# Photo advice and a plea for tolerance

So I followed a Daring Fireball link to this post, where “marketing guy” Ryan Cash says the new white iPhone 4 is thicker than the black one.

He may well be right, but the four photos in his post certainly don’t show it. The first one is taken from the wrong angle, as is the second. The third one, where a black and a white phone are stacked, couldn’t possibly be used to determine a small difference in thickness. And the fourth one is focused on the background instead of the phones. (I’m not going to steal or hotlink his photos; go ahead and follow the link to see them.)

A brief perusal of my Flickr feed will tell you I’m no artist when it comes to photography, but one thing I can do—because my job requires it—is take pictures that clearly document the size, shape, and condition of equipment. So here’s a tip for Mr. Cash if he really want to show a thickness difference.

Wait. Before I do that, let me digress and say that the best way to prove a difference in thickness is to measure the goddamn thicknesses. But perhaps this is too technical a solution for a “marketing guy.”

Another digression: Strictly speaking, showing a difference between the thickness of one white iPhone and one black iPhone doesn’t show that white iPhones as a class are thicker than black iPhones. Look up the word tolerance (there, I did it for you).

Neither of these digressions are meant to imply that Cash is wrong. He’s seen the phones and held them in his hands, which puts him a step ahead of me. But there’s a difference between saying something and proving it. Which is probably also lost on a “marketing guy.”

Anyway, here’s how to take the picture. Set the phones side-by-side on a clean flat surface (as he’s already done). Lay a straightedge across the thicker one so it extends out over the thinner one. Make sure the background is of a contrasting color or brightness, and photograph the gap between the straightedge and the thinner phone. Part of the thicker phone should be in the frame, so viewers can see how the straightedge is being held.

The advantage of this technique is that you don’t have to align the camera perfectly with the surfaces of the phones. The gap is visible over a wide range of viewing angles. Here, for example, is a photo of my phone and my son’s—both black iPhones bought back in June.

Well, whaddaya know! A black iPhone is thicker than a black iPhone!

Which one is thicker? Take your pick. My son’s phone varies from 0.370″ to 0.372″ around the edges, and mine varies from 0.369″ to 0.373″, so I could’ve made either one look thicker in the photo. As it happens, I forgot to take note of which was left and which was right.

Update 4/29/11
Forgot to mention how I got the thicknesses quoted above. I used an old-school set of Mitutoyo dial calipers that I’ve had for 15-20 years. A micrometer may have been a better tool because they allow you to reach in and get the thickness at an interior spot, but I didn’t have one handy when I wrote the post. The calipers were fine for getting edge thicknesses.

Which gets back to the tolerance thing. Not only is there going to be variation from one phone to another, there’ll be variation from place to place on the same phone. If someone does go out a measure a few white iPhones and finds them to be several (not just a few) thousandths of an inch thicker than black iPhones1, I’ll be convinced there’s a difference.

I won’t think the difference matters, but I’ll be satisfied that it exists.

Update 4/29/11
I’m very nearly satisfied by this post at TiPb (also linked to at Daring Fireball).

• Actual measurements? Check.
• A difference of several thousandths of a inch? Check, but just barely. The quoted 0.2mm difference is between 0.007″ and 0.008″.
• A photo that shows a real difference? Check. It would be nice if the photo were in focus, but it shows the difference pretty clearly. Bonus points for including photos of you measuring the phones with a set of digital calipers.
• A comparison with a new black iPhone. No check here. The black iPhone measured by TiPb was bought in July. As I said in the footnote, the proper comparison is between current black and white iPhones.

An odd thing that’s sort of disguised by TiPb’s decision to measure in metric and mine to measure in US customary units: the thickness values for the black TiPb phone are distinctly smaller than the measurements I got. Theirs range from 0.365″ to 0.370″. Mine, over two phones, range from 0.369″ to 0.373″. Perhaps this is a difference in instrument calibration, perhaps a difference in measurement technique. Or maybe the iPhone-to-iPhone difference is bigger than we expect.

In any event, TiPb’s post is far more convincing than Cash’s. The last bit of info will come if someone with a newer black iPhone takes measurements.

Apple, of course, has the actual thickness tolerance figures, but it’s unlikely they’ll ever see the light of day.

Update 5/2/11
Consumer Reports now weighs in on this extremely critical issue and says the black and white iPhones are the same thickness and have a photo that backs it up. I’d feel more confident in the CR measurements if the photo that accompanies the assertion didn’t show

1. exactly the same thickness for both phones, which is something of a coincidence; and
2. the caliper blades running across the phones’ home buttons—not the best place to take a measurement.

Also, the CR item says that others were claiming a 2 mm thickness difference, a significant misreading of the reported 0.2 mm.

So who’s right? I have no idea, but it would be nice if someone doing this sort of reporting actually knew how to how to take measurements, understood the role of manufacturing tolerances, and recognized that single instances don’t form much of a statistical basis.

Thanks to @potatowire on Twitter for pointing me to this.

1. Current black iPhones. It’s entirely possible that Apple’s tweaked the manufacturing process in the past ten months and all the iPhones being produced now are thicker than the ones from last summer.

## 6 Responses to “Photo advice and a plea for tolerance”

1. sm says:

Good call on the tolerance issue. A more precise measure, and one a marketing guy wouldn’t know about, would be a caliper. A straight edge will do in a pinch, though.

2. I decided to review the first white iPhone I could get my hands on, with an honourable mention of Dr. Drang of course:

http://www.macfilos.com/home/2011/4/29/first-review-of-the-fabulous-new-white-iphone.html

3. Pete says:

Shouldn’t you also do the photo two (or more) times, switching the position of the phones so you’re sure the gap doesn’t result from assuming your “clean, flat surface” is cleaner and flatter than it really is? (And that your straightedge is really straight, etc…)

4. Steko says:

Is it at all plausible that some of the errors are because they’re comparing a brand new white iphone 4 with a 10 month old black iphone 4?

5. Marc in Chicago says:

It’s problematic that there is a tiny bit of light shining underneath the left iPhone in the photograph accompanying this article. That means either the iPhone is curved or the straight edge is not straight.