Review: Solmeta N3 Geotagger
Recent Nikon DSLRs has built-in support for automatically attaching positioning metadata to an image if fitted with on-camera GPS receiver. This review is for the Solmeta N3 Geotagger, which were first shown at CES 2012 in January.
The first GPS receiver designed to communicate with a Nikon DSLR was Nikon's own GP-1. Nikon hasn't updated the design since it is appeared in 2008, and it is getting long in the tooth, needing at least 3-4 minutes to obtain an initial position under good conditions.
Solmeta already has a line of geotaggers for Nikon, and the N3 model, which is the subject of this review, is a second generation device. It is supposed to obtain a faster lock on satellites, and to use less power, than first generation devices such as the Nikon GP-1. Unlike Nikon's unit, it also has a three axis compass which provide heading data, and it will remember last outdoor position when you move indoors (or otherwise becomes out of range from GPS satellites). Unlike the other Solmeta models, the N3 does not contain a battery. Instead, it is powered by the camera's battery. I think this is an improvement. I prefer bringing one or two spare Nikon batteries along in the camera bag, rather than having to charge the non-replaceable Solmeta battery.
I ordered my device from Solmeta's Hong Kong website on July 13, 2012, and received my unit six days later, on July 19. The photo below shows the content of the box.
Because the unit is connected to the camera through its accessory unit, a corded remote release is included. It plugs into the side of the N3 with a 2.5 mm stereo jack, just like some remote releases for Canon do. The kit also contains a small plastic adaptor to attach the unit to the camera neck strap instead of the accessory shoe (but it will not provide correct heading data in this position). Finally, there is a nylon pouch for storage, and a printed user manual.
Unlike the other Solmeta models, which comes with a cord that plugs into the device, the N3 comes with a coiled cord that is permanently fixed. For someone owning several Nikon bodies with different GPS-plugs, this may be seen as a step backwards, as you cannot move the geotagger to a different body by replacing the cord. However, I've read that the detachable cords on previous Solmeta models was not well-built and that the connection socket was susceptible to breaking, so I think the fixed cord is an improvement.
Since the cord is fixed, Solmeta makes three versions of the device, the only difference between them is that a different cord is used to connect the unit to the camera's accessory terminal.
- Geotagger N3-A can be used with a Nikon D4, D3S, D3, D800, D800E, D700, D300, D300s, D2X, D2XS, D2HS and D200.
- Geotagger N3-B can be used with a Nikon D90.
- Geotagger N3-C can be used with a Nikon D7000, D5100, D5000, D3200 and D3100.
At USD 189 plus another USD 30 for shipping from Hong Kong to Norway, the Solmeta Geotagger N3 is more expensive that most Nikon GP-1 alternatives. Is it worth the extra money? Read on to find out.
My camera is a Nikon D700, so I got the Solmeta Geotagger N3-A. Unlike the units made by rival Dawn Technology, the Solmeta coiled cord on the unit I bought has a plug that sticks out like a sore thumb on the D700 and any other model that uses Nikon's 10-pin plug. While Nikon's GP-1 has the the same design, this is unfortunate as it may prevent you from mounting an L-plate.
Except for the plug angle, the cord seems well made and is of perfect length. The coiled construction gives some flexibility if you want to have the device attached to the camera neck strap instead of the accessory shoe. The 10-pin plug has a screw-lock collar to secure it to the socket. This ensures that the cord will not fall out even inadvertently.
Update 2012-07-30: The European importer of Solmeta geotaggers, MBK GmbH in Germany, has persuaded Solmeta to change this detail, and informs me that it is now able to supply units with the improved N3-A cord (see image to the right).
The N3 itself slides nicely into the Nikon D700's accessory shoe. There is no locking collar, but it fits tightly and is light enough (51 grams) to not slide out easily. If it should fall off, the screw-lock collar on the 10-pin plug will prevent it falling to the ground.
The mechanical construction is not remarkable. The casing is made of plastic. While it has a solid feeling, it is certainly not built like a tank. The 2.5 mm jack socket for the remote release has no cover and the button for calibrating the compass does not have weather seals. I do not think this unit should be used outside in wet conditions.
Setting the unit up is very easy. After unpacking the unit and putting it into the accessory shoe of a Nikon D700 and securing the connecting cord in the camera's 10-pin accessory port, I turned the camera on.
The Solmeta N3 geotagger was instantly detected, my Nikon D700 acknowledged its presence by the letters “GPS” appearing on the top LCD of the camera, as shown on the right. The letters will be flashing when the geotagger is searching for satellites to lock on, and steady when the geotagger is locked on GPS satellites and providing valid GPS data.
After I turned on the camera, the LED at the back of the Solmeta N3 unit immediately started flashing red, which means that it is searching for an initial satellite lock. After about forty seconds, the LED turns into solid green. This means that the unit has obtained a lock on GPS satellites and is ready to provide GPS data to the camera. Except for calibrating the compass (discussed below), this is all there is to setting up the unit.
The only display on the unit is a colour LED on the back. Below is an outline of how the six different states the unit can be is reflected in the colour and mode of this LED.
|LED colour||LED mode||Meter||Description|
|Red||Flashing||On||Searching for GPS signals.|
|Green||Flashing||On||Searching for GPS signals, have 2D lock.|
|Flashing||Off||Sleep mode (camera or meter is off).|
|Steady||On||GPS lock obtained, unit is ready.|
|Green and red||Alternating||On||Indoors mode (remembers last position).|
|Orange||Flashing||On||Compass calibration mode.|
The colour of the LED is simple to see, even in bright daylight.
Because Nikon's own GP-1 and many other geotagging devices draw a lot of current if they are kept on continuously, Nikon's DSLRs has a power-saving function for GPS units. With the device connected (otherwise the GPS sub-menu will not appear) press the Menu button on the D700 and navigate to Setup menu → GPS. Here you will find an option called Auto Meter Off. The default is that this function is enabled. When enabled, the exposure meter is automatically turned off after some time of inactivity (how long can be changed with custom setting c2). Because the exposure meter is linked to powering the GPS unit, the GPS unit will be also be powered down when the camera's exposure meter is off. If you disable this option, the GPS unit will be powered for as long as it is connected to the camera, which means that the geotagger will never enter sleep mode as long as it is connected.
With Nikon's GP-1, disabling this will use up the D700's battery in about a day. With the Solmeta N3, the drain on the batteries in the same time is negligible.
By keeping Auto Meter Off disabled you're telling the N3 GPS unit to remain alive at all times. The benefit here is that the GPS unit will continuously search for satellites in order to keep track of where you are located. This means that as long as you are in range of satellites, the geotagger will be instantly ready for use when you turn on the camera.
Unlike its bigger brother, the more expensive Solmeta Pro 2, the Solmeta N3 does not have a display that shows positions in real time. However, your camera's display can do this. Press the Menu button on the D700 and navigate to Setup menu → GPS → Position. This will give you a live display of GPS data, as illustrated to the right. (This menu option is not available with D2X, D2XS, D2HS, and D200 cameras, and the D200 does not store heading in EXIF.) The unit is very sensitive, so taking a single step in any direction will result in a change in GPS position data.
You can use this display for testing that the device is functional and respond to position changes. The camera Position display is also handy for calibrating the compass.
Unlike the Nikon GP-1, the Solmeta N3 have a built-in 3-axis electronic compass that allows heading data (0-359°) to be stored in the image file along with latitude, longitude, altitude and UTC time. Unlike the four other data points, the heading data is not derived from GPS satellites. There is almost no information about the compass in the technical documentation, but I believe it is a built-in magnetometer that reacts to the earths magnetic field. It will show the direction you are facing when taking a photograph.
The compass requires calibration to be useful. The manual recommends that you calibrate before using the device, and re-calibrate if you move the device to another camera and if you move the more than 100 miles (160 km). The calibration button is located on the side of the device, just behind the coiled cord connection.
The manual's instructions for compass calibration is not very clear, but after some experimentation, I've ended up with the following routine, which seems to work:
- Mount the N3 in the camera's accessory shoe and connect it to the camera. Turn the camera on.
- Press the calibration button briefly. The LED will start flashing orange, indicating that the unit is in compass calibration mode.
- Hold the camera with the N3 mounted in your hands. Rotate the camera slowly with uniform speed 360° twice around X-axis (pitching forward), Y-axis (rotating horizontally), and Z-axis (tilting sideways). Please note, do the rotation twice on each direction, and let each rotation take about 5-10 seconds. Try to rotate the camera around a vertical or horizontal axis.
- After the procedure is completed, press the calibration button briefly again to end the calibration and return the N3 to normal operation.
The Solmeta N3 heading is reported as magnetic north, which differs from true north by an angle known as the magnetic declination. Since the magnetic declination changes over time (and not in a linear fashion), the uncorrected magnetic headings recorded by the Solmeta N3 may be somewhat inaccurate and difficult to use when looking a heading metadata some years after the photo was taken, unless you remember to record magnetic declination for the place and time the photo was taken in some other metadata field. However, the current magnetic declination in Oslo is about 2° 6', so this should not be a major concern.
I used an old Silva Expedition magnetic compass left over from my orienteering days as reference when calibrating. By comparing the heading displayed on the back of the camera's LCD with the heading indicated by the Silva compass, I could immediately verify that the calibration routine had “worked”.
One of reasons calibration is necessary is that the camera body itself is a magnetic object. When I put the Nikon D700 next to the Silva the needle turned about 70° away from magnetic north. Calibration makes the built-in magnetometer in the N3 compensate for the magnetic disturbance from the camera body.
To be frank, first time calibration of the compass was not easy. I had to try several times before the N3 produced sensible heading data. Unfortunately, there is no error indication when the compass is out of calibration and produces bogus heading data, so it helps to have an external reference to verify that calibration has been successful. Also, with practice, calibration becomes easier. Now, I am usually successful the first time.
As far as I am able to tell, once calibrated, the N3 stays calibrated as long as it stays on the camera. When it is moved to a different camera, it needs recalibration.
The Solmeta N3 lived up to its promise of being fast in getting an initial lock on satellites. After turning it on in a new location, I usually had a lock in less within one to two minutes, even in an urban canyon or under dense tree cover. If I had a reasonably clear view of the sky, it would typically lock in on satellites in less than 40 seconds from a cold start. This is a lot faster than Nikon's GP-1.
Also, the power drain on the camera's battery seems to be negligible, even with the Auto Meter Off function disabled.
The picture on the right shows GPS metadata embedded in a photo as it appears on the rear LCD of my Nikon D700. This is from a series of test shots I did of a sculpture of a charging bull that is located in a roundabout in Hegermanns gate in Torshov, Oslo, Norway.
Below is a dump of all GPS information in the same photo extracted with Phil Harvey's ExifTool. It is of course the same as the data shown on the camera's LCD display, but it is more detailed. For instance, it shows that the device was locked in on 10 satellites at the time. It also uses a different format for position data.
GPS Version ID : 188.8.131.52 GPS Latitude Ref : North GPS Longitude Ref : East GPS Altitude Ref : Above Sea Level GPS Time Stamp : 18:24:04 GPS Satellites : 10 GPS Img Direction Ref : Magnetic North GPS Img Direction : 95.04 GPS Map Datum : WGS 84 GPS Date Stamp : 2012:07:20 GPS Altitude : 61 m Above Sea Level GPS Date/Time : 2012:07:20 18:24:04Z GPS Latitude : 59 deg 56' 4.42" N GPS Longitude : 10 deg 46' 0.61" E GPS Position : 59 deg 56' 4.42" N, 10 deg 46' 0.61" E
Below, I look more closely at the speed and accuracy of all the metadata reported by the Solmeta N3 geotagger:
To check the position coordinates, I took the position as reported by the N3, and checked how this position appeared on the satellite images available through Google Maps:
The green arrow on the Google Maps satellite image reproduced below shows my position when I took the photograph of the sculpture located between Hegermanns gate 12 and 14:
If you upload the photo to a photo sharing site that links GPS metadata to map data (e.g. Flickr), the photo may automatically appear on a map, as is shown below. (If you compare Flickr's map below with Google's satellite image above, you'll notice that Flickr's map does not render the roundabout around the bull sculpture very accurate.)
I repeated this test in several locations, and in almost all cases, I found the position information (latitude and longitude) to be spot on.
So far, after about 700 images, I've experienced a single glitch where the position data for a sequence of photos (all taken from roughly the same spot) was scattered around about 40-80 meters away from my true location. The LED was solid green and according to the metadata, the geotagger was fixed on 9 satellites, so I have no real explanation for why the N3 suddenly became inaccurate. Position data for all images taken on the location just before the problematic spot, and the location just after, in the same type of terrain, were accurate to within two meters.
To check the accuracy of the altitude data, I took advantage of the fact that the Norwegian National Rail Administration tags almost every station it operates with its exact altitude. So I took a short trip up to scenic Nordagutu which is located exactly 112.1 meters above average sea level.
When I am standing up with the camera at eye level, the geotagger on top of the camera is 1.7 meters above the ground level, so the altitude for photos taken by me when upright at Nordagutu is 112.1+1.7=113.8 meters. I took 10 photos during my stay at Nordagutu. They were taken from different positions, but all from the same altitude. Browsing the metadata from the N3, altitude reported for this photos varies from 110 meters to 117 meters, with an average of 112.8 meters. The photo on the left, showing the sign on the station, is tagged with 115 meters (i.e. “wrong” by about 1.2 meters). I've since repeated the test with the N3 at other locations where the altitude is known, and gotten similar results.
However, it is well known that GPS units is not always reliable reporters of altitude. Since they compute altitude above average sea levels by computing the distance to satellites using a geodesic model, they are prone to report wildly inaccurate altitudes when all available satellites are near the horizon. In addition, both imperfections in the geodesic model and satellites' deviations from orbit may cause errors. In other words, do not attempt to land a plane using the Solmeta N3 as an altimeter.
When testing, I also found that that the altitude metadata computed by the N3 is less reliable in the first two minutes or so after the green LED has been lit after switching the unit on after arriving at a new location. Then altitude can be off by as much as 20 meters. For accurate altitude data, the unit apparently needs an unobstructed view of the sky, and it takes a longer time to get a satellite fix to produce reliable altitude data than to get a satellite fix to produce an accurate latitude and longitude.
The Solmeta N3 features a 3-axis compass. The three axes are important to make sure heading data is valid no matter what angle you hold the camera.
Out of the box, the compass was useless. The compass needs careful calibration before first time use, or after being moved to a different camera body.
The Silva magnetic compass I used as a reference is small and do not provide very fine grained resolution, but just like the Solmeta N3, the Silva has magnetic north as its 0° reference. As far as I am able to tell, the heading data from the N3 is accurate to within a few degrees. Just shifting the camera a tiny amount resulted in a change in heading of 2-3°, so any inaccuracy may well be down to my fairly crude measurement technique. For more precise measurements, I would need a larger compass and to have the camera + geotagger on a tripod with a locking ball-head. However, the accuracy of heading data produced by the N3 proved to be more than good enough for knowing in what general direction the camera was pointing when the photo was taken.
The time-stamp is not from the camera's clock, but is standard time (UTC) from the satellites. It has atom-clock accuracy. You can convert it to local time if you know the offset between UTC time and local time. In the case of Norway, the local time is CET summertime, which is is UTC +2 hours.
I also made a few tests to verify that the included corded remote shutter release worked. It did, but it has a plastic feeling and is probably not a high quality product. It is not a big deal, I probably will not need to use a remote shutter release and a geotagger at the same time very often.
The Solmeta N3 lives up to its promise of being a fast GPS geotagging unit with a low drain on the camera's battery. Save for a single glitch, it is very accurate in reporting position (latitude and longitude) and UTC time. It is less accurate when reporting altitude and heading, but these data is nevertheless good enough to be useful.
It carries a higher price tag than most of the other third party alternatives to the Nikon GP-1 geotagger, but in my opinion the feature set, the accuracy, speed and low power consumption is worth the extra money.