I'm going to write about the GPS behavior of SRAM's Quarq.
Based on my actual experience with equipment,
I've used Polar's heart rate monitors with GPS function and
Pioneer's SGX-CA500, and
when I set running speed and distance based solely on GPS,
it's clear that running distance is (assuming speed updates every second)
NOT an accumulation of speed readings moment by moment.
While current speed displays the result of GPS communication,
distance measurement is being done separately from that.
Polar's running distance updates in 0.01 km (10 meter) increments,
but the current speed shows a "lingering" effect for several seconds after stopping.
What I mean is, for example, if you go from 30 km/h
and stop at a traffic light,
the display shows 28.2...21.3...17.8...9.6...
3.4.....0 km!
Like that—even though you're clearly stopped,
the displayed speed just won't drop to zero quickly.
If running distance were based on the accumulation of displayed speed,
this "lingering" accumulation should result in 10 meters of distance
and add 0.01 km during the stop,
but I've never once observed that happening.
As more definitive evidence:
when entering a reasonably long tunnel (depending on speed, 200+ meters),
the GPS signal is lost, and
when it's reacquired just after exiting the tunnel,
the running distance suddenly increases by roughly the tunnel's length.
If the tunnel is 600 meters long, about 600 meters gets added,
with running distance jumping as the two GPS points (signal lost and reacquired) are connected.
Even if the actual tunnel trajectory curves somewhat,
since it's connected as a straight line, some error occurs,
but it's not really a problem.
If running distance were an accumulation of speed display,
it would be strange for the 0 km/h reading during the tunnel
to suddenly add the straight-line distance amount when exiting the tunnel.
About butterfly mode.
With GPS position measurement,
when you stand completely still without moving,
the GPS current position history
doesn't stay fixed at one point—
instead, like a pen tip scribbling within a narrow range,
it traces a line that looks like a butterfly's flight path.
I call this "butterfly mode,"
and it seems the difference in how GPS signals are processed
lies with the receiving end (the cycle computer side).
Polar's drift is more like a snail
rather than a butterfly,
while Pioneer's CA500
flutters around quite a range (radius of tens of meters).
As for SRAM's Quarq—it was quite an energetic butterfly.
After acquiring the GPS signal
and the map appeared on the screen,
I started measuring without moving it at all.
In the image above, speed shows 0.0 km,
but without moving or touching it at all,
sometimes it shows 1.4 km or similar.
Even while stationary, the speed doesn't stay at 0.0 km—
this is the same with Pioneer's CA500.
In this respect, Polar does it well:
although there's a long lingering effect when stopped, once it shows 0.0 km,
unless you clearly start moving (probably around 3 km/h),
the display stays at 0.0 km.
About 13 minutes after starting measurement,
it was already showing 0.1 km traveled,
so I stopped measuring.
Looking at the movement history,
the butterfly illegally trespassed to the trophy shop across the street (→here).
I forgot to capture it, but zooming in more on the map
shows a chaotic trajectory around the starting point.
As for this distance, excluding the butterfly trajectory,
the simple straight-line distance is less than 0.1 km.
Polar's sensors like heart rate, speed, and cadence
only support Bluetooth.
Incidentally, two generations ago they only supported their own proprietary standard.
Quarq supports Bluetooth and also
ANT+ in addition.
ANT+ broadcasts information to all nearby receiving devices
when there are multiple pairing targets on the transmitting side,
but Bluetooth is one-to-one,
so in the image above, since I connected with Polar first,
the Quarq doesn't display heart rate.
When I disconnected the heart rate monitor, heart rate immediately appeared on the Quarq side.
Once that happens, unless you disconnect Quarq's connection,
heart rate won't display on Polar's side.
The reason I continue to support Polar no matter how its market share narrows in the cycling world
is because it's the original pioneer of heart rate monitors in sports,
and especially because I trust its calorie burn calculation.
So even now, I gauge other manufacturers' calorie burn from devices with heart rate function as
"higher/lower than Polar."
Though most people probably don't know this anymore,
there was a long era when heart rate monitors from anyone but Polar
were essentially toys.
Well, compared to actual medical devices,
maybe Polar is toy-like too.
With Polar, you first input your date of birth.
This automatically becomes an age field that updates.
Additionally, by inputting gender, height, weight data and
self-reported exercise frequency (hours per week),
the calorie burn value is
"for someone of this age, gender, height, and weight
exercising at this reported frequency level
at this heart rate,
the calorie burn per measurement interval would be this amount,"
and those previously mentioned items become the calculation basis.
But with Quarq, the only item required for calorie burn calculation is body weight,
and the calculation basis is actually power output.
In other words, even if you display heart rate on Quarq,
it's not used as the basis for calorie burn calculation.
And calorie calculation requires a power meter.
Since I haven't adopted a power meter,
I can't conduct ongoing research on whether Quarq's power-based calorie burn
tends to be higher/lower than Polar.
Well, I'll adopt one eventually.
So currently, I've
removed the heart rate transmitter from
Quarq's pairing.
I have Quarq mounted on my disc road bike,
which is my regular-use bike and
the only one with a 31.8 mm handlebar clamp diameter
(besides that there's a sub-4 kg bike with 31.8 mm,
others are 26.0 mm or 25.8 mm).
For commuting on the disc road, I don't use the heart rate monitor,
and with heart rate monitors it's annoying to delete logs,
but Quarq lets you choose to save or delete logs at the end of measurement,
so for commutes I usually don't save them.
When I want to keep GPS logs from rides of meaningful distance,
I also use the Polar heart rate monitor.
I want to manage my ride logs all in one place
in Polar Flow (Polar's cloud management software),
and while exporting Quarq's ride logs to Strava is easy—basically a service the manufacturer provides—
transferring to Polar Flow is somewhat tedious due to file extension issues. I'll cover that in a separate post.
So before I start riding, I confirm both Polar and Quarq
have acquired GPS signal, and
start logging with just a 2-3 second difference,
then ride off. The GPS distance data
doesn't differ drastically between them,
proving both are accurate, but
I've noticed a certain tendency.
Polar displays distance in 0.01 km increments,
while Quarq displays in 0.1 km increments.
When Polar's second decimal place is 7-9,
Quarq's distance is larger,
and when it's 1-3, Polar's distance is larger—
that's the tendency I see.
For example, when Polar shows 9.27 km,
Quarq shows 9.3 km.
This isn't just coincidence—
it's based on sustained observation over time.
Somehow, Quarq's running distance seems to be
about 0.05 km (50 meters)
ahead of schedule
(more precisely, it likely rounds the 0.05–0.14 km range to 0.1 km).
So when I started Polar and Quarq as close to simultaneously as possible (within 1 second)
and rode straight and carefully, I confirmed that
when Polar switches from 0.03 km to 0.04 km,
Quarq shows 0.1 km.
I've tested this several times.
Even without using Polar simultaneously,
I've confirmed that Quarq shows 0.1 km within 0.06 km (60 meters)
even when riding somewhat faster.
↑Like this, for routes I've created or pins I've placed,
the remaining distance to the goal often differs by 0.1 km
between the numeric distance display and
the upcoming elevation profile,
which suggests the rounding method for distances under 0.1 km
differs depending on the display item.
So that butterfly trajectory I mentioned earlier
that "trespassed" to the trophy shop across the street
is probably actually about 70–80 meters of distance.
Even during a ~1 minute stop at a traffic light,
we can't rule out butterfly mode occurring,
but even over about 100 km of riding,
Polar and Quarq's distances maintain a state where
Quarq is about 0.05 km ahead
and never diverge significantly.
I have the auto start/stop function (which automatically stops time tracking when stopped) turned off
on both Polar and Quarq, and don't stop for brief traffic lights,
so during traffic waits both experience fine butterfly mode similarly,
and they might appear correct simply because both are equally off.
From sustained observation of butterfly mode,
Polar seems to have less variance
(it's snail-like as I mentioned before).
SRAM Quarq is expected to become more prevalent
as an accessory with the new RED (E1) going forward,
but Quarq standalone is already available from distributors,
so I'm having fun verifying it.
Separate from the program named "Baltan Starmen,"
I created another program.
It's on the right, so I swipe.
A two-line Zaku II emoji art made with the character range available in Quarq's text input.
Please accept this.
Based on my actual experience with equipment,
I've used Polar's heart rate monitors with GPS function and
Pioneer's SGX-CA500, and
when I set running speed and distance based solely on GPS,
it's clear that running distance is (assuming speed updates every second)
NOT an accumulation of speed readings moment by moment.
While current speed displays the result of GPS communication,
distance measurement is being done separately from that.
Polar's running distance updates in 0.01 km (10 meter) increments,
but the current speed shows a "lingering" effect for several seconds after stopping.
What I mean is, for example, if you go from 30 km/h
and stop at a traffic light,
the display shows 28.2...21.3...17.8...9.6...
3.4.....0 km!
Like that—even though you're clearly stopped,
the displayed speed just won't drop to zero quickly.
If running distance were based on the accumulation of displayed speed,
this "lingering" accumulation should result in 10 meters of distance
and add 0.01 km during the stop,
but I've never once observed that happening.
As more definitive evidence:
when entering a reasonably long tunnel (depending on speed, 200+ meters),
the GPS signal is lost, and
when it's reacquired just after exiting the tunnel,
the running distance suddenly increases by roughly the tunnel's length.
If the tunnel is 600 meters long, about 600 meters gets added,
with running distance jumping as the two GPS points (signal lost and reacquired) are connected.
Even if the actual tunnel trajectory curves somewhat,
since it's connected as a straight line, some error occurs,
but it's not really a problem.
If running distance were an accumulation of speed display,
it would be strange for the 0 km/h reading during the tunnel
to suddenly add the straight-line distance amount when exiting the tunnel.
About butterfly mode.
With GPS position measurement,
when you stand completely still without moving,
the GPS current position history
doesn't stay fixed at one point—
instead, like a pen tip scribbling within a narrow range,
it traces a line that looks like a butterfly's flight path.
I call this "butterfly mode,"
and it seems the difference in how GPS signals are processed
lies with the receiving end (the cycle computer side).
Polar's drift is more like a snail
rather than a butterfly,
while Pioneer's CA500
flutters around quite a range (radius of tens of meters).
As for SRAM's Quarq—it was quite an energetic butterfly.
After acquiring the GPS signal
and the map appeared on the screen,
I started measuring without moving it at all.
In the image above, speed shows 0.0 km,
but without moving or touching it at all,
sometimes it shows 1.4 km or similar.
Even while stationary, the speed doesn't stay at 0.0 km—
this is the same with Pioneer's CA500.
In this respect, Polar does it well:
although there's a long lingering effect when stopped, once it shows 0.0 km,
unless you clearly start moving (probably around 3 km/h),
the display stays at 0.0 km.
About 13 minutes after starting measurement,
it was already showing 0.1 km traveled,
so I stopped measuring.
Looking at the movement history,
the butterfly illegally trespassed to the trophy shop across the street (→here).
I forgot to capture it, but zooming in more on the map
shows a chaotic trajectory around the starting point.
As for this distance, excluding the butterfly trajectory,
the simple straight-line distance is less than 0.1 km.
Polar's sensors like heart rate, speed, and cadence
only support Bluetooth.
Incidentally, two generations ago they only supported their own proprietary standard.
Quarq supports Bluetooth and also
ANT+ in addition.
ANT+ broadcasts information to all nearby receiving devices
when there are multiple pairing targets on the transmitting side,
but Bluetooth is one-to-one,
so in the image above, since I connected with Polar first,
the Quarq doesn't display heart rate.
When I disconnected the heart rate monitor, heart rate immediately appeared on the Quarq side.
Once that happens, unless you disconnect Quarq's connection,
heart rate won't display on Polar's side.
The reason I continue to support Polar no matter how its market share narrows in the cycling world
is because it's the original pioneer of heart rate monitors in sports,
and especially because I trust its calorie burn calculation.
So even now, I gauge other manufacturers' calorie burn from devices with heart rate function as
"higher/lower than Polar."
Though most people probably don't know this anymore,
there was a long era when heart rate monitors from anyone but Polar
were essentially toys.
Well, compared to actual medical devices,
maybe Polar is toy-like too.
With Polar, you first input your date of birth.
This automatically becomes an age field that updates.
Additionally, by inputting gender, height, weight data and
self-reported exercise frequency (hours per week),
the calorie burn value is
"for someone of this age, gender, height, and weight
exercising at this reported frequency level
at this heart rate,
the calorie burn per measurement interval would be this amount,"
and those previously mentioned items become the calculation basis.
But with Quarq, the only item required for calorie burn calculation is body weight,
and the calculation basis is actually power output.
In other words, even if you display heart rate on Quarq,
it's not used as the basis for calorie burn calculation.
And calorie calculation requires a power meter.
I can't conduct ongoing research on whether Quarq's power-based calorie burn
tends to be higher/lower than Polar.
Well, I'll adopt one eventually.
So currently, I've
removed the heart rate transmitter from
Quarq's pairing.
I have Quarq mounted on my disc road bike,
which is my regular-use bike and
the only one with a 31.8 mm handlebar clamp diameter
(besides that there's a sub-4 kg bike with 31.8 mm,
others are 26.0 mm or 25.8 mm).
For commuting on the disc road, I don't use the heart rate monitor,
and with heart rate monitors it's annoying to delete logs,
but Quarq lets you choose to save or delete logs at the end of measurement,
so for commutes I usually don't save them.
When I want to keep GPS logs from rides of meaningful distance,
I also use the Polar heart rate monitor.
I want to manage my ride logs all in one place
in Polar Flow (Polar's cloud management software),
and while exporting Quarq's ride logs to Strava is easy—basically a service the manufacturer provides—
transferring to Polar Flow is somewhat tedious due to file extension issues. I'll cover that in a separate post.
So before I start riding, I confirm both Polar and Quarq
have acquired GPS signal, and
start logging with just a 2-3 second difference,
then ride off. The GPS distance data
doesn't differ drastically between them,
proving both are accurate, but
I've noticed a certain tendency.
Polar displays distance in 0.01 km increments,
while Quarq displays in 0.1 km increments.
When Polar's second decimal place is 7-9,
Quarq's distance is larger,
and when it's 1-3, Polar's distance is larger—
that's the tendency I see.
For example, when Polar shows 9.27 km,
Quarq shows 9.3 km.
This isn't just coincidence—
it's based on sustained observation over time.
Somehow, Quarq's running distance seems to be
about 0.05 km (50 meters)
ahead of schedule
(more precisely, it likely rounds the 0.05–0.14 km range to 0.1 km).
So when I started Polar and Quarq as close to simultaneously as possible (within 1 second)
and rode straight and carefully, I confirmed that
when Polar switches from 0.03 km to 0.04 km,
Quarq shows 0.1 km.
I've tested this several times.
Even without using Polar simultaneously,
I've confirmed that Quarq shows 0.1 km within 0.06 km (60 meters)
even when riding somewhat faster.
↑Like this, for routes I've created or pins I've placed,
the remaining distance to the goal often differs by 0.1 km
between the numeric distance display and
the upcoming elevation profile,
which suggests the rounding method for distances under 0.1 km
differs depending on the display item.
So that butterfly trajectory I mentioned earlier
that "trespassed" to the trophy shop across the street
is probably actually about 70–80 meters of distance.
Even during a ~1 minute stop at a traffic light,
we can't rule out butterfly mode occurring,
but even over about 100 km of riding,
Polar and Quarq's distances maintain a state where
Quarq is about 0.05 km ahead
and never diverge significantly.
I have the auto start/stop function (which automatically stops time tracking when stopped) turned off
on both Polar and Quarq, and don't stop for brief traffic lights,
so during traffic waits both experience fine butterfly mode similarly,
and they might appear correct simply because both are equally off.
From sustained observation of butterfly mode,
Polar seems to have less variance
(it's snail-like as I mentioned before).
SRAM Quarq is expected to become more prevalent
as an accessory with the new RED (E1) going forward,
but Quarq standalone is already available from distributors,
so I'm having fun verifying it.
Separate from the program named "Baltan Starmen,"
I created another program.
It's on the right, so I swipe.
A two-line Zaku II emoji art made with the character range available in Quarq's text input.
Please accept this.