Photography, Reviews

The Lumu Power iOS light meter is so accurate it is will successfully replace your Sekonic or Gossen

Europe is home to some pretty innovative developers. In Slovenia, Lumulabs is a small operation that developed the Lumu Power light meter, a Kickstarter success that started well over a year ago, successfully ending in June with the first shipment of the finished devices. The Lumu Power light meter is a hardware sensor combined with an iOS app. It has both a fast-response silicon photo diode and true colour sensor in a package the size of a big marble. The Lumu Power light meter sensor is half a sphere with a stainless steel housing and polycarbonate Lexan diffusers.

Brief intro to incident metering

An industry-standard incident meter uses a white hemispherical dome to read the light falling on a subject. A digital camera measures light reflected from a subject. While that often works out quite well, reflected light measurements may be off, because the meter can be fooled. For example, a dark subject on a bright white background (e.g. snow) confuses a camera’s built-in meter, resulting in exposure errors. When using incident metering, the lightness or darkness of your subject and its background don’t matter as the meter reads the light emanating from the sky or from an artificial light source before it gets to the subject.

Incident metering should be more accurate than reflective but it does take some getting used to. The most important thing is holding the device in the light that is falling on the subject when you take a reading. With high-contrast areas on the subject, you may have to decide which area you want to expose properly — which is a creative decision, not one that is “correct”.

If you and your subject are both in open sunlight, you can take a reading from where you’re standing, near the camera. However, if you are standing in the shade and your subject is in direct sunlight, you must take a measurement in the same light as the subject.

When you are taking an incident reading from the subject’s position, the meter should be held so that the white dome faces the camera lens, not the light source. Changing the angle of the meter gives different results, which can be used creatively. Incident metering has its limits. It can’t be used when you’re shooting backlit subjects. This includes any situation in which the light can’t fall on the dome when it is pointed at the camera.

The Lumu app

For a quick overview of the app’s interface, I created a movie on Youtube that captures the screens and briefly explains them. It’s far from polished, partly because you cannot stream from an iOS device to the ScreenFlow screen recording application on the Mac when the Lightning port is taken by another device. I ,therefore, reverted to taking screenshots on the iPad, which then randomly froze the app. In short, it’s not a pretty sight, but it gives you an idea of how feature-complete the app is, how flexible it is to use and what it looks like.

You can find that video here:

Please bear in mind the app has been updated already once since I made this video, so while most of it is still valid, some features have changed, improved or added to.

The app is a free download in the App Store. Without the Lumu Power attached, you can use it as a reflective spot meter, but the smallest spot you can measure with it is at about an angle of 10 degrees. A Gossen Starlite offers spot metering at 1 and 5-degree angles, so I won’t discuss this free functionality any further — also as it’s outside the scope of the review.

How accurate is the Lumu Power?

The first thing I checked was the accuracy of the Lumu Power and whether the device measures up to its published specifications. These are quite impressive:

  • A measuring range of EV -4 to 20 at ISO 100 and an accuracy of +/- 0.1 EV.
  • The light receptor: a fast-response silicon photo diode, complies with CIE spectral luminous efficiency.
  • For exposure metering, the Lumu Power uses a hemispherical diffuser and has cardioid-type responses (cardioid means “heart-shaped” — the signal is picked up mostly from the front, but to a lesser extent the sides as well).
  • The colour temperature light receptor is a True Color Sensor, compliant with CIE 1931 Color standard Human eye perception.
  • Colour temperature sensitivity is limited to visible light and has a dynamic range of 1:1,000,000.
  • The accuracy of the colour temperature sensor lies within < 0.6 deltaE with a repeatability of < 0.2 deltaE.
  • To measure colour temperature and luminance, the Lumu Power uses a flat diffuser with a cosine-type response.
  • The luminance measuring lies between 0.15 and 250,000 lux, with an accuracy of +/- 3%.

I started by testing the luminance and colour capabilities of the Lumu Power.

Luminance and colour temperature

My reference equipment was an X-Rite i1Pro 2, controlled by basICColor’s Display software, which allows you to measure the luminance level and coloristic characteristics of ambient light. Given that professional photographers and printing companies depend on the i1Pro 2 for the colour accuracy of their output, it can be trusted to be very accurate. Mine has been recently calibrated and its measurement error is less than .2 deltaE. If you don’t know what that is, you can find deltaE explained here. Suffice it to say a deltaE of 2.3 is invisible to human perception (read the section on “Just Noticeable Difference”).

The Lumu Power has a deltaE of less than 1 in all cases where I measured luminance and colour. In the Lumu Power app, this means you will find no perceivable measurement errors when using the Illuminance, Colour Temperature and Chromaticity functionality.

While the colour temperature is represented in degrees Kelvin and a green or magenta shift for which you have to compensate towards the opposite, the app also supports foot-candles. You can also choose to analyse the luminance and colour temperature by having the app draw a curve of your measurements and averaging the results.

Finally, colour in photography often relates to setting up filters and the Lumu app supports two-light setups with colour differences given in standard colour correction gel nomenclature. An update has been planned to optionally show the green/magenta shift in Wratten numbers, plusgreen/minusgreen values, etc.

One thing you won’t be able to measure with the Lumu Power is the colour temperature of your speedlights. Either the app isn’t ready for it, or the colour sensor is too slow. I tried it in all sorts of combinations, with or without continuous reading and it doesn’t work.

I also tried to fool the Lumu Power, using awful fluorescent lighting in different colours — even mixing two of these together — but while my Sony A700 didn’t know what white balance to set for any of these lamps, the Lumu Power had no problems at all.

Exposure metering

Exposure metering is done in the Photo Ambient, Photo Flash, Cine/Video and Photo Spot metering categories of the app. I compared the results of Lumu Power’s measurements with those of my Sony A700’s built-in light meter as well as with those of a Gossen Spot Meter f/Spot Incident Flash & Cine Meter Starlite 2.

When metering for ambient light, the Lumu Power performed like a charm, with measurements consistently spot-on and only differing by approx. 1% in a few circumstances compared to the Starlite 2 readings. I expected low light metering to cause errors, but even in light as low as 7 lux, the Lumu Power kept its cool and performed accurately.

Dark objects on dark backgrounds measured correctly, with exposure consistently better than the camera’s built-in meter system. The same applied to dark objects on bright white backgrounds.

For ambient light exposure metering, the app offers three modes: single, multi and pinhole camera. The second one drew my attention. It allows you to take up to four measurements and calculate an average exposure, or a delta-EV — the difference between the measurements in EV. In the case of that high-contrast subject I was talking about earlier, I just took three measurements and averaged them to get a proper exposure for the whole thing — all across the stark differences of light.

Flash metering was exact too. I did not experience wrong measurements. The app shows a ratio of the ambient vs. flash light and it lets you analyse the flash light curve — the output over time of the flash as you fire it.

Cine/Video metering lets you set an angle instead of a shutter speed. It allows you to set aperture, angle or shutter speed, frame rate and ISO. There’s also EV and Filter compensation, but that’s it. There’s no RGB-parade like what you get with the Cine Meter II I reviewed earlier or a vector scope, but I don’t know of any dedicated light meters that have this either. You can, of course, always switch to the colour diffuser and read the chromaticity and colour temperature values continuously.

Real-world experiences

My iOS device to test the Lumu Power with was an iPad Air 2. An iPad is not the best device to use as a light meter. It’s unwieldy to use in the field. By contrast, it’s perfectly usable in a studio environment. Even in the field, you can make it a more efficient user experience by buying a data-transmitting Lightning extension cable.

When inserting the Lumu Power into the Lightning port, the sensor can start the Lumu app automatically. It does ask your permission to launch the app each time you insert it — unless you turn off this feature in the app’s settings.

Used with an iPhone or iPod Touch — which both are more or less the same form factor and size of a dedicated light meter — the Lumu Power is, in my opinion, a must-have accessory for any serious photographer, no matter whether you’re an amateur or a professional.

Each mode enables you to take notes of measurement results, complete with location metadata and the ability to take a shot with your iPhone, iPod or iPad of the place where you are for visible reference. Notes can then be read by opening them from the home screen. What I would love to have seen also is the ability to save and share all of the averaging data.

Lumu Power comes out very well when compared with the top models of Gossen and Sekonic. The only two things it can’t do is measure flash duration and trigger flashes via radio, like a Sekonic L-858D-U.

But the Sekonic will cost you north of $800, while a Lumu Power will be yours for $299.