## Tuesday, June 5, 2012

### Measuring Your Own Pupillary Distance

Pupillary distance (PD) is the fundamental optical measurement for eyeglasses. It's the distance between the pupils of your eyes. This is important because every prescription optical lens has an optical center. I'll let Allaboutvision.com explain it:

The optical center of your eyeglass lenses is the part that gives you the truest vision, and it should be directly in front of your pupils. To determine how to place the lenses in your frames so the optical center is customized for your eyes, the eyeglasses lab needs to know the distance between your pupils, or PD.
It can be tricky to measure your own PD, somewhat akin to trying to cut your own hair. Dispensers need lots of practice to be able to measure PDs correctly, and even experienced opticians have difficulty taking their own in a mirror.
Most online optical providers present you with several ways to go about this important task. Some suggest the simplest method, which is to have your prescribing eye doctor or an optician take the measurement for you. Alternatively, they offer step-by-step explanations of how to take your own PD in the mirror, or how to have a friend take it for you.
There is only one reason you would your want your own PD measurement: your want to buy glasses online, without the expense of an optician's expertise. Ok, fine. I'm a professional optician and I have reasons to hate that people do this. But I just love opticianry. What irritates me more than the erosion of respect for and practice of professional opticianry are the endlessly reused wrong methods of obtaining an accurate PD perpetuated all over the anything-for-a-dollar online glasses websites. So I'm going to tell you why their instructions are wrong, and then I'm going to tell you simply how to do it right.

There are two systems proliferated online to obtain your own PD: manually measure yourself, or via software manipulate an uploaded photo of yourself to produce the data.

To measure yourself the common instructions are (courtesy of random online glasses site justeyewear.com)
Have ready: a straight ruler*, a pencil, and a mirror
1. Facing the mirror, place the ruler on the bridge of your nose, bringing the start of the ruler directly below the center of one eye’s pupil.
2. Looking straight into the mirror, hold the ruler (keeping it steady and parallel to the floor!) and mark the location on the ruler of the other eye’s pupil.
3. Measure the distance in millimeters between the two marks. This is your PD.
4. Repeat this process a couple of times to ensure you have an accurate measurement.
If you have a willing assistant, you can simplify the process even further by having your friend measure your PD using a ruler. (Your job will be to stand still.)
Whether measuring with someone or alone, be sure to hold the ruler steady and parallel to the floor!

 Does this inspire you with confidence or what?

You may get close to your actual PD using this method, but prescription eyewear is neither horseshoes nor hand grenades, so maybe a more precision method is called for.

The second method is to upload a photograph of yourself with some sort of size reference device in the photo. This was initially developed as the iPhone app Pupil Meter using a credit card as the size-reference device. Since most credit cards are fixed size, once you know the size of the card one can calculate the distance between the pupils in the image.

The idea is sound. It's a simple ratio equation. But the app is notoriously flawed and inaccurate. Witness it's awesome 1 star rating and the many uncomplimentary comments.

The system caught on, though, and websites developed their own more sophisticated versions of the same measuring system. I've tried a dozen and the best-designed one I found is this PD Self-Test. This uses you own current glasses as the reference device. You just measure the size of the glasses and highlight your pupils in the image using their slick interface and it does the calculation.

I tried it out. Eight times. With eight different photographs. It measured me from a 61mm to 65mm PD. So call it accurate within ±4mm. How bad can that be?

The whole point of getting a PD measurement is to have the optical center of the lens centered in front of the eye.  If it is decentered, that induces prism: "A lens with prism correction displaces the image, which is used to treat muscular imbalance or other conditions that cause errors in eye orientation." Double vision and focusing difficulties can be caused because your eyes don't work together perfectly, and prism can be prescribed by doctors to correct that and force your eyes to focus together properly. However, prism that is not prescribed and unnecessary will do the opposite: it will create double vision and focusing problems.

So unnecessary prism is bad. But nothing can be measured perfectly, even the most accurate eyewear has some margin of error. How much is too much? Well let's figure out what a pair of glasses I would order myself using the above PD Self Test measurements would be like. If my PD ended up being decentered by the test's 4mm error how much prism would I get?

The Prentice Rule is a simple optician's equation to calculate exactly that. The formula is:
$P = cf$
P is the amount of prism in diopters D
c is decentration in centimeters
f is lens power in diopters D

We are looking for the prism P. I know the decentration is 4mm, which is 0.4cm. My Rx lens power is a moderate 2.5D (-2.50, but signs don't matter for this calculation). P = 0.4*2.5 = 1.0D. To put this in perspective 0.25D is typically the smallest increment prescribed by doctors, generally because it is the smallest amount that will have noticeable impact on vision in most people. 1.0D is four times that. That is literally suffering four times the prescribed dosage. The ANSI standards for prism deviation in a spectacle lens is no more than 0.67D, or at slightly higher power ±2.5mm. The lenses made off that PD calculation at 4mm decentration and 1.0D fail both standards and are worthy only of the garbage bin.

Here's a kicker. My actual PD, measured many times by every method from a ruler held to the face to the Visioffice digital measuring system, is 69mm. The online calc didn't even come close. If I had used it's average result of 63mm, that would not possibly, but exactly cause my resulting glasses to be decentered 6mm, inducing a whopping 0.6*2.5=1.5D unnecessary prism, more than double ANSI standards. Those glasses would certainly cause double vision, strain, and focusing failure.

How can the system be that off? Well, because you have two PDs. When you focus on objects within 20 feet, your eyes converge and the distance between your pupils decreases. How far away is the cameraphone or webcam you're using to take the image for the online systems? 2 feet? 4 feet? Those systems are measuring your near PD, not the 20+ foot distance PD you want in your glasses. They don't even get the near PD right either.

There is no accuracy to online PD systems. It is improbable that glasses made from theses PDs even accidentally end up being correct.

So you've hung out through my lecture and you want to know how to do it right. It's ridiculously simple. I don't know why no online opticals recommend this. I'm sure people simply assume a slick looking digital system like the one above is more accurate than any manual system, and are inspired with confidence in their cut-rate online supplier. But manual is simply the best for doing this on your own. Here it is:
1. Wear your glasses. (Or any glasses if you don't have your own. Even just try on demo glasses at the mall if you are that bold.)
2. Have a felt-tip maker handy.
3. Focus on a single object in the far distance (anything farther than 20 feet works, but farther is better).
4. Raise the marker to your right lens and precisely put a dot on it directly over the distant object.
5. Repeat for your left eye. If done correctly, with both eyes open the two dots should overlap into a single dot over the distant object. If not repeat making the markings until they do form a single dot.
6.  Measure the distance between the two dots on your lenses with a millimeter ruler.
7. That's your distance PD.
8. If you need a near PD for reading or computer glasses, just do the same procedure but focus instead on the object you will be looking at, either reading material or computer monitor.
I've done this myself many times and always replicated the same 69mm result ±1mm. It works because the dots you are putting on the lenses mark the points through which you are actually looking, which is just exactly what PD measurement is used to determine.

I don't mind educating about this because despite the online opticals' claims, PD is not the only barrier between you and high quality eyewear. There are 30 other ways the online guys can and will ruin your glasses. Only a knowledgeable and skilled optician can make great eyewear. For excellent vision, a perfect comfortable fit, and beautiful eyewear that will last at least a year, you just gotta see an optigeek like me.

But if you want to bring me your own PD measured by my method, I'll take it. But you wouldn't mind if I just double-checked that PD myself first though, right?