Category Archives: The Arts

Bad Architectural Photoshop
Posted on March 20, 2022 by CHC   ᐃ ᐁ

This is another in an occasional series of posts about bad design, a longstanding source of disgruntlement for me.*  In past editions, I’ve covered bad software interfaces, bad signs, bad fonts and bad hurricane forecast maps.  This time, I’m going to briefly touch upon a topic of particular interest to me, since many of my posts (including my most recent one) revolve around it — I am talking about photoshop mashups and, in particular, the lighting of subjects and objects in those images.

There are countless examples of bad photoshop on the internet, most of which are easily spotted with a casual glance.  Whenever you see an image that seems “off” or unnatural or too-good-to-be-true, it means that someone at the photoshop desk has not done their job.  It is probably because they were in a rush (if I feel generous about it) or because they don’t have the training (if I feel otherwise).  The typical reason that things just don’t look right is inconsistent and unrealistic lighting, something that takes time to get right.**

I am going to pick apart just one image here.  What set me off about this particular one is who created the image — an architectural firm in upstate New York.  (I am not spelling out its name, but here you go.)  Now, you would think that architects would only hire the best graphic designers they could find, as they have their professional reputations to protect.  But you would be wrong (and not for the first time, if I may gently remind you).

Anyway, this firm “partnered” with New York State’s parks department to develop some renderings of a proposed riverside park in downtown Rochester.  Here is the drawing that caught my critical eye:

The Democrat & Chronicle caption says the rendering shows a “possible river-level view” of the falls area, but this would be “possible” only if our planet orbited three or more suns.  (That planet would instead be Triskelion; see Star Trek Episode 46.)  If I may explain…

We have one sun, and it is basically “infinitely” far away.  As a result, when the sun is out and you look around at the shadows it casts, you will find all shadows point in the same direction.  Sadly, this is clearly not the case in the architect’s rendering of this scene.

Which, if any, of those shadows are realistic?  To find out, we first figure out where the sun is supposed to be.  The answer is revealed by another amateur mistake: someone shot the background photo while facing the sun without using a lens hood, causing telltale flares, which the graphic artist could not or did not remove.  I highlighted the flares in Figure 1:

Tracing the lens flares back to their source, we find the sun is at about 35 degrees from the southeast horizon (note: the waterfall faces north) as would be the case on a mid-June day at 9 AM.  Give the artist credit — the shadow of the boy in the center of the drawing is about the right length for that time of day.

But a closer look at the running boy (Detail 2) reveals that his shadow is at the wrong angle.  Because the boy is centrally positioned in the frame, the line of his shadow should point directly toward the sun [A].  But the shadow as drawn points well to the right of the sun.

Things get more confusing when we examine the lighting and shadows on the boy himself.  The light on his face and body is not coming from the sun behind him but from some invisible source [B] above and to his right.

The rest of the drawing has similar lighting issues, which I will now enumerate (Figure 3):

[A] The river banks are sunlit, but this is not possible given the sun’s position in the sky.
[B] The seated man is illuminated by a source in back of him, not by the sun.
[C] A tree casts a shadow on the bench but there is no tree between the bench and sun.
[D] The couple walking with the small girl are illuminated not by the sun but by a source in front of them and to the right.
[E] The shadows cast by that group, by the wheelchair couple, and by the bench, point to some unseen light source above the waterfall.
[F] The couple holding the two small children are illuminated by some source on the right.
[G] The shadows cast by that group have the correct angle but are too long.

I was going to add that the woman with the bicycle at far right is not illuminated at all, but this is probably accurate given that the locale is perpetually-cloudy Rochester, New York.

But this is just a one-off, right?  Other architects certainly aren’t this casual — or are they? Here’s a example (Figure 4) from a project closer to home called, ironically, Artful Way…

[A] The people in purple clothes are translucent.  They must be ghosting us.
[B] Plus, they cast no shadows.  (Further proof of the same.)
[C] The fence casts a much stronger shadow than does the person standing behind it.
[D] This last one is more subtle and doesn’t involve lighting.  The designer presents an oblique view of the deck, together with a very close perspective; as a result, the deck floor appears to slope downward and to the right.  That’s not how people perceive the surfaces that they stand on.  I would have chosen a more straight-on, farther-away point-of-view.

So there you go.  Good thing I’m not on a planning board or nothing would ever get built.

_________________

* The online definition of “disgruntlement” is “a state of sulky dissatisfaction” which to me actually sounds a little romantic.
** Another common photoshop faux pas is “too much fine detail” which competes with “too many lights” for the Emperor Joseph II Award in Graphic Arts.
Be the next to comment | Read other posts in The Arts
The “Hang It on Two Studs” Calculator
Posted on September 16, 2021 by CHC   ᐃ ᐁ

Asked and Answered 3.4

Hello, and here we are again.  I thought I was done with this series on hanging pictures, but it seems physics never dies — it just gets more complicated.

Some commenters on my previous articles (Why Frames Tilt Forward, The “Hang It with Two Hooks” Calculator, and The Physics of Hanging Pictures) asked how they could hang items on wall studs, if the studs are off-center from the desired hanging spot.  This seemed to be a rather specialized topic, and beyond the scope of my series, so I deferred until now.  But a recent commenter rekindled my interest and finally inspired me to take a look.

Before I proceed, however, I must mention that I’m not the first to address this problem.  The number-one result (as of now) I found in searches for “hang item on off-center studs” is this article on instructables.com by an author named MolecularD.  The author describes the principles involved and offers a set of equations (minus the math) that are meant to show the reader where to place the wall hooks.  Unfortunately, some readers commented that they did not get the desired result when they followed the author’s instructions.

The solution provided in the inscrutables.com article is such a complicated equation that there is no way for me to verify it without essentially solving the problem myself.  Which is what I will do now, taking a somewhat simpler, more intuitive approach.

The four consecutive views in Figure 1 demonstrate the concept:

FIGURE 1: THE CONCEPT
Concept of Hanging a Frame on Two Studs

View (A) depicts a frame hanging on a wall, centered at our desired position (dotted line), using a wire on a single hook.  Because of the symmetry of the system, there is no tendency for the frame to rotate one way or the other.  Ignore for now the fact that the wire extends above the top of the frame.

View (B) shows the studs in the wall behind the frame (we use a stud-finder to spot them).  The two studs are different distances from the center of the frame.  We drive a nail into the center of each stud, just touching the underside of the wire.  This does not cause the frame to rotate.

In View (C), we attach a piece of wire (blue) to the original wire, from the point where the first nail touches the wire to where the second nail touches the wire, without any slack.  The load is now shared between the central hook and the nails in the studs.  But this still does not cause the frame to rotate.

In View (D), we snip away the original wire where it touched the nails, leaving our new wire in place.  The nails in the studs now assume all the load, with the higher nail bearing more than the lower.  Still the frame does not rotate, so we have found the solution.

Obviously, I don’t expect readers to repeat these steps to hang their pictures — this was just a demonstration of concept.  Instead I will offer a calculator, with instructions for taking measurements, placing the hooks and cutting the wire, to help the reader achieve the final result.

That is, if you really insist on using studs.  Personally, I think it would be easier in most cases to forget about the studs and use the Hang-It-With-Two-Hooks calculator that I presented in my earlier article.  You would fasten the hooks to the wall with toggle bolts, which can hold a significant amount of weight when paired with the appropriate hooks.  (This video shows how to install them.)  But in the end, it’s your call.

The Setup

Oh, you’re still here!  This must mean that you really, really want to use two studs to hang your item.  Okay then, onto the intricate details.  Please consult Figure 2 (below) to get a sense of the important lengths and measures:

FIGURE 2: USING TWO STUDS TO HANG A PICTURE
Diagram of Frame Hung on Two Studs

Start by measuring the height H and the weight of the item you want to hang.  Then mark the spot 0n the wall corresponding to the top-center of the item.  All other measurements will refer to this point.

Next, use your stud-finder to measure XA, the distance from top-center to the center of the closest stud, and XB, the distance from top-center to the center of the next-closest stud.

Now inspect your hanging hardware.  You want to (ideally) hide all your hardware behind the item you are hanging, which means the higher hook (A) should not show.  Therefore, you should choose a value for ZA, the distance from the top of the frame to the bottom of Hook A, that is slightly greater than the length of the hook.

While you are it, measure the length (D) of the D-rings attached to the item.  If you plan to attach the wire directly to the item, then this length is zero.

Your next measurement is WD, the distance between the D-ring attachment points.  If you have not yet attached the D-rings to your item, then mark the spots where you think they should be attached, and measure the distance between those marks.

Note that I have not asked you to specify Y, the distance from the top of the frame to the D-ring attachment point, or ZB, the distance from the top of the frame to the bottom of Hook B, or S, the length of wire to cut.  These values will be returned by the calculator.

There is one last thing you may have noticed on the diagram: to make the item hang true, you need to install a guide hook below Hook A to equalize the slack in the wire — and the forward tilt — on the left and right sides.  More on this later.

The Calculator

My geometric/algebraic solution may be found in this attachment.  Additional comments  and assumptions about the problem are provided in the appendix at the end of this post, for those who are interested.  But I’m sure most of you want to get right to the calculator, so here we go.

First enter the height and weight of the frame, then enter the position of the studs closest to the top center.  The calculator will identify which hook (left or right) is A and which is B. Now enter the remaining four values.   The default wire angle of 30 degrees should be fine in most cases.

Results update when a new value is entered and are reported to the nearest eighth-inch.  The calculator advises how to position the guide hook on the A side, and how to attach the wire to the frame.  The length of wire S includes 6 extra inches — 3 inches per side — for tying the wire ends to the D-rings.

If the calculator flags one of your entries as out-of-bounds, don’t ignore it.  The calculator will not report any results if the ratio Y/H is greater than 1/3 (that is, the attachment point is too low) and it will warn you if the estimated wire tension exceeds 25 lbs.  If the default wire angle results in a reasonable value for Y/H (1/3 to 1/5) then go with it, assuming the wire tension is not too high.

Words to the Wise

You ask, do I have to use the guide hook?  If your item weighs much of anything, then yes.  The farther that Hook B is from the center, the more the item will tilt forward at Hook A,  since there is more slack in the wire on that side.  And the more front-heavy the item, the more uneven the forward tilt will be.  The guide hook helps keep the wire close to the wall on the A side.

It may be a challenge to hang your item on three hooks.  I suggest you find a helper, if only for you to have someone to kvetch to while trying it.  (Nonetheless, watch your language.)  You might start by feeding the slack of the wire through the guide hook and onto Hook A.  Then slide the item toward Hook B and feed the wire over Hook B.

I end with my usual disclaimer.  My calculator makes it easier for a person to hang an item on two off-center studs using hooks and wire.  But whether this method is suitable in your situation is a judgment only you can make.  You assume full responsibility for your project, including selecting the appropriate hardware.  I offer this calculator as a convenience but accept no liability for damage of any kind, even if the suggestions offered here are followed exactly.  If you’re not confident how things are going to work out, you can do a mock-up in your garage before marking your walls.

With that out of the way, good luck.  I would be interested to hear about your successes, failures or problems.  As always, your suggestions and feedback are welcome.

[ A NOTE FROM THE AUTHOR ]

If you find value in this post and would like to express your appreciation,
you can buy me a coffee or martini!  Click below to add a tip to my coffee fund.

This takes you to my Buy Me A Coffee page. Thanks! – CHC

If you have found this post informative or entertaining, you might enjoy the other posts on this blog.  Check out my favorite posts and consider signing up for update notices.

Appendix: Notes on the Calculations

The result we are most interested in is:

ZB = ZA+ (XB – XA) tan θ

where θ is the wire angle, tan θ = (Y – ZA)/(WC – XA) and WC = ½ WD.

The formula for ZB assumes that Y, the D-ring attachment point, is a given.  But I don’t ask the user to specify Y directly, as this involves a judgment call.  Ideally, the ratio Y/H would be about 1/5 (the “one-fifth rule”) to minimize forward tilt of the item.  But in some cases, this might lead to too small a wire angle and create too much tension in the wire.

So, what I did in the calculator is have you specify the wire angle, with 30° as the default. The minimum entry is 20° and the maximum is the angle corresponding to Y/H = 1/3 (the “one-third rule”).  The wire angle is used to calculate Y as described in the attachment.

Read 9 comments and add yours | Read other posts in Asked & Answered, The Arts
Cartooning 101 vs The New Yorker
Posted on October 21, 2020 by CHC   ᐃ ᐁ

The New Yorker weekly magazine, long renowned for its single-panel cartoons, features a  cartoon caption contest in its online edition.  The premise sounds like fun — but this being the notoriously fussy New Yorker, the simple task of crowning a weekly contest-winner is a month-long ordeal:

Week 1:  Readers are shown an un-captioned cartoon drawn by a New Yorker contributor.  They are given seven days to submit the most outrageously funny caption known to man, or at least to themselves after a glass or two of Bordeaux.

Week 2:  The cartoon is reprinted and various submitted captions — possibly randomly selected, but preferably randomly curated — are presented.  Readers may register their opinions of these captions, each in turn, by clicking UNFUNNY, SOMEWHAT FUNNY, or FUNNY.  (Be assured, UNFUNNY is the button one’s finger will hover over.)  Exactly what happens to these votes is not entirely clear. 

Week 3:  Unlucky New Yorker staff members and interns are tasked with reviewing all the submitted captions and picking three finalists, according to some unpublished criteria.

Week 4:  The cartoon and its three finalist captions are presented to New Yorker readers so that they may vote on the winner.

Week 5:  The cartoon, the winning caption, and the runners-up — as determined by the previous week’s votes — are printed on the New Yorker‘s cartoon caption contest page.

Now, submitting a caption and then being forced to wait weeks to find out if yours is even a finalist… this is buzz-kill enough in the internet instant-gratification era.  But I haven’t even touched on one’s (slim) chances of winning the New Yorker‘s caption contest without knowing the secret sauce.  From what I can tell, a winning caption must first and foremost pay homage to the quirky concerns of upscale New York City residents.  That is the ticket that gets you in the door.  But you would be mistaken if you think your entry will score higher if it adheres to time-honored cartoon conventions — in fact, you may be ill-served by that approach.  Here are a couple of examples of what I mean.

The Spacious Apartment

The cartoon at left (click to enlarge) shows a property manager or rental agent delivering the opening lines of her sales pitch about the apartment pictured.  The lights in the unit are off and the prospective tenants have just stepped through the door.  The couple cannot see the window and its planetary view, as the door is in their way.  Rather, their concerned gazes are focused on something on the other side of the room.  What could be over there?

One would think that the winning caption should answer the question: what is it that the prospective tenants see that concerns them (and which the agent hopes to explain away)?  The scene in the window provides a clue but cannot itself be the answer, being that the couple has yet to look in that direction.  Nonetheless, here are the captions selected by the  New Yorker as the three finalists:

“It’s the closest you’ll get to Manhattan in your price range.”

“Keep in mind the neighborhood is expanding.“

“And, if you open the window, the view will take your breath away.“

Note that none of these captions adequately explains or even bothers to address the source of the couple’s concerned gazes.  Is there an alien being across the room?  Is there another floor-to-ceiling window?  Or maybe there is nothing over there at all, no walls or floors or ceilings… just empty space.  We don’t know.  None of these captions advance the story.

I clicked through at least thirty reader-submitted captions for this cartoon.  I would say that nearly a third of them incorporated some variation on “the view is out of this world.” This of course is both the most obvious and least intellectually-satisfying response, sort of the Neil Diamond of captions.  That this cliché did not make the list of finalists is of some credit to the New Yorker, but then Neil Diamond was more Brooklyn than Manhattan.

In any event, all of the finalist captions violated an important precept of Cartooning 101, which is that the drawing and the caption must work together seamlessly.  Vital elements in the drawing must be referenced in the caption, and vice versa, else there is dissonance.  One can’t simply ignore the expressions on characters’ faces.*

William Tell Revisited

Our second cartoon (click to enlarge) offers a variation on the tale of the Swiss marksman William Tell.  You may recall that Tell was forced by a local magistrate to shoot an apple off the head of his son, so that their lives may be spared.  Here, however, a cherry has been placed atop the boy’s head rather than the legendary apple.  In an unusual move for the New Yorker, the cherry has been colored red to draw the viewer’s attention.  Tell looks worried; his son wears a narrow-eyed scowl.  The background elements are unimportant.

So what is the boy saying to his worried father?  Here are the New Yorker‘s three finalists:

“I’m just saying, after this haircut, it’s difficult to trust you.“

“I now see why you only get me every other weekend.”

“You know, pumpkins are in season too.”

Let’s consider these in turn.  In the first caption (the winner), the cherry is totally ignored; the caption would have the same relevance if the target were an apple.  So let’s just forget about the target and make fun of the boy’s hair shall we?  As the caption is only marginally funny in any case, I give it a D-plus.

The second caption also fails to acknowledge the cherry.  Yes, being the son of William Tell may have its occupational hazards, but this caption would have worked just as well with the original apple.  If I were to channel my inner New Yorker — and I had just finished a good martini at Rockefeller Center — I might grant this one a middling C.

The third caption at least makes reference to the size of the target, so partial credit for that.  However, the boy’s remark implies that it is his father who has chosen the target, which contradicts the legend.  And if I may nitpick, there were no pumpkins in Europe when the William Tell narrative arose.  Pumpkins — native to America — became known to Europeans only after a 1584 expedition by Jacques Cartier, over a hundred years later.  Furthermore, pumpkins and cherries are not in season at the same time, not even in modern-day Switzerland.  So the whole premise is unsupported by both fact and legend.  This means I will have to give the third caption a C-minus for sloppy research as well as dearth of funniness.

At this point, you are probably anticipating that I am going to offer my own killer captions for these cartoons.  No, not today.  As they say, those who can, do; those who can’t, teach.  Or they run the caption contest at the New Yorker.

_________________

* Don’t get me started about all the contest entries in which the caption is delivered by a character whose mouth is closed.  However, that’s not the New Yorker‘s fault — perhaps blame Bordeaux.
Read 5 comments and add yours | Read other posts in Humour, The Arts
Content © Craig H Collins 2026  ᐃ ᐁ