I really, really wanted to like the Garmin Fenix 2. I’ve been using a sturdy, but slowly disintegrating Forerunner 305 since 2006 or whenever they came out. My latest one has a dead button and doesn’t sync reliably, plus a short battery life. So, I was looking for a go-to GPS unit for my running, and figured it would be the Fenix 2 since I do a little bit of navigating on the watch, the 910 and 920 don’t seem up to that task, since you can’t put courses or track on them, from what I’ve read (and from what Garmin told me on the phone.)
The Fenix 2 has gotten a lot of press for its claimed 50 hour battery life. It’s also gotten absolutely pounded by dozens (if not hundreds) of users online for being a pretty lousy device as far as GPS accuracy and precision go, not to mention random lockups during really long events, and other seizures that erase hours worth of hard effort for the endurance athletes Garmin is targeting with this device.
Bad first impression
My own Fenix 2 experience didn’t start well. The first time I plugged the thing into my computer, it pulled up Garmin express then the watch screen launched into some kind of debug/system mode and seemed to lock up. It took me 20 minutes of fiddling with it to get it to restart and behave normally. I downloaded the latest firmware, installed, and thought I was good to go.
After that, I took a hike with my little girl, wearing both the new Fenix 2 and the old Forerunner 305. We were on the typical, lightly-wooded trail, most leaves are down, and in 3 miles or so, the Fenix 2 managed to read something like 5% long. I know the trail well, and knew the 305’s distance to be pretty close.
I turned off the WAAS settings in the Fenix and the next couple of runs, it seemed to come in closer to the distances that I know from experience (hundreds of bike and running trips) to be accurate.
Not so great later in the week, either
So, I decided to try the Fenix 2 out at Saturday’s Rehoboth Beach Marathon. I wore both the Fenix 2 and the ForeRunner 305 on my left arm. On my right arm, I wore a Microsoft Band, which is a GPS-capable fitness band, totally unsuited for use as a race watch, but fun to have to count steps on marathon day.
To make the long story short, the Fenix’s distance fell apart pretty quickly. The course was along salt marshes on the seashore for the first half, with a little bit of sparsely-pined sand dunes. The marshes are pretty much wide open. The dune tree cover is pretty light.
The Fenix read a little bit longer than the 305 (and the MB) during the first, open-sky part of the race. The MB was definitely closest to hitting the mile markers correctly. But for the first half of the race, the error seemed to be acceptable, within the normal, half-percent range (long, always) that I’m used to from the Garmin 305. (I’ve analyzed all my marathons from that unit, and it has averaged half a percent long.)
Why long? Well, when the GPS track goes haywire, it tends to zig and zag a bit, and with the unit assuming you’re running the zigzag and not the straight line you are actually running, the distance adds up.
However, the Fenix 2 really began to fall apart at about mile 13 when we headed back into the light tree cover. Soon, it was alerting me to half mile splits/laps that were a long, long way in front of the mile markers. Looking at the GPS tracks, you can see the Fenix 2 plotting a really horrible line compared to the much older Forerunner.
Pretty firm evidence
Here’s the half mile around mile 13, Forerunner 305 on the top, Fenix 2 below. Look how the out and back from the old unit are just about right on top of each other, and how they’re nice and straight, tracking my line on the trail. Compare that to the horrendous track by the expensive, new Fenix 2, below, which zigzags and tracks off into the woods from my real line. As you can see, this is not a heavily wooded trail. The trees are short and they don’t even cover the trail.
Here’s the last turnaround, as recorded by the Forerunner 305 on the left, the Fenix 2 on the right.
Here’s the half mile near mile 21 and 25, as recorded by the Forerunner 305 on the left, Fenix 2 on the right.
I think this section of out-and-back shows just how bad the Fenix 2’s GPS traces are. As soon as I headed into the light tree cover at the side of the field, the unit completely lost track of me, sticking me 10-15 meters to either side of the trail on some of those turns. This unit comes from Garmin’s outdoor division, so this should be especially humiliating to them. What good is a multisport, hiking GPS unit that can’t do better than this, that can’t do better than a 6 year-old Garmin GPS that’s on its last legs?
By the end of the race, the Fenix had the 26.22 mile route at 25.57 miles. The Forerunner 305 had it at 26.33 miles. The Microsoft Band read 26.31. So, the Fenix 2 was off by 0.35 miles, or almost 1.5%. That might not sound like much but it’s wrong by more than 2 minutes worth of effort at a 6 minute per mile marathon pace. That’s huge if you’re depending on your watch to give you anything close to useful information on the course. With the Forerunner 305, knowing it is usually long by 0.1 to 0.2 of a mile, I know the average pace displayed on the watch is within 2-3 seconds of my real pace. With the Fenix 2, it was off by 5-7 seconds, even more after particularly gnarly half-mile traces.
GPS inaccuracy is only the tip of the iceberg with this thing.
Something I noticed early on is that the Fenix 2 is finicky about how you press the buttons. This is new to me. If you click the button enough to feel it and hear the click, it still may not actually register the click. (No, I’m not talking about the press-and-hold functions as for the menu button. I’m talking about single-click functions, like the all-important start-stop button.)
At the end of my marathon yesterday, the Fenix was the first watch I turned off. However, a couple minutes after I’d crossed the line, as I stood chatting with a friend in the chute and lamenting that I’d missed a new PR by only 2 seconds (thanks to the inaccuracies of the Fenix, I had plenty of gas to spend to pick up those seconds, but the Fenix’s data was so far off, I didn’t bother using it, thinking it would be a waste of time, and at the end of a 110-mile week, an unnecessary injury risk…) I noted the Fenix 2’s clock was still running. It hadn’t registered the button press. Think I’m making this up? Watch this:
Finally, and perhaps worst of all, has one of the most slow and unfriendly user interfaces I’ve ever used in an electronics device. Just try and take a look at an event you’ve completed. Words cannot begin to describe just how bad the Fenix 2 is at showing you the information you have paid $450 to record. Words fail, but maybe this video won’t.