  The idea was to find a way to make an image "pop."  There are even lenses known to have properties that helped make an image behave this way.  The Takumar 50mm f/1.4 screw-mount lens comes to mind.  The lens coatings had a yellow cast.

In digital we can perhaps begin to understand why this approach usually worked for us old film photographers.  Take a look at the image above.  It's starting to "pop."  The colors converted to grays are beginning to "feel" more or less correct.  The only problem I see is that the reds don't yet "feel" right and the yellows are a little too "hot."


Conversion using a Channel Mixer Yellow-Green Filter -


Goal! Suddenly it feels as if we've found a very good solution. 

Applying a Yellow-Green filter in Rawtherapee we see the various color intensities expressed as clearly delineated grays.  The only comment I would make is that perhaps the blues and teals could be just a touch darker.  But this is easily fixed by gently darkening in these two colors prior to conversion to black and white.


Other Filters -

Early photographic emulsions captured only the blue end of the visible spectrum of light.  This is why all early photographs show skies as white or very light.  It wasn't until the advent of panchromatic emulsions that skies in photographs became what they are today.

If you want to emulate early emulsions you can start by using a blue filter.  Here is how Rawtherapee expresses this filter.

In the early part of the 20th century,  Saint Ansel made a fine image of Half Dome in Yosemite Valley.  He wrote about the making of this image and how he felt he'd hit upon a good solution to make the image "pop."  He used panchromatic film to capture the full spectrum of light and a red filter to darken the sky.

So in this spirit, sometimes a red filter is the right tool.  Here is how Rawtherapee expresses this filter.


In closing I can't stress strongly enough that people should do their own image conversion comparisons using the tools they normally use to process their images.  I have found that different tools implement black and white filters differently.

------------------
For further reading as additional background on the topic of how we "see" things in Black and White, please refer to this.

Thirty Years of Looking at Photographic Gear and Lenses ~ Summary of Findings

If you've read my scribblings and blather from the early days you know that I've been poking at this subject for over 30 years.

I tend to put a lot of, dare I say too much, energy into photography, camera equipment, and image processing.  In my defense, however, every day seems to bring some new understanding or some spark of insight.

It recently dawned on me that there might be a simple short summary of my "findings."  I'm not sure how this would help, but, I think it might be worth the effort to articulate a few things.  Years and years of cogitation boiled down to practically nothing.  Just a few sentences.

As someone famously said, make things as simple as you can, and no simpler.  This is my attempt to be simple and no simpler.  In this light (ahem), I feel confident enough to articulate three basic, foundation insights into the tools of photographic image creation.

Here they are -

1) The limiting factor for sharpness and resolution in an image is not the lens.  The limiting factor is, instead, the light sensitive material used to capture that image (film or digital sensor).   

NOTE: Most lenses, regardless of focal length, aperture, and coverage that I have owned, held, looked at and used, from small through ultra-large format, easily out-resolve whatever is capturing those photons of light.

 
2) The "interesting" thing about lenses is not how sharp they are (see #1 above), but how they transition into and out of focus.  

NOTE: This is where some of the real "magic" in a lens can be found.  It will show in the final results, whether we understand and recognize this or not.


3a) When shooting film - expose for the shadows and process for the highlights

3b) When shooting digital - expose for the highlights and process for the shadows

There it is.  My "findings" in a 30 year nutshell. 

I hope this helps.

Flickr ~ the long slow goodbye?

Something kept me awake last night.  I thought a lot about the current pace of change in the online world.
 
Earlier this month I received an email from Smugmug, the current owners of Flickr.  They were offering a discount to open a Smugmug account.  I thought that was curious since I thought I was already part of the Smugmug Group.

Then came an email from one of the Big Wigs at Smugmug.  It sounds like Flickr is failing to meet financial expectations.  As a last gasp, the new owners of Flickr are pleading with users to buy more subscriptions.

I've been a Flickr member for 15 years and am a rather heavy user.  I have over 28 thousand images posted and 13 million views.

My use pattern quickly developed in the early days.  Flickr hosted everything I wanted to share.  It has never been a backup site (a massive image cloud, if you will).  Rather, it was a site for publishing finished works and the make connections with like-minded image-making crafts/arts-people.

My blogs, Facebook, Tumblr, Instagram, and PX500 pages all pointed to or originated from my Flickr account.  I have links to Flickr sprinkled nearly everywhere I've had an online presence.

Over time, however, my online participation changed.

I closed and deleted my Facebook account after I realized Zuckerberg and his management team were more interested in selling out than they were to expressing, enabling, or extending truth telling, truth sharing of democratic impulses in the country they are incorporated in.

By extension, because Facebook owns Instagram and has an even more onerous business style, I cancelled them, too.  It wasn't just the personal information selling that concerned me, it was their loose interpretation of copyright protections.

Similarly, I deleted my Tumblr pages after Yahoo sold to Verizon and became "Oath."  Tumblr had been a free-wheeling environment where I could find just about anything image-related.  There were sometimes interesting image ideas that I could borrow from.  But with the Verizon acquisition came a Puritanical lock-down on certain image forms thus casting out any art expression management didn't like.  My image never tended toward anything controversial, but it was the principal of the matter.

At one time it seemed like PX500 was doing good things.  They had their image sharing platform and they offered guidance on creating good images as well as conducting interviews with prominent photographers.  Two things happened in similar time to cause me to close my accounts there, too.  First was the huge security breach that they failed to tell users about for quite a long time.  Second was sale of PX500 to the Chinese.  That did it.  I was done.  Out of there.

Now comes word that Flickr is not meeting financial goals.  The implication is that the platform may not be around much longer for paid subscribers.

This leads to a number of questions -

• What will become of the images I've posted to Flickr?
• Will the site simply go "dark?"
• Will Smugmug provide a migration path to their other platform?
• What do I do about all my Flickr links that are embedded in things I've written
• How do I manage links and information on sites I have contributed to but have no control over where my contact information is published?

Most of this I can manage, I hope.  Yet this doesn't really address my concern for where to share my images.

Should I move to Smugmug and bet against history that they continue to live?  Or is there another platform I haven't already considered?  Or should I just pull the plug and "go dark?"

Happy Holidays!  It's been an interesting year.

UPDATE: The CEO updated his comments. 

UPDATE2: The CEO added yet another comment.

Out of Focus Rendition ~ a summary of thoughts and findings

What a year it's been.  The end result being that I have moved decisively to considering more than just lens resolution when thinking about commercially available optics for photography.

With the help of some comments made over on DPReview (which, unfortunately, I can no longer find) and Nikon's "Thousand and One Nights" series I have had the opportunity to explore something that truly distinguishes one lens from another.

I continue to poke at the topic in spite of the mobile phone having become the primary image maker for the vast majority of people.  Thus far I feel that phone camera images have a "synthetic" feel to them.  They still feel much like the early point and shoot camera images did.  They tend to have a "water-color-y" look and feel, but this is quickly changing.

I don't have the budget to rush out and buy the latest gadget, which encourages me to continue to explore interesting properties of "old school", "traditional" photographic equipment and, of course, lenses.  For the moment I am sticking with stand alone cameras with APS-C sized sensors (Sony NEX and Axxx) and old manual focus lenses (Nikon, Takumar) and, much of the time, focal reducers (Lens Turbo II). 

This year I learned the most interesting thing about commercially designed and manufactured lenses is not how sharp they are.  Sharpness in the purest sense of the word is rather trivial to achieve.  Lenses more than one hundred years old are quite often sharp.

Other optical properties are at last as important as resolution.  Field flatness, chromatic aberration control, and distortions are lens faults commonly found, even in modern optics.  While processing can do nothing to correct field flatness issues, distortions and chromatic aberration controls are usually just a short step away from being corrected in software (either in-camera, or during processing on a computer).

As for field flatness, I recently learned a lot more then I ever knew.  Fixed focal length lenses are often touted as having flatter fields than zoom lenses.  I have found it depends on the lens and distance to the subject, but to be, in the broad sense, true.  I have several wide angle fixed focal length lenses that suffer from as much field curvature as a few zoom lenses I have.

Conversely, zoom lenses are frequently criticized for being "soft" in the corners.  But this depends on where you focus the lens.  Most of us focus a lens near the center of the field.  As an experiment try focusing a zoom lens in the corner and you will likely find that it actually is quite sharp there.  The effect is a clear demonstration of field curvature and not of a zoom lens being "soft".

In all my years of looking at this only one family of lenses ever tested soft when I expected it to be sharp.  I have owned far too many of these thinking I'd somehow picked up "bad" copies.  The lens is the famous but utterly uselessly soft down to f/8 Zeiss Tessar 50mm f/3.5 or f/2.8.  I paid 7USD for the last copy I had and now know what I know and, well, I will never buy another, no matter how cheap.

Tessars don't need to be soft, and in fact most aren't. I had a 200mm Nikkor-M f/9 large format lens that was razor sharp from wide open.  Similarly, every single Kodak Commercial Ektar f/6.3 (all tessar formula lenses) I ever owned and used was incredibly sharp, again, from wide open.  So the Zeiss 50mm Tessar problem wasn't with the optical configuration.  It was something else.

Related to famous marque identification is the phenomenon where someone "in the know" claims some lens or other to be an un-discovered gem.  It's crazy watching how the market responds like sheep in a herd.

Nearly anything labeled Zeiss or Leica appear to "hold their value" on the open market.  It doesn't seem to matter if the lens is actually "good" or not.  No matter.  The brand etched onto the lens barrel seems one factor that drives pricing.

Then there are lenses which have special "optical effects."  Petzval lenses went from relatively cheap and unknown to highly valued and expensive when someone talked up the "swirly" out of focus rendition.  I watched as an American whipped out a rather impressive stack of Euros at a French photo swap to pay a German for a small collection of brass-era lenses, some of which were Petzvals.

Similarly, the Helios 40 85mm f/1.5 "Sonnar" design Russian lens has the power to drive crazy prices.  True Petzval lenses are rare and might be worth what one pays, but the Helios?  Really?  There are a ton of them out on the market today but no one is asking the less than 100USD they might have gone for before they were "discovered."

Then there is the "soap bubble bokeh" craze that, even now, appears to be raging out of control.  Certain lenses are so expensive that it takes my breath away.

The herd is "all in," as they say.  And for what?  A three element lens that was deliberately designed to give over-corrected spherical aberrations behind the point of focus so that one or two aperture clicks down from wide open the thing would be acceptably sharp?  Yikes!  If what you really want is "soap bubble bokeh" I can point you to three lenses that might set you back all of 25USD that still do that trick.  But find one of those "special" lenses and you might pay a bunch of money for it.

No.  I try to avoid herd-thinking and wallet-denting pricing for an imaging fad.  None of those things are all that interesting to me.  Besides, my fixed income life has put a halt to chasing highly touted optical "pixies."

However, the thing I find most interesting about lenses is how they transition from in-focus to out-of-focus.  That is where the optical "magic" lay.  That is where lenses can distinguish themselves, one from another.  And like field flatness, there are no "corrections" for this in software.  The effect is inherent in the lens, whatever it is.

Optical formula has little or no effect on out of focus rendition.  This includes Plasmat, Planar, Xenar, Xenotar, Tessar, Ektar, Artar, Sonnar, Gauss Wide Field and many many other basic, and now classic optical formulas.  Lens element layout is not a real, nor very useful predictor of out of focus rendition.

I smile when I read marketing literature claim such and such a lens is "classic Sonnar," for example, and such and such lens will give the "classic Sonnar" out of focus rendition.  Such claims are wildly misleading.

Returning to my large format lens example, the Nikon Nikkor-M 200mm f/9 tessar formula lens was not only sharp from wide open, but the out of focus rendition was creamy beautiful.  Not so on either account the poor old Zeiss Tessar 50mm previously mentioned, which is famously "soap bubble bokeh" over-corrected for spherical aberrations behind the point of focus.

How spherical aberrations are treated in the out of focus areas of an image are determined by the lens designer and the calculations they make, in the lens curvatures they specify, and not by the number of elements nor the configuration they are arranged in.



Coming back to the practical, experiential world, to my eyes, I find I prefer -

• Neutrally corrected spherical aberration lenses for most subjects including street scenes and transportation (cars, trains, motorcycles)
• Under-corrected spherical aberration lenses for portraiture, still-life, and gardens (trees, shrubs)

My current favorite neutral spherical aberration corrected lenses include -

• Nikon Micro-Nikkor 55mm f/2.8 and f/3.5 
• Three Nikon zoom lenses
• Nikon 75-150mm f/3.5 E-series 
• Nikkor 80-200mm f/4.5N Ai 
• Nikkor 100-300mm f/5.6 AiS

My current favorite under-corrected for spherical aberration lenses behind the point of focus include -

• Nikon Nikkor 50mm f/1.8 AiS
• Nikkor-P 105mm f/2.5
• Nikkor 85mm f/1.8 H, HC, or K

If I had unlimited resources and access to other lenses from other manufacturers it would be interesting to see how they compare to my current collection of Nikkors.  Alas, such things must be left to others interested in the topic or for me in another lifetime.

What to pack and other First World problems...

... so... we're packing our things in preparation for leaving Paris for the winter.  I never thought I'd see the day when I, too, could be considered a Snow Bird by flying (as it were) south for the winter.  But there's the truth of things.

A couple friends asked which cameras I would be taking.  My answer was this -

I ran a test of my lenses to see how much I "gained" in terms of IQ by going with the old Nikkors. 

Alas, It's really difficult to tell much difference. I know where to look so I know the old lenses out-perform the new stuff. But it's not really significant.

Particularly this beautiful 50mm f/1.8 SEL (APS-C only) Sony that I have. It turns out, she's a 'beaut. 

Just one 20mm Nikkor UD f/3.5 out-weighs an A5000 + two AF plastic lenses. Weight. 

And I'm getting old. So... the AF lenses are what I'm taking. The Nikkors I'll continue to use, as you say, around town and close to home.


I wrote too soon and promised things that will never ever come to pass.

A little thought came when I looked more closely at the Out Of Focus Rendition of the Sigma 30mm f/2.8 EX DN E I was considering taking.  Frankly, the OOFR is scary bad when compared against a Micro-Nikkor 55mm f/2.8 Ai.  Looking at some of my recent work with the lens convinces me of the Micro-Nikkor's brilliance in separating a subject beautifully from the background.

At which point I felt the Sony A5000 with the Sony 16mm f/2.8 SEL, the Sony A6000 with the gorgeous Sony 50mm f/1.8 SEL OSS, and the Sony NEX-7 with a Lens Turbo II focal reducer and the Micro-Nikkor 55mm would do the trick.

Slippery Slope Alert: 
Since the Lens Turbo II is now part of the "kit", why not, I asked myself, carry not only the NEX-7 with the aforementioned Micro-Nikkor and for just a bit more weight carry the lovely Nikkor-O 35mm f/2 that I recently picked up for a song along with it?  That way I could carry one workhorse camera along with a slightly wide and slightly long lens.  Maybe I could leave the Sony A6000 at home?

Er.  Right.

The NEX-7 is new to me and I'm not sure I'd put full faith and trust in it.  Particularly since we'll be gone for three months.  Anything could happen, right?  Ah, heck!  The A6000 is small and light, so just put a body cap on it and take is as a backup camera.  Just in case, and all that.

You see where this is headed, right?  Yes.  I was sure you would.  But.  It gets worse.

Quickly reviewing the list of places we want to visit I see several art communities, small villages, a couple car and motorcycle events, and, of course, Monaco.  Indeed, the NEX-7 could do well in these situations.  The Micro-Nikkor could practically live on the camera.  Or maybe the Nikkor-O could do most of the imaging duties.  No worries.  I'd have both with me.  All is good... um... hold on...

We want to visit Menton during the Orange Festival (which is at the very same time as the Carnival in Nice).  Part of the Menton partying includes an orchid show.  Hmmm... I've been wanting to shoot some flowers using... oh gawds... here we go... some more...

Now I'm at the point of taking the A5000 with the ultra-wide angle pancake lens.  It's super light and will be nice when I photograph cars from above.  I'll have the NEX-7 with the focal reducer and the two lenses, 35mm and 55mm.  The A6000 will be sitting there as a backup, so why not mount up a wonderful Nikkor-P pre-Ai 105mm f/2.5 and take the little Nikon 12mm extension tube, too?

There.   Everything is covered.  All imaginable situations (based on prior history of being on the Riviera during the winter) have been considered.  Alas, there are a few more days before we leave and maybe a new thought will come to mind?  I sure hope not.

That we all should have such problems, right?

Tones and Tints ~ Creating Look-Up Tables (LUTs) for use in RawTherapee

Background ~

About a year ago I started to have trouble with a hard-drive that was failing on my super-fast laptop.  My father's HP (different model than mine) did the same thing a year earlier.  Maybe HP laptops overheat the hard-drive?  Or perhaps they had a long run of "bad" drives?

Fed up and disgusted with how Microsoft Windows rots from the inside out and how HP buries it's hard-drives deep into their laptops, I jettisoned the whole thing and returned to my roots.  One of the things I did in my work life was bring Linux to software developer's attention.  In this case it was easy for me to grab a slow, under-powered, and unused Dell laptop and load it with Linux and have it perform decently.

I have used the Gimp for over fourteen years and am used to Open Source Software.  The biggest restrictions for me with the Gimp is it's inability to open RAW files and to work in anything but at an 8-bit color depth.

The Gimp offers at least two useful approaches to change a base image's colors, "Sample Colorize" and a wide range of options implemented in G'Mic (Color Presets, Color Grading, Film Emulation). 

Around the time my Win10 HP box was causing trouble I downloaded RawTherapee.  It opens a vast variety of RAW formats and provides a default color depth of 16bits.  While at first it seemed difficult to understand and use, I quickly caught on.  Naturally one of the first goals I was to emulate film tones in the RawTherappe 16bit space.

Fortunately G'Mic has a set of emulations for RawTherapee.  I installed them, but instantly wished for more.  There are some really interesting LUTs in G'Mic Color Presets on the Gimp and over the years I have created some nice color samples from beautifully tinted black and white photographs.  These are what I wanted over in RawTherapee.


LUT creation and transfer ~ 

Using the (privacy) Force I did a bit of research and quickly understood how I could transfer some of the Color Presets and film emulations I like as well as my black and white tints from the Gimp into RawTherapee.   

I used the following materials -

• Linux Mint OS (any Linux distribution should do - and something similar should be achievable with Windoze and Apple)
• A copy of the base 12 or 16 bit HaldCLUT image to work from (the LUT base)
• In the G'Mic RawTherapee film emulation package I have I found a 12 bit version of the base LUT in .../HaldCLUT/ Hald_CLUT_Identity_12.tif
• There is a 16bit LUT in png format found here under Pat David's first comment
• the Gimp (in my case v2.10)
• G'Mic installed (the latest version)
• black and white Sample Colorized step wedges
• RawTherapee
• G'Mic film emulations (used for its directory structure if you don't want to create your own - besides the G'Mic film emulations are a pretty nice place to start)

To transfer LUTs or Film Emulations from the Gimp -

• Open an unaltered base LUT in the Gimp
• Open G'Mic -> Color -> Color Presets
• Apply the desired Color Preset to the open LUT image
• NOTE: I like to keep the default layer active and have G'Mic write the changed color pattern as a new (inactive) layer.  That way I don't have to destroy and re-open the base image.  
•  Keeping G'Mic open, move to the Gimp after the new LUT layer is available and save the Color Preset modified layer named as something meaningful to you and in .png format (this is essential)
• Repeat these last two steps for as many Color Presets as you would like to transfer into RawTherapee

To create black and white tint LUTs in the Gimp -

• Open an unaltered base LUT in the Gimp
• Open Color -> Map -> Sample Colorize
• Sample Colorize the open LUT image using a BW tint step wedge
• NOTE: I like to make a copy of the base as a new layer and Sample Colorize the new layer.  That way I don't have to destroy and re-open the base image.  
• Save the Sample Colorized LUT layer named as something meaningful to you and in .png format (this is essential)
• Repeat these last two steps for as many Color Presets as you would like to transfer into RawTherapee

Making these updated LUTs available in RawTherapee -

• Verify the location of the LUT directory you have pointed RawTherapee to
• If you're not sure where the film emulations are found, open RawTherapee -> open Settings and note the LUT directory location
• Using a terminal or, better, Folder view, Change Directory to the LUT directory and note there are (at least) two sub-directories inside the HaldCLUT directory
• Color
• Black and White
• Inside each of these directories are further sub-sub-directories.   At this point you can choose an existing sub-sub-directory, or, as in my case, create new sub-sub-directories using a meaningful naming convention (such as GimpLUTs or anything that differentiates your new LUTs from the existing collection)
• Copy your newly created .png LUTs from their current locations into the new RawTherapee directory structure locations.
• Restart RawTherapee if it's still for some reason open
• Go to Film Emulations and verify your new LUTs are where you expected them to be
• Open an image file and apply your new LUTs to verify they look like they did in the Gimp

If you are fluent in Linux system and application management you will quickly recognize there are several ways of achieving similar results.  What I've tried to provide here is a recipe based on an existing directory structure, that being provided by G'Mic.  Of course you can create your own directory structure and forego the use of G'Mic altogether.  Just point RawTherapee at your top directory and the system should be able to understand your intent.

A little something called Super-resolution [part deux]

After looking at handheld super-resolution image making a friend suggested that image stitching, rather than image stacking, could lead to useful improvements in image file size.  So, naturally, I had to take a closer look.

Before we look at the results, let's compare the two approaches.

Image Stacking for Super-Resolution -

• Shooting handheld
• Fire-off a half dozen images of the scene using a high-speed multiple exposure mode
• In processing -
• Align the images
• Stack the images as layers in PS or Gimp
• 2X bi-cubic 600DPI (minimum) up-rez every image
• Set the opacity of each layer
• Flatten the image
• Apply an Unsharp Mask of 2 pixels

Pros -

• Image noise is reduced significantly
• Light/Dark transition zones are smooth and "creamy"
• 2x up-rez gives a somewhat useful, if not exactly brilliant, increase in viewable detail
• If all else fails, at least there is an image to begin with, up-rez'd or not.  

Cons -

• The flip-side of the third entry under "Pros", even limited to 2x up-rez, the increase in viewable detail isn't exactly brilliant
• Limited to photographing static subjects
• Limited to 2x up-rez (or thereabouts)
• How do Sony, Pentax, and Olympus make their camera sensor "wiggle" up-rez function work so well?  If you know the answer, I would be happy to learn a bit more about how they achieve such beautiful results in-camera.  In principal the two approaches (IBIS sensor motion and handheld image stacking) should be fairly equivalent, but clearly they're not.  I'd like to see if there might be a way to refine the handheld multiple exposure approach.

Image Stitching for Super-Resolution -

• Set exposure to something at accurately expresses highlight and shadow detail - use this combination of aperture, shutterspeed, and ISO for taking all the "section"
• Shoot small-ish "sections" of the scene where images sequentially overlap eachother, making sure you've covered the entire scene (see: Breznier Method)
• In processing -
• Load the images into an image stitching software (Photoshop, Hugin, etc.)
• Allow the software to generate an output file
• Crop, finish up in something like Photoshop, the Gimp, RawTherapee, or DarkTable

Pros -

• Enormous sensor-level resolution images can be created (example - the original in this case is 20,000 pixels in the long dimension)
• Depending on the software, various projection methods give optional final image "looks" - ie: you can go so far as to emulate fisheye distortions if you wanted to (example - not fisheye, but look at the ceiling "pull")

Cons - 

• Images need to be planned with final images visible only after processing - which means it really helps to "pre-visualize" a scene
• Depending on the software and accuracy of shooting image "sections" there may be distortions (example of failing to accurately rotate the lens around the nodal point)
• Limited to static subjects
• Slow setup time - suggest manual exposure metering to help the "sections" stitch correctly and to keep the overall final scene exposure even and correct

 

Comparison of Resultant Images -

[If you click on the image it'll take you to the Flickr hosting site. Once there, look at the file at full resolution. In many cases the differences between lenses is small and likely can't be seen until you take a squint at the comparison at 100 percent.]



The obvious conclusion is this - even though the stitched image is 1400 pixels shorter in the long dimension than the stacked up-rez'd image, the stitched image clearly resolves small details better than the stacked image.



My friend is, of course, correct.  Check out the section titled "4-Way Focusing Rail..."  The image stitch approach can be very nice indeed, but only if you plan ahead.