GPS Accuracy: Suunto's SPARTAN SPORT *AND* ULTRA vs TomTom's Runner 3 (Spark 3)

Suunto Spartan Ultra vs. Suunto Spartan Sport vs TomTom Runner 3.

Well, you’d expect two identical lines and one that is a bit different wouldn’t you? Two near-identical Suuntos?

Not quite. This is my inadvertent test to belatedly see how the ULTRA handles GPS Ephemeris Data.

I’m tapering, so this was a short run (4k) and it was after an easy bike faffing around for an hour to ‘warm up’ the GPS.

All 3 watches were worn ‘properly’. The ones on my left wrist was the TomTom and the Suunto Spartan Sport. The Ultra was on the other wrist. So they were 50cm or so closer to the tree cover that I encountered!!! Not much of a factor but I mention it anyway, no doubt my body could get in the way ‘a bit’ from a couple of satellites.

This map is zoomed out, losing detail. But which watch is which?

In places they were near-identical. this is a thin, man-made track through grass. So they are within a metre of each other, with no tree cover, towards the end.

Yet with the image on the right we are running under tree cover near the start. The blue and green lines are probably both about right and the red is a good 5m out. Maybe 10m in places. That must be the TomTom right? (Obviously the answer is going to be no!)

The red line is the Spartan ULTRA. All 3 watches were on the latest firmware and, as I said earlier, they had definitely all acquired the GPS signal with the relavent icon showing.

We’d seen as well in <this> and <this> test that the Spartan ULTRA probably OUTperformed the other watches on those occasions, just. But not this time. Strange.

What to make of it? Well the ULTRA *WAS* on a different wrist to the other two. Albeit the supposedly BETTER wrist.

As well I think it reasonable to expect some machine-level inaccuracy ie if you compared 10 identical devices then they would all be different in their GPS tracks.

To expect REPEATABILITY is nice but maybe we expect too much for an inherently difficult thing.

Edit: I think I have the explanation. For the ‘warm-up’ bike I forgot to turn on the ULTRA. so what I might be seeing here with the inaccuracy of the ULTRA at the start (in the small portrait orientated image, above) is it collecting GPS ephemeris data (see below or google it). It is NOT collected on the watch if the watch is ‘just’ worn as a watch in open air, it has to be looking for a GPS signal ie with a sport profile running. This is more than just assisted-GPS which most devices have, including Suunto. All devices will have to do this in some form (A GPS Watch company employee suggested to me that some sports watch companies will turn watches on to gather then info without you knowing, so that the GPS is fully accurate from the start).

I’ll keep plugging away. But remember that the few runs/rides that I make DO NOT constitute a scientific test. they are just my experience, warts and all.

The Moral Of This Story: Do a 15 minute warmup of you and the watch.

Key:

Blue=Runner 3

Red=ULTRA

Green=SPORT

Read This, from <here>:

Almanac and Ephemeris Data as used by GPS receivers
(4 July 1998)

The  satellites  broadcast  two  types  of  data,   Almanac   and
Ephemeris.   Almanac  data is course orbital parameters  for  all
SVs.  Each SV broadcasts Almanac data for ALL SVs.  This  Almanac
data  is  not  very precise and is considered  valid  for  up  to
several  months.   Ephemeris data by comparison is  very  precise
orbital  and  clock correction for each SV and is  necessary  for
precise  positioning.  EACH SV broadcasts ONLY its own  Ephemeris
data.   This data is only considered valid for about 30  minutes.
The Ephemeris data is broadcast by each SV every 30 seconds.

When the GPS is initially turned on after being off for more than
30 minutes,  it “looks” for SVs based on where it is based on the
almanac  and current time.  With this  information,   appropriate
SVs can be selected for initial search.

When  the  GPS receiver initially locks onto a  SV,   the  Garmin
display then shows “hollow” signal strength bars.  At this  time,
the Ephemeris data has yet to be completely collected.  Once  the
ephemeris data is collected from EACH SV in turn,  the associated
signal  strength  bar will turn “solid” black and then  the  data
from that SV is considered valid for navigation.

If power is cycled on a GPS unit,  and when turned back on,   the
Ephemeris data is less than 30 minutes old,  lock-on will be very
quick since the GPS does not have to collect new Ephemeris  data.
This is called a “warm” start.

If  it  is later than 30 minutes,  this is  considered  a  “cold”
start and all Ephemeris data will have to be recollected.

If  the GPS has moved more than a few hundred miles  or  accurate
time  is lost,  the Almanac data will be invalid and if  you  are
far  enough off,  none of the SVs that the Almanac thinks  should
be  overhead will be there.  In such case,  the GPS will have  to
“sky search” or be reinitialized so it can download a new Almanac
and start over.

(Note:   Yes!  We know this is somewhat  simplified  information.
Yes,  we know that the Ephemeris data may not have to be  updated
as  often  as  the  G-12XL does it to  get  data  to  the  G-12XL
accuracy.)