Garmin has advertised that new (Edge for cycling and Forerunner for running) 205 & 305 models will perform much better in the woods and in urban canyons. There has also been a claim that the new chip technology will also enable quicker satellite acquisition. Do the new units deliver on that promise?
My test protocol was to run multiple laps of a one-mile loop wearing both the new Edge 305 and my Forerunner 201. I’ve measured (and remeasured) the loop with a wheel and believe it to be pretty close to 1.01 miles. For 5 laps (with the Edge carried in an iPod holster on the upper arm and the old Forerunner head-mounted for optimal accuracy: The Edge recorded 5.04 miles and the Forerunner 5.07 miles. So, perhaps, the Edge is a bit more accurate.
A clearer indication of the performance of the two units can be seen in how accurately the recorded GPS tracks follow the actual path run and the consistency of each lap’s track. In the two illustrations, I have highlighted two areas where there is a noticeable improvement with the new Edge unit.
|
|
|
In the Woods. The oval highlights a section where the trail winds through fairly thick woods. Notice how the Edge tracks are more consistent. The Edge tracks also come closer to following the actual path of the trail.
Around Buildings. The rectangle highlights where the trail runs along the side of a gymnasium and then makes a sharp turn around the corner of the building. On four out of five laps the Edge does a much better job of following the actual trail around the corner of the building. One lap shows an odd zig zag… perhaps the result of a GPS signal echoing off the side of the gym?
In the Open. Curiously, the tracks of the head-mounted Forerunner are tighter (each track more closely overlays the other tracks) in open terrain than those for the arm-mounted Edge. On the other hand, the Edge tracks follow more accurately the actual path. Perhaps even with the more sensitive chip set, the human body still plays a significant role in blocking GPS satellite signals. Further testing is needed to verify whether the new units are as sensitive to where they are worn as the prior Forerunner series.
In Doors. Barely… I was able to get the system to acquire satellites when placed on the windowsill of a large bay window. Away from the window, the Edge will not acquire enough satellites to work in my wood-frame house.
Satellite Acquisition. The new chip set does not acquire satellites any quicker on a cold start, e.g., at a new location or after several days. A restart at the same location, even after a few hours, is very quick. There is a neat new feature… one of the custom displays available is an estimate of location accuracy. I have noticed that the satellite acquisition display clears as soon as the device has you generally located. Watching the accuracy field, I see the estimated accuracy continue to improve as the device continues to acquire more satellites for another minute or more, from 200 feet to less than 30 feet.
Barometric Altimeter. This feature is only available with the Edge 305, i.e., will not be available in the Forerunner line. It promises to enable much more accurate elevation profiles of runs and rides. My initial tests with the Edge in a iPod holster worn on the arm were disappointing. The altimeter was not at all accurate... displayed altitudes were off by hundreds of feet. I have had far better results with the Edge fully exposed to the air... attached to a band of a headset radio using the Garmin-supplied mounting bracket and cable ties. The altimeter sensor is on the back side of the device and appears to need to be fully exposed to work correctly. More later...
On April 17th, I ran 20 laps of a certified 1.77 mile loop in Delaware Park in Buffalo, NY. The event was the April 17th running of the BPAC Six Hour Distance Classic. The table below shows various distances recorded for my 20 laps. Both GPS values are longer than the official race distance but are surprisingly close. I know I ran slightly farther than the official distance as I did not always follow the shorted path, for example as I avoided some highly crowned sections of the road.
| Forerunner display at race end | 35.70 miles |
| TopoFusion total from track data | 35.66 miles |
| Official race distance (20 x 1.77 mi) | 35.40 miles |
The TopoFusion screen snap below shows the 20 laps overlaid on an aerial photo of the park. Conditions for accurate results were near optimal... a head-mounted Forerunner, limited tree cover and no tall building to block view of the sky. The single spur track at the bottom of the picture is not an error... the start was offset about 20 feet from the track followed on all the other laps. The wider track along the bottom of the track is also not an error... my running track varied in that section during the course of the race.
The next screen snap shows results of the very nice Lap Analysis feature in TopoFusion. With the Forerunner attached to the head band of my headset radio, it was not convenient to record laps. This was no problem as, after the fact, TopoFusion can extract lap distances & times from the GPS track data. Notice the "pretty good" accuracy of the results. Every lap fell with the range 1.76 to 1.79 miles. Not bad repeatability at all. The Stopped Time on laps 7 and 15 were a couple of porta-potty visits and tell me just how much time I lost on those laps.
It is well known that GPS altitude measurements are not very precise. The TopoFusion screen snap below shows the GPS profile for the same 20 laps. The profile shown is a bit misleading. Notice that difference between the highest point and lowest point on the course is only 40-50 feet. Errors in the GPS altitude measures show in the profile and inflate the Total Climbing and Total Descent values.
The TopoFusion application includes a Climbing Analysis feature. The Straight DEM (digital elevation model) value is obtained by using 3D elevation data downloaded from a NASA JPL server. The 965 feet seems about right... if there is actually about 45 feet difference between the highest point and the lowest point, then 20 laps x 45 feet would give 900 feet of total climbing.
Clicking the Add Elevation to Track button will permanently add the downloaded DEM data to your GPS track data. The snap below shows the profile of the same 20 laps with downloaded elevation data. Pretty boring, eh?
Tom Perry
April 23,2005