Edge 705 and ant+ Powertap

[Howard]

I read somewhere that there is an issue with auto pause on the Edge
and loading .tcx files into WKO+. What is the specific issue? Is
there also an issue with turning auto pause off and pushing the start/
stop button? Thank you.
--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[Dave Harris]

It's more of a feature than an issue...

Stopped time is not recognized as such. It doesn't matter if auto-pause is
on or off - due to the "smart recording" feature of the 705 wko currently
assumes there is no stopped time. Smart recording stores points at random
intervals based on when it determines one is needed (and not before) but
that does not mean the time between points was stopped time.

Hence the conundrum for WKO programmers. I am pretty sure they are working
on it ;)

I have not tried manually pausing the Edge so can't say what happens there.

> -----Original Message-----
> From: ***@googlegroups.com [mailto:***@googlegroups.com] On
> Behalf Of Howard
> Sent: Monday, December 15, 2008 7:41 AM
> To: Wattage
> Subject: [Wattage] Edge 705 and ant+ Powertap
>
>
> I read somewhere that there is an issue with auto pause on the Edge
> and loading .tcx files into WKO+. What is the specific issue? Is
> there also an issue with turning auto pause off and pushing the start/
> stop button? Thank you.


--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[Tom_Anhalt]

On Dec 15, 7:01 am, "Dave Harris" <***@gmail.com> wrote:
> It's more of a feature than an issue...
>
> Stopped time is not recognized as such.  It doesn't matter if auto-pause is
> on or off - due to the "smart recording" feature of the 705 wko currently
> assumes there is no stopped time.  Smart recording stores points at random
> intervals based on when it determines one is needed (and not before) but
> that does  not mean the time between points was stopped time.
>
> Hence the conundrum for WKO programmers.  I am pretty sure they are working
> on it ;)
>
> I have not tried manually pausing the Edge so can't say what happens there.
>

I have ;-) It doesn't appear to make a difference in WKO+ if you
manually hit the "stop" button when you're paused...well, unless you
happen to forget to hit the button again when you start up, that
is :-)

Aaah...that makes sense about the conundrum caused by the so-called
"smart-recording" being the issue. Perhaps the WKO+ folks can give an
option to specify if it was so-called "smart" recorded or 1s recorded
when bringing in a .tcx file? If it's so-called "smart" recorded,
then ignore the autopauses? I don't know...just speculating "out
loud" here...I'm sure there's more to it than that..

Even so...I've quickly "adapted" to just going into the WKO+ chart and
highlighting and "cutting" out the offending sections. I don't worry
about stops at signal lights and such...but just cut out the longer
"pauses". Let's put it this way...it's not what I'd consider a "deal
breaker" in the use of a 705 to record a PM output. Annoying?
Yes...Deal breaker? No. YMMV :-)
--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[r***@aol.com]

Just a note of clarification here as well.

DO NOT use smart recording if you actually want any meaningful power data.
Random data recording is just that... random.
I already have a random number generator on my bike(HAC4).....

1 second recording if you have a 705.

And yes, we are working on dealing with all the 705 issues.




Hunter

Buy your Wattage based training plan ONLINE!
www.trainingpeaks.com/hunter

The Peaks Coaching Group
"We care about your success!"
www.peakscoachinggroup.com

TrainingPeaks WKO+
The Ultimate Analysis Software
www.trainingpeaks.com


-----Original Message-----
From: Dave Harris <***@gmail.com>
To: ***@googlegroups.com
Sent: Mon, 15 Dec 2008 10:01 am
Subject: [Wattage] Re: Edge 705 and ant+ Powertap




It's more of a feature than an issue...

Stopped time is not recognized as such. It doesn't matter if auto-pause is
on or off - due to the "smart recording" feature of the 705 wko currently
assumes there is no stopped time. Smart recording stores points at random
intervals based on when it determines one is needed (and not before) but
that does not mean the time between points was stopped time.

Hence the conundrum for WKO programmers. I am pretty sure they are working
on it ;)

I have not tried manually pausing the Edge so can't say what happens there.

> -----Original Message-----
> From: ***@googlegroups.com [mailto:***@googlegroups.com] On
> Behalf Of Howard
> Sent: Monday, December 15, 2008 7:41 AM
> To: Wattage
> Subject: [Wattage] Edge 705 and ant+ Powertap
>
>
> I read somewhere that there is an issue with auto pause on the Edge
> and loading .tcx files into WKO+. What is the specific issue? Is
> there also an issue with turning auto pause off and pushing the start/
> stop button? Thank you.





--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[Tom_Anhalt]

On Dec 15, 7:29 am, ***@aol.com wrote:
> Just a note of clarification here as well.
>
> DO NOT use smart recording if you actually want any meaningful power data.
> Random data recording is just that... random.
> I already have a random number generator on my bike(HAC4).....
>
> 1 second recording if you have a 705.
>

Hence, my constant reference to it as "so-called" smart recording ;-)

+1 on the 1s only!

> And yes, we are working on dealing with all the 705 issues.
>

As we knew you would :-) Thanks!
--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[Jens]

The problem with the pauses in WKO+ graph don't change wether you use
auto pause or not. It doesn't depend on the smart recording feature
either, which by the way you shouldn't use when recording power, as
Hunter already mentioned.

The Edge records the absolute time (not trip time) to every data
point. Every data point is then displayed exactly when it happened. So
you can have only 2 datapoints and a WKO+ graph with 15 minutes pause
if datapoint one was recorded at 10.00 and datapoint two at 10.15 for
example.

IIt's really only an issue how the software deals with the recording
format. And this shouldn't be a big problem to solve, I think.

Jens

--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[Tom_Anhalt]

On Dec 15, 3:33 pm, Jens <jens-***@gmx.de> wrote:
> The problem with the pauses in WKO+ graph don't change wether you use
> auto pause or not. It doesn't depend on the smart recording feature
> either,  which by the way you shouldn't use when recording power, as
> Hunter already mentioned.
>
> The Edge records the absolute time (not trip time) to every data
> point. Every data point is then displayed exactly when it happened. So
> you can have only 2 datapoints and a WKO+ graph with 15 minutes pause
> if datapoint one was recorded at 10.00 and datapoint two at 10.15 for
> example.
>
> IIt's really only an issue how the software deals with the recording
> format. And this shouldn't be a big problem to solve, I think.
>
> Jens

OK...and one would think that the "auto pause/unpause" feature of the
Garmin inserts some sort of tag in the data stream, right? So isn't
just a matter of WKO+ being set up to recognize those tags?

Oh well...I'm sure Hunter and the gents know what's what and are
working hard :-)
--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[racerfern]

The Garmin 705 does not record pauses so there are no tags. It doesn't
even record speed. The 705 knows your location, the distance you
traveled and it knows your elevation. The software reading the file
calculates speed by determining the distance from one lat/lon point to
the next and the time it took to get from one point to the next.

If you have a series of points within a given area then you are
paused. However WKO doesn't deal with that correctly. If you switch
from distance to time on the "X" scale then the graphs appear
correctly in WKO. In SportTracks for instance you can declare at what
speed you are paused. I use below 2.5mph.

Here is one "trackpoint" from a data file:

<Trackpoint>
<Time>2008-12-14T01:33:44Z</Time>
<Position>
<LatitudeDegrees>33.663987</LatitudeDegrees>
<LongitudeDegrees>-117.794439</LongitudeDegrees>
</Position>
<AltitudeMeters>33.241</AltitudeMeters>
<DistanceMeters>10175.854</DistanceMeters>
<HeartRateBpm>
<Value>140</Value>
</HeartRateBpm>
<Cadence>82</Cadence>
<SensorState>Present</SensorState>
<Extensions>
<TPX xmlns="http://www.garmin.com/xmlschemas/
ActivityExtension/v2">
<Watts>254</Watts>
</TPX>
</Extensions>
</Trackpoint>

On Dec 15, 3:45 pm, Tom_Anhalt <***@yahoo.com> wrote:
> On Dec 15, 3:33 pm, Jens <jens-***@gmx.de> wrote:
>
> > The problem with the pauses in WKO+ graph don't change wether you use
> > auto pause or not. It doesn't depend on the smart recording feature
> > either,  which by the way you shouldn't use when recording power, as
> > Hunter already mentioned.
>
> > The Edge records the absolute time (not trip time) to every data
> > point. Every data point is then displayed exactly when it happened. So
> > you can have only 2 datapoints and a WKO+ graph with 15 minutes pause
> > if datapoint one was recorded at 10.00 and datapoint two at 10.15 for
> > example.
>
> > IIt's really only an issue how the software deals with the recording
> > format. And this shouldn't be a big problem to solve, I think.
>
> > Jens
>
> OK...and one would think that the "auto pause/unpause" feature of the
> Garmin inserts some sort of tag in the data stream, right?  So isn't
> just a matter of WKO+ being set up to recognize those tags?
>
> Oh well...I'm sure Hunter and the gents know what's what and are
> working hard :-)
--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[Tom_Anhalt]

On Dec 15, 5:20 pm, racerfern <***@gmail.com> wrote:
> The Garmin 705 does not record pauses so there are no tags. It doesn't
> even record speed. The 705 knows your location, the distance you
> traveled and it knows your elevation. The software reading the file
> calculates speed by determining the distance from one lat/lon point to
> the next and the time it took to get from one point to the next.
>
> If you have a series of points within a given area then you are
> paused. However WKO doesn't deal with that correctly. If you switch
> from distance to time on the "X" scale then the graphs appear
> correctly in WKO. In SportTracks for instance you can declare at what
> speed you are paused. I use below 2.5mph.
>
> Here is one "trackpoint" from a data file:
>
> <Trackpoint>
>             <Time>2008-12-14T01:33:44Z</Time>
>             <Position>
>               <LatitudeDegrees>33.663987</LatitudeDegrees>
>               <LongitudeDegrees>-117.794439</LongitudeDegrees>
>             </Position>
>             <AltitudeMeters>33.241</AltitudeMeters>
>             <DistanceMeters>10175.854</DistanceMeters>
>             <HeartRateBpm>
>               <Value>140</Value>
>             </HeartRateBpm>
>             <Cadence>82</Cadence>
>             <SensorState>Present</SensorState>
>             <Extensions>
>               <TPX xmlns="http://www.garmin.com/xmlschemas/
> ActivityExtension/v2">
>                 <Watts>254</Watts>
>               </TPX>
>             </Extensions>
>           </Trackpoint>
>

Hmmm...so what the heck is it doing when I stop and start and it beeps
and displays "auto pause" and "auto resume"...just making a bunch of
noise for no reason? What would be the point of even having that
option (i.e. "auto pause") in the head unit? That sounds a bit
silly...
--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[racerfern]

Not really noise for no reason. The unit FW knows to stop counting
time towards the totals and to stop subtracting results from the
averages, so your "moving" totals are correct. It's when you download
that the trouble starts. Garmin's Training Center handles pauses and
stops correctly. But what good is it since it can't even read the
power data and can't be tailored to view data the way you might want?

Garmin wants to migrate to Garmin Connect which should be able to
handle power data, but that entire project is a total disaster while
trying to roll Motionbased data into the new site. That's why I'm
staying with SportTracks. It has a quirky interface, but very
customizable.

On Dec 15, 5:47 pm, Tom_Anhalt <***@yahoo.com> wrote:
> On Dec 15, 5:20 pm, racerfern <***@gmail.com> wrote:
>
>
>
> > The Garmin 705 does not record pauses so there are no tags. It doesn't
> > even record speed. The 705 knows your location, the distance you
> > traveled and it knows your elevation. The software reading the file
> > calculates speed by determining the distance from one lat/lon point to
> > the next and the time it took to get from one point to the next.
>
> > If you have a series of points within a given area then you are
> > paused. However WKO doesn't deal with that correctly. If you switch
> > from distance to time on the "X" scale then the graphs appear
> > correctly in WKO. In SportTracks for instance you can declare at what
> > speed you are paused. I use below 2.5mph.
>
> > Here is one "trackpoint" from a data file:
>
> > <Trackpoint>
> >             <Time>2008-12-14T01:33:44Z</Time>
> >             <Position>
> >               <LatitudeDegrees>33.663987</LatitudeDegrees>
> >               <LongitudeDegrees>-117.794439</LongitudeDegrees>
> >             </Position>
> >             <AltitudeMeters>33.241</AltitudeMeters>
> >             <DistanceMeters>10175.854</DistanceMeters>
> >             <HeartRateBpm>
> >               <Value>140</Value>
> >             </HeartRateBpm>
> >             <Cadence>82</Cadence>
> >             <SensorState>Present</SensorState>
> >             <Extensions>
> >               <TPX xmlns="http://www.garmin.com/xmlschemas/
> > ActivityExtension/v2">
> >                 <Watts>254</Watts>
> >               </TPX>
> >             </Extensions>
> >           </Trackpoint>
>
> Hmmm...so what the heck is it doing when I stop and start and it beeps
> and displays "auto pause" and "auto resume"...just making a bunch of
> noise for no reason?  What would be the point of even having that
> option (i.e. "auto pause") in the head unit?  That sounds a bit
> silly...
--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[Howard]

Thanks for all the discussion. What I plan to do while WKO+ is
working on an update is to ride with both the powertap head unit and
the edge. I'll download the edge to sportstrack and the powertap cpu
to WKO+.

Howard

--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[hmspain]

Once you update your PT hub to ANT+, assuming you have a Garmin 705,
you will have a couple choices to make.

1) Pull the speed from the Garmin (like you have always done), or pull
the speed from the PT hub (along with the power).

2) Pull the cadence from the Garmin cadence device (like you have
always done), or pull the cadence from the hub.

Based on the previous discussion, I think the decision is clear, pull
the speed/cadence (and power of course) from the hub. Let the Garmin
do what it does best, that is record the track. No need for another
battery powered device on the bike (the Garmin cadence device), just
pull this data from the hub.

BTW, when I was still running the non ANT+ hub, I would import the 705
data into SportTracks and then overlay the PT data :-).

On Dec 16, 7:24 am, Howard <***@atlanticbb.net> wrote:
> Thanks for all the discussion.  What I plan to do while WKO+ is
> working on an update is to ride with both the powertap head unit and
> the edge.  I'll download the edge to sportstrack and the powertap cpu
> to WKO+.
>
> Howard
--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[racerfern]

I still do that and will continue to do so until they fix the NZAP
issue.

On Dec 16, 8:01 am, hmspain <***@gmail.com> wrote:
> BTW, when I was still running the non ANT+ hub, I would import the 705
> data into SportTracks and then overlay the PT data :-).

--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[Howard]

How do you choose where to pull the cadence and speed from? I have a
cycleops cadence sensor on the bike as well as the Garmin speed/
cadence sensor. I would love to get rid of some extra hardware.
What do you suggest? Thanks

On Dec 16, 11:01 am, hmspain <***@gmail.com> wrote:
> Once you update your PT hub to ANT+, assuming you have a Garmin 705,
> you will have a couple choices to make.
>
> 1) Pull the speed from the Garmin (like you have always done), or pull
> the speed from the PT hub (along with the power).
>
> 2) Pull the cadence from the Garmin cadence device (like you have
> always done), or pull the cadence from the hub.
>
> Based on the previous discussion, I think the decision is clear, pull
> the speed/cadence (and power of course) from the hub.  Let the Garmin
> do what it does best, that is record the track.  No need for another
> battery powered device on the bike (the Garmin cadence device), just
> pull this data from the hub.
>
> BTW, when I was still running the non ANT+ hub, I would import the 705
> data into SportTracks and then overlay the PT data :-).
>
> On Dec 16, 7:24 am, Howard <***@atlanticbb.net> wrote:
>
>
>
> > Thanks for all the discussion.  What I plan to do while WKO+ is
> > working on an update is to ride with both the powertap head unit and
> > the edge.  I'll download the edge to sportstrack and the powertap cpu
> > to WKO+.
>
> > Howard- Hide quoted text -
>
> - Show quoted text -
--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[hmspain]

I just checked my Garmin 705, and it allows you to "see" the heart/
power/cadence (heart from the heart strap, and power/cadence from the
hub). The question is, how do you instruct the Garmin 705 to pull the
speed from the hub rather than depend on this calculation on the
Garmin 705?

On Dec 16, 8:01 am, hmspain <***@gmail.com> wrote:
> Once you update your PT hub to ANT+, assuming you have a Garmin 705,
> you will have a couple choices to make.
>
> 1) Pull the speed from the Garmin (like you have always done), or pull
> the speed from the PT hub (along with the power).
>
> 2) Pull the cadence from the Garmin cadence device (like you have
> always done), or pull the cadence from the hub.
>
> Based on the previous discussion, I think the decision is clear, pull
> the speed/cadence (and power of course) from the hub.  Let the Garmin
> do what it does best, that is record the track.  No need for another
> battery powered device on the bike (the Garmin cadence device), just
> pull this data from the hub.
>
> BTW, when I was still running the non ANT+ hub, I would import the 705
> data into SportTracks and then overlay the PT data :-).
>
> On Dec 16, 7:24 am, Howard <***@atlanticbb.net> wrote:
>
> > Thanks for all the discussion.  What I plan to do while WKO+ is
> > working on an update is to ride with both the powertap head unit and
> > the edge.  I'll download the edge to sportstrack and the powertap cpu
> > to WKO+.
>
> > Howard
--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[Trimix]

Yesterday I ran the PT and the 705 while riding a computrainer. I
turned off the sat function on the 705. I rode a course on the
computrainer and I notice the speed a on the 705 flashed in and out.
The miles ridden was also way off. In 20 mile at least 9 miles. If the
garmin can pull from the hub why is it so off on the 705.

ren

On Dec 16, 12:08 pm, hmspain <***@gmail.com> wrote:
> I just checked my Garmin 705, and it allows you to "see" the heart/
> power/cadence (heart from the heart strap, and power/cadence from the
> hub).  The question is, how do you instruct the Garmin 705 to pull the
> speed from the hub rather than depend on this calculation on the
> Garmin 705?
>
> On Dec 16, 8:01 am, hmspain <***@gmail.com> wrote:
>
> > Once you update your PT hub to ANT+, assuming you have a Garmin 705,
> > you will have a couple choices to make.
>
> > 1) Pull the speed from the Garmin (like you have always done), or pull
> > the speed from the PT hub (along with the power).
>
> > 2) Pull the cadence from the Garmin cadence device (like you have
> > always done), or pull the cadence from the hub.
>
> > Based on the previous discussion, I think the decision is clear, pull
> > the speed/cadence (and power of course) from the hub.  Let the Garmin
> > do what it does best, that is record the track.  No need for another
> > battery powered device on the bike (the Garmin cadence device), just
> > pull this data from the hub.
>
> > BTW, when I was still running the non ANT+ hub, I would import the 705
> > data into SportTracks and then overlay the PT data :-).
>
> > On Dec 16, 7:24 am, Howard <***@atlanticbb.net> wrote:
>
> > > Thanks for all the discussion.  What I plan to do while WKO+ is
> > > working on an update is to ride with both the powertap head unit and
> > > the edge.  I'll download the edge to sportstrack and the powertap cpu
> > > to WKO+.
>
> > > Howard
--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[Tom_Anhalt]

On Dec 17, 7:46 am, Nolan <***@garmin.com> wrote:
> There is a problem in the current Edge release related to computing
> the distance and speed data from the new Powertap hub.  We know about
> the issue and have a correction.  It should be released soon as we
> continue to verify the release.  It is true that there was a small
> filter on all the power data.  During early testing we would see
> obvious spikes in power.  This made the Edge appear unreliable even
> though the data was coming from the meters.  So a small filter was
> added to reduce obvious invalid wattage values that were coming from
> the meters.  We are evaluating if this filter needs to be present for
> all power meters.
>

Thanks for the input Nolan! I'd like to add to the request that NO
filtering be done to the data coming from the ANT+ compatible power
meters.

However, could you guys PLEASE put a filter on the GPS-based speed, or
somehow otherwise fix the "speed spikes up/down" recorded when using
1s sampling? Now THOSE are some "obviously invalid" speed values ;-)
--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[Stuart Lynne]

On Wed, Dec 17, 2008 at 1:08 PM, Tom_Anhalt <***@yahoo.com> wrote:

>
>
>
> On Dec 17, 7:46 am, Nolan <***@garmin.com> wrote:
> > There is a problem in the current Edge release related to computing
> > the distance and speed data from the new Powertap hub. We know about
> > the issue and have a correction. It should be released soon as we
> > continue to verify the release. It is true that there was a small
> > filter on all the power data. During early testing we would see
> > obvious spikes in power. This made the Edge appear unreliable even
> > though the data was coming from the meters. So a small filter was
> > added to reduce obvious invalid wattage values that were coming from
> > the meters. We are evaluating if this filter needs to be present for
> > all power meters.
> >
>
> Thanks for the input Nolan! I'd like to add to the request that NO
> filtering be done to the data coming from the ANT+ compatible power
> meters.
>
> However, could you guys PLEASE put a filter on the GPS-based speed, or
> somehow otherwise fix the "speed spikes up/down" recorded when using
> 1s sampling? Now THOSE are some "obviously invalid" speed values ;-)
>

Other than hardware or electronics design issues... is there any reason that
a time based sample is used?

Wouldn't, all things being equal, samples based on (for example) pedal
revolution? Maybe one sample per revolution? Or even two?

I realize that this means that you also have to send the reference time
along with the measured value, but I suspect that even at high cadence the
Ant+ network could handle it. Say a max RPM of 240, or 4 x 2 samples per
second, at say about 32bytes per message...

Maybe supplement so that at low cadence you do more samples etc...


--
__________O___________
_______-\<,____________
_____(_)/_(_)___________
_________________________
***@enposte.net

--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[Dave Harris]

The ANT+ receivers only report what they are sent. So in this case that's a
PT request, not Garmin.

From: ***@googlegroups.com [mailto:***@googlegroups.com] On Behalf
Of Stuart Lynne
Sent: Wednesday, December 17, 2008 2:18 PM
To: ***@googlegroups.com
Subject: Fwd: [Wattage] Re: Edge 705 and ant+ Powertap


Other than hardware or electronics design issues... is there any reason that
a time based sample is used? 

Wouldn't, all things being equal, samples based on (for example) pedal
revolution? Maybe one sample per revolution? Or even two?

I realize that this means that you also have to send the reference time
along with the measured value, but I suspect that even at high cadence the
Ant+ network could handle it. Say a max RPM of 240, or 4 x 2 samples per
second, at say about 32bytes per message... 

Maybe supplement so that at low cadence you do more samples etc... 




--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[Tom_Anhalt]

On Dec 18, 6:30 am, Nolan <***@garmin.com> wrote:
> How much variability is really "real" in power data.  Does a users
> pedaling efficiency cause a large fluctuation in the data?  If I ride
> on a trainer at a constant rpm how much variance should there be in
> the data?   I am new to power and have not analyzed the myriad of
> files and data that all the people on this list do.   From a user
> standpoint it looks like the meter is lying to me if I can pedal on a
> trainer at a constant rpm and get 20 watt differences in the data each
> second.   So is this why a display filter is used to placate the user
> that the data is more constant but still record all the differences in
> the actual saved data?   If all filtering is removed from the saved
> data does an Edge need to  have a display filter to compensate for the
> user experience?
>

As others have said so eloquently (Hi Dave Harris!) the variability in
the power data is REAL, especially coming from the PT due to the fixed
time base calculation, as opposed to the calculation being performed
on an integer pedal cycle basis (i.e. SRM, CinQo, or even Polar
<sorta>).

So...in short...filtering of the power data prior to recording?
Bad. :-(

User selectable "smoothing" (i.e. simple moving average) of the
DISPLAYED power (e.g. no smoothing, 2s, 5s, 10s, 30s, etc. averages)?
Good :-)

--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[ggalat]

Ren

Is your distance only off when riding on the trainer?

I am doing the same thing, and seeing almost the exact same data,
except when riding idoors, when the satellites don't see me, and the
device reverts to getting distance/speed data from the PT hub.

Geoff

On Dec 18, 12:48 pm, Trimix <***@robinhill.com> wrote:
> I am also riding with powertap cpu and 705 side by side. The distance
> rode is way off which I mention before and I see they are working on
> it. I have also notice the powertap cpu reacts faster than the
> 705.Basically the power number are pretty close between the two. If
> anyone working on this issue wants data showing the two units working
> at the same time they can email me directly.
>
> Ren
>
> On Dec 17, 4:08 pm, Tom_Anhalt <***@yahoo.com> wrote:
>
>
>
> > On Dec 17, 7:46 am, Nolan <***@garmin.com> wrote:
>
> > > There is a problem in the current Edge release related to computing
> > > the distance and speed data from the new Powertap hub.  We know about
> > > the issue and have a correction.  It should be released soon as we
> > > continue to verify the release.  It is true that there was a small
> > > filter on all the power data.  During early testing we would see
> > > obvious spikes in power.  This made the Edge appear unreliable even
> > > though the data was coming from the meters.  So a small filter was
> > > added to reduce obvious invalid wattage values that were coming from
> > > the meters.  We are evaluating if this filter needs to be present for
> > > all power meters.
>
> > Thanks for the input Nolan!  I'd like to add to the request that NO
> > filtering be done to the data coming from the ANT+ compatible power
> > meters.
>
> > However, could you guys PLEASE put a filter on the GPS-based speed, or
> > somehow otherwise fix the "speed spikes up/down" recorded when using
> > 1s sampling?  Now THOSE are some "obviously invalid" speed values ;-)- Hide quoted text -
>
> - Show quoted text -
--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[Nolan]

There is a problem in the current Edge release related to computing
the distance and speed data from the new Powertap hub. We know about
the issue and have a correction. It should be released soon as we
continue to verify the release. It is true that there was a small
filter on all the power data. During early testing we would see
obvious spikes in power. This made the Edge appear unreliable even
though the data was coming from the meters. So a small filter was
added to reduce obvious invalid wattage values that were coming from
the meters. We are evaluating if this filter needs to be present for
all power meters.




On Dec 16, 3:04 pm, Trimix <***@robinhill.com> wrote:
> Yesterday I ran the PT and the 705 while riding a computrainer. I
> turned off the sat function on the 705. I rode a course on the
> computrainer and I notice the speed a on the 705 flashed in and out.
> The miles ridden was also way off. In 20 mile at least 9 miles. If the
> garmin can pull from the hub why is it so off on the 705.
>
> ren
>
> On Dec 16, 12:08 pm, hmspain <***@gmail.com> wrote:
>
>
>
> > I just checked my Garmin 705, and it allows you to "see" the heart/
> > power/cadence (heart from the heart strap, and power/cadence from the
> > hub).  The question is, how do you instruct the Garmin 705 to pull the
> > speed from the hub rather than depend on this calculation on the
> > Garmin 705?
>
> > On Dec 16, 8:01 am, hmspain <***@gmail.com> wrote:
>
> > > Once you update your PT hub to ANT+, assuming you have a Garmin 705,
> > > you will have a couple choices to make.
>
> > > 1) Pull the speed from the Garmin (like you have always done), or pull
> > > the speed from the PT hub (along with the power).
>
> > > 2) Pull the cadence from the Garmin cadence device (like you have
> > > always done), or pull the cadence from the hub.
>
> > > Based on the previous discussion, I think the decision is clear, pull
> > > the speed/cadence (and power of course) from the hub.  Let the Garmin
> > > do what it does best, that is record the track.  No need for another
> > > battery powered device on the bike (the Garmin cadence device), just
> > > pull this data from the hub.
>
> > > BTW, when I was still running the non ANT+ hub, I would import the 705
> > > data into SportTracks and then overlay the PT data :-).
>
> > > On Dec 16, 7:24 am, Howard <***@atlanticbb.net> wrote:
>
> > > > Thanks for all the discussion.  What I plan to do while WKO+ is
> > > > working on an update is to ride with both the powertap head unit and
> > > > the edge.  I'll download the edge to sportstrack and the powertap cpu
> > > > to WKO+.
>
> > > > Howard- Hide quoted text -
>
> - Show quoted text -
--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[Chris Cleeland]

On Wed, Dec 17, 2008 at 9:46 AM, Nolan <***@garmin.com> wrote:
>
> It is true that there was a small
> filter on all the power data. During early testing we would see
> obvious spikes in power. This made the Edge appear unreliable even
> though the data was coming from the meters. So a small filter was
> added to reduce obvious invalid wattage values that were coming from
> the meters. We are evaluating if this filter needs to be present for
> all power meters.

IMHO, the head unit shouldn't be filtering it at all. Record what's
sent. Let post-ride analysis after download perform filtering. How
does your head unit know that a wattage value is obviously invalid?
Only apply lossless filters that can be reliably inverted in order to
get the original data stream.

--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[DailyEvil]

On Dec 17, 10:57 am, "Chris Cleeland" <***@gmail.com>
wrote:
>
> IMHO, the head unit shouldn't be filtering it at all.

The head unit shouldn't be filtering the data it *records* at all. It
should record exactly what it receives, no more, no less.

But, a user-selectable filter on the DISPLAY is a useful feature.
Having it be user-tunable so it can be adjusted for different workouts
is critical.

--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[Chris Cleeland]

On Wed, Dec 17, 2008 at 10:23 AM, DailyEvil <***@gmail.com> wrote:
>
> On Dec 17, 10:57 am, "Chris Cleeland" <***@gmail.com>
> wrote:
>>
>> IMHO, the head unit shouldn't be filtering it at all.
>
> The head unit shouldn't be filtering the data it *records* at all.
> But, a user-selectable filter on the DISPLAY is a useful feature.

Good point.

> Having it be user-tunable so it can be adjusted for different workouts
> is critical.

You mean, like, being able to turn it off?

--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[Jaime Menendez de Luarca]

Is it possible to adjust Zero Torque in a PT Using the G 705?

2008/12/17 Chris Cleeland <***@gmail.com>

>
> On Wed, Dec 17, 2008 at 10:23 AM, DailyEvil <***@gmail.com> wrote:
> >
> > On Dec 17, 10:57 am, "Chris Cleeland" <***@gmail.com>
> > wrote:
> >>
> >> IMHO, the head unit shouldn't be filtering it at all.
> >
> > The head unit shouldn't be filtering the data it *records* at all.
> > But, a user-selectable filter on the DISPLAY is a useful feature.
>
> Good point.
>
> > Having it be user-tunable so it can be adjusted for different workouts
> > is critical.
>
> You mean, like, being able to turn it off?
>
> >
>

--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[DailyEvil]

On Dec 17, 11:27 am, "Chris Cleeland" <***@gmail.com>
wrote:

> You mean, like, being able to turn it off?

Yes, on/off as a minimum. But, being able to choose the length of time/
number of samples to average over is useful too... a long flat TT has
different requirements than 30 second sprints. The athlete might not
choose to adjust for each effort, but it's nice to have.


--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[Nolan]

How much variability is really "real" in power data. Does a users
pedaling efficiency cause a large fluctuation in the data? If I ride
on a trainer at a constant rpm how much variance should there be in
the data? I am new to power and have not analyzed the myriad of
files and data that all the people on this list do. From a user
standpoint it looks like the meter is lying to me if I can pedal on a
trainer at a constant rpm and get 20 watt differences in the data each
second. So is this why a display filter is used to placate the user
that the data is more constant but still record all the differences in
the actual saved data? If all filtering is removed from the saved
data does an Edge need to have a display filter to compensate for the
user experience?


On Dec 17, 3:08 pm, Tom_Anhalt <***@yahoo.com> wrote:
> On Dec 17, 7:46 am, Nolan <***@garmin.com> wrote:
>
> > There is a problem in the current Edge release related to computing
> > the distance and speed data from the new Powertap hub.  We know about
> > the issue and have a correction.  It should be released soon as we
> > continue to verify the release.  It is true that there was a small
> > filter on all the power data.  During early testing we would see
> > obvious spikes in power.  This made the Edge appear unreliable even
> > though the data was coming from the meters.  So a small filter was
> > added to reduce obvious invalid wattage values that were coming from
> > the meters.  We are evaluating if this filter needs to be present for
> > all power meters.
>
> Thanks for the input Nolan!  I'd like to add to the request that NO
> filtering be done to the data coming from the ANT+ compatible power
> meters.
>
> However, could you guys PLEASE put a filter on the GPS-based speed, or
> somehow otherwise fix the "speed spikes up/down" recorded when using
> 1s sampling?  Now THOSE are some "obviously invalid" speed values ;-)
--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[Greg Steele]

The variability is real.

To use your example, at a constant trainer rpm, there probably won't
even be huge wheelspeed variations despite the 20w variability.

I would be happy to do a quick test tonight of this with torque
analysis (a raw dump from the srm) that will show that what occurs in
the 20w variation, despite constant cadence, is that the "pushes" at
the downstroke are higher. Actually, what really happens is that the
entire torque curve shifts up, but the downstrokes especially.

So storing the data as sent is vital to the edge being (for this group
at least) a viable option of an ant+sport powermeter hea unit.

Filtering the data on the display is another thing entirely. IMHO,
the ability to turn this on/off/control it would be nice. I use my
srm and powertaps with smoothing turned off. I feel if I wanted
smoothed data, I could use HR.

G

On 12/18/08, Nolan <***@garmin.com> wrote:
>
> How much variability is really "real" in power data. Does a users
> pedaling efficiency cause a large fluctuation in the data? If I ride
> on a trainer at a constant rpm how much variance should there be in
> the data? I am new to power and have not analyzed the myriad of
> files and data that all the people on this list do. From a user
> standpoint it looks like the meter is lying to me if I can pedal on a
> trainer at a constant rpm and get 20 watt differences in the data each
> second. So is this why a display filter is used to placate the user
> that the data is more constant but still record all the differences in
> the actual saved data? If all filtering is removed from the saved
> data does an Edge need to have a display filter to compensate for the
> user experience?
>
>
> On Dec 17, 3:08 pm, Tom_Anhalt <***@yahoo.com> wrote:
>> On Dec 17, 7:46 am, Nolan <***@garmin.com> wrote:
>>
>> > There is a problem in the current Edge release related to computing
>> > the distance and speed data from the new Powertap hub. We know about
>> > the issue and have a correction. It should be released soon as we
>> > continue to verify the release. It is true that there was a small
>> > filter on all the power data. During early testing we would see
>> > obvious spikes in power. This made the Edge appear unreliable even
>> > though the data was coming from the meters. So a small filter was
>> > added to reduce obvious invalid wattage values that were coming from
>> > the meters. We are evaluating if this filter needs to be present for
>> > all power meters.
>>
>> Thanks for the input Nolan! I'd like to add to the request that NO
>> filtering be done to the data coming from the ANT+ compatible power
>> meters.
>>
>> However, could you guys PLEASE put a filter on the GPS-based speed, or
>> somehow otherwise fix the "speed spikes up/down" recorded when using
>> 1s sampling? Now THOSE are some "obviously invalid" speed values ;-)
> >
>

--
Sent from my mobile device

--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[Chris Mayhew]

On Thu, Dec 18, 2008 at 9:39 AM, Greg Steele <***@gmail.com> wrote:

> smoothed data, I could use HR.

I thought you wanted data?




--
Chris Mayhew
www.jbvcoaching.com

--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[Greg Steele]

Not to reply to my own post, but I forgot I had this image on my server:

http://wattagetraining.com/100-300wTA.png

This shows torque vs. crank angle for constant 100w, 200w, and 300w
pedalling at constant cadence.

You will notice it acts as I described below.

Thanks,

g

p.s. this is 30s averaged data, I also can show individual crank
revolution variations...if needed.

Greg Steele wrote:
> The variability is real.
>
> To use your example, at a constant trainer rpm, there probably won't
> even be huge wheelspeed variations despite the 20w variability.
>
> I would be happy to do a quick test tonight of this with torque
> analysis (a raw dump from the srm) that will show that what occurs in
> the 20w variation, despite constant cadence, is that the "pushes" at
> the downstroke are higher. Actually, what really happens is that the
> entire torque curve shifts up, but the downstrokes especially.
>
> So storing the data as sent is vital to the edge being (for this group
> at least) a viable option of an ant+sport powermeter hea unit.
>
> Filtering the data on the display is another thing entirely. IMHO,
> the ability to turn this on/off/control it would be nice. I use my
> srm and powertaps with smoothing turned off. I feel if I wanted
> smoothed data, I could use HR.
>
> G
>
> On 12/18/08, Nolan <***@garmin.com> wrote:
>> How much variability is really "real" in power data. Does a users
>> pedaling efficiency cause a large fluctuation in the data? If I ride
>> on a trainer at a constant rpm how much variance should there be in
>> the data? I am new to power and have not analyzed the myriad of
>> files and data that all the people on this list do. From a user
>> standpoint it looks like the meter is lying to me if I can pedal on a
>> trainer at a constant rpm and get 20 watt differences in the data each
>> second. So is this why a display filter is used to placate the user
>> that the data is more constant but still record all the differences in
>> the actual saved data? If all filtering is removed from the saved
>> data does an Edge need to have a display filter to compensate for the
>> user experience?
>>
>>
>> On Dec 17, 3:08 pm, Tom_Anhalt <***@yahoo.com> wrote:
>>> On Dec 17, 7:46 am, Nolan <***@garmin.com> wrote:
>>>
>>>> There is a problem in the current Edge release related to computing
>>>> the distance and speed data from the new Powertap hub. We know about
>>>> the issue and have a correction. It should be released soon as we
>>>> continue to verify the release. It is true that there was a small
>>>> filter on all the power data. During early testing we would see
>>>> obvious spikes in power. This made the Edge appear unreliable even
>>>> though the data was coming from the meters. So a small filter was
>>>> added to reduce obvious invalid wattage values that were coming from
>>>> the meters. We are evaluating if this filter needs to be present for
>>>> all power meters.
>>> Thanks for the input Nolan! I'd like to add to the request that NO
>>> filtering be done to the data coming from the ANT+ compatible power
>>> meters.
>>>
>>> However, could you guys PLEASE put a filter on the GPS-based speed, or
>>> somehow otherwise fix the "speed spikes up/down" recorded when using
>>> 1s sampling? Now THOSE are some "obviously invalid" speed values ;-)
>> >>
>>
>

--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[sergen]

I'm would definitely plead for Garmin to include a 'DISPLAY' filtering
feature - for example, the fluctuations in wattage on the display can
be annoying when doing time trial efforts on a flat course when trying
to hold a certain wattage. I suppose it comes down to personal
preference but I don't want to see power fluctuations of 20w.

But as for the 'RECORDED' data that's a totally different matter. No
filters please, I want to see it warts and all...


On Dec 18, 2:39 pm, "Greg Steele" <***@gmail.com> wrote:
> The variability is real.
>
> To use your example, at a constant trainer rpm, there probably won't
> even be huge wheelspeed variations despite the 20w variability.
>
> I would be happy to do a quick test tonight of this with torque
> analysis (a raw dump from the srm) that will show that what occurs in
> the 20w variation, despite constant cadence, is that the "pushes" at
> the downstroke are higher.  Actually, what really happens is that the
> entire torque curve shifts up, but the downstrokes especially.
>
> So storing the data as sent is vital to the edge being (for this group
> at least) a viable option of an ant+sport powermeter hea unit.
>
> Filtering the data on the display is another thing entirely.  IMHO,
> the ability to turn this on/off/control it would be nice.  I use my
> srm and powertaps with smoothing turned off.  I feel if I wanted
> smoothed data, I could use HR.
>
> G
>
> On 12/18/08, Nolan <***@garmin.com> wrote:
>
>
>
>
>
> > How much variability is really "real" in power data.  Does a users
> > pedaling efficiency cause a large fluctuation in the data?  If I ride
> > on a trainer at a constant rpm how much variance should there be in
> > the data?   I am new to power and have not analyzed the myriad of
> > files and data that all the people on this list do.   From a user
> > standpoint it looks like the meter is lying to me if I can pedal on a
> > trainer at a constant rpm and get 20 watt differences in the data each
> > second.   So is this why a display filter is used to placate the user
> > that the data is more constant but still record all the differences in
> > the actual saved data?   If all filtering is removed from the saved
> > data does an Edge need to  have a display filter to compensate for the
> > user experience?
>
> > On Dec 17, 3:08 pm, Tom_Anhalt <***@yahoo.com> wrote:
> >> On Dec 17, 7:46 am, Nolan <***@garmin.com> wrote:
>
> >> > There is a problem in the current Edge release related to computing
> >> > the distance and speed data from the new Powertap hub.  We know about
> >> > the issue and have a correction.  It should be released soon as we
> >> > continue to verify the release.  It is true that there was a small
> >> > filter on all the power data.  During early testing we would see
> >> > obvious spikes in power.  This made the Edge appear unreliable even
> >> > though the data was coming from the meters.  So a small filter was
> >> > added to reduce obvious invalid wattage values that were coming from
> >> > the meters.  We are evaluating if this filter needs to be present for
> >> > all power meters.
>
> >> Thanks for the input Nolan!  I'd like to add to the request that NO
> >> filtering be done to the data coming from the ANT+ compatible power
> >> meters.
>
> >> However, could you guys PLEASE put a filter on the GPS-based speed, or
> >> somehow otherwise fix the "speed spikes up/down" recorded when using
> >> 1s sampling?  Now THOSE are some "obviously invalid" speed values ;-)
>
> --
> Sent from my mobile device
--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[Dave Harris]

Nolan,

Thanks for joining in the fun here.

Power is highly variable. You'll find that anyone seriously making use of
power data will want the data set exactly as it came from the measuring
meter, warts and all. We've got all sorts of tools and experience to deal
with its variable nature.

I found that the smoothing applied to PT data received at the 705 actually
alters the results of this post processing - some of the data is lost.

Are you familiar with precession in the PT? The bit counters will correct
my numbers <g>, but basically the PT samples hub torque at some frequency, I
think it's 60hz, calculates an associated power value and stores it. Every
1.26 (now 1.0) seconds it reports the average power over those previous 1.26
(or 1 s) values. So, at a cadence of 75 (or 60) each data point is a
measure of a full pedal revolution.

In reality, instantaneous power production is more of a sin wave with peaks
near crank horizontal and valleys near crank vertical. So, it's possible
(likely, actually) in the PT to get a data stream with numbers fluctuating.
Unless your cadence is right at the sweet spot of 75 (or 60) each data point
was generated from some fractional integral of pedal strokes - and those
values will "seesaw". Over the longer term, say 5 sec - they average out
just fine.

The current 705 smoothing reduces kJ/ride by (on average) 2.25% in my files.
It is much, much more than this for brief periods of highly variable data.

It is essential for the accuracy of power tap data that these fluctuations
be preserved in recorded data. Other power meters that integrate torque
over pedal revolutions may not be affected by the current smoothing
algorithm (hi Tom). Hence, since PT just released its ANT+ support this is
the first time the issue has come public.


> -----Original Message-----
> From: ***@googlegroups.com [mailto:***@googlegroups.com] On
> Behalf Of Nolan
> Sent: Thursday, December 18, 2008 7:31 AM
> To: Wattage
> Subject: [Wattage] Re: Edge 705 and ant+ Powertap
>
>
> How much variability is really "real" in power data. Does a users
> pedaling efficiency cause a large fluctuation in the data? If I ride
> on a trainer at a constant rpm how much variance should there be in
> the data? I am new to power and have not analyzed the myriad of
> files and data that all the people on this list do. From a user
> standpoint it looks like the meter is lying to me if I can pedal on a
> trainer at a constant rpm and get 20 watt differences in the data each
> second. So is this why a display filter is used to placate the user
> that the data is more constant but still record all the differences in
> the actual saved data? If all filtering is removed from the saved
> data does an Edge need to have a display filter to compensate for the
> user experience?
>
>
> On Dec 17, 3:08 pm, Tom_Anhalt <***@yahoo.com> wrote:
> > On Dec 17, 7:46 am, Nolan <***@garmin.com> wrote:
> >
> > > There is a problem in the current Edge release related to computing
> > > the distance and speed data from the new Powertap hub.  We know
> about
> > > the issue and have a correction.  It should be released soon as we
> > > continue to verify the release.  It is true that there was a small
> > > filter on all the power data.  During early testing we would see
> > > obvious spikes in power.  This made the Edge appear unreliable even
> > > though the data was coming from the meters.  So a small filter was
> > > added to reduce obvious invalid wattage values that were coming
> from
> > > the meters.  We are evaluating if this filter needs to be present
> for
> > > all power meters.
> >
> > Thanks for the input Nolan!  I'd like to add to the request that NO
> > filtering be done to the data coming from the ANT+ compatible power
> > meters.
> >
> > However, could you guys PLEASE put a filter on the GPS-based speed,
> or
> > somehow otherwise fix the "speed spikes up/down" recorded when using
> > 1s sampling?  Now THOSE are some "obviously invalid" speed values ;-)
>

--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[Tom_Anhalt]

Nolan...(or is it David? Just curious based on the email address),

One quick question. Is the filtering of the saved power data applied
to power data coming from ANY of the ANT+ compatible power meters, or
was it only applied to certain models? Based on the reports of the
Garmin being able to determine what power meter it's talking to (as
evidenced by different setup screens available for different models) I
was wondering if it was handling the data streams for various power
meters differently.

The only reason I ask is that (as can be seen in this thread and the
posts between Dave Harris and I) when comparing a PT output to a CinQo/
Garmin output, I'm not seeing the same sort of differences that Dave
is seeing comparing PT vs. PT/Garmin recordings.

No matter what though...as we've emphasized already, I'd highly
encourage you to change the firmware so that NO filtering is done on
the incoming data for ANY power meter. In fact, for certain power
meters that actually transmit data at faster rates than 1s (such as
the CinQo with it's 4hz data rate) I'd encourage you to either record
ALL the data, or at least save at each 1s point an average of the 4
data points coming in each second.

Thanks for listening!

On Dec 18, 6:30 am, Nolan <***@garmin.com> wrote:
> How much variability is really "real" in power data.  Does a users
> pedaling efficiency cause a large fluctuation in the data?  If I ride
> on a trainer at a constant rpm how much variance should there be in
> the data?   I am new to power and have not analyzed the myriad of
> files and data that all the people on this list do.   From a user
> standpoint it looks like the meter is lying to me if I can pedal on a
> trainer at a constant rpm and get 20 watt differences in the data each
> second.   So is this why a display filter is used to placate the user
> that the data is more constant but still record all the differences in
> the actual saved data?   If all filtering is removed from the saved
> data does an Edge need to  have a display filter to compensate for the
> user experience?
>
> On Dec 17, 3:08 pm, Tom_Anhalt <***@yahoo.com> wrote:
>
> > On Dec 17, 7:46 am, Nolan <***@garmin.com> wrote:
>
> > > There is a problem in the current Edge release related to computing
> > > the distance and speed data from the new Powertap hub.  We know about
> > > the issue and have a correction.  It should be released soon as we
> > > continue to verify the release.  It is true that there was a small
> > > filter on all the power data.  During early testing we would see
> > > obvious spikes in power.  This made the Edge appear unreliable even
> > > though the data was coming from the meters.  So a small filter was
> > > added to reduce obvious invalid wattage values that were coming from
> > > the meters.  We are evaluating if this filter needs to be present for
> > > all power meters.
>
> > Thanks for the input Nolan!  I'd like to add to the request that NO
> > filtering be done to the data coming from the ANT+ compatible power
> > meters.
>
> > However, could you guys PLEASE put a filter on the GPS-based speed, or
> > somehow otherwise fix the "speed spikes up/down" recorded when using
> > 1s sampling?  Now THOSE are some "obviously invalid" speed values ;-)
--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[Nolan]

The filtering is currently done on all power meters. Any differences
would be based on the nature of the data. It is not a time filter but
a sample filter. As such the time based systems would be more
affected by the filter than an event system (based on cadence).


On Dec 18, 9:56 am, Tom_Anhalt <***@yahoo.com> wrote:
> Nolan...(or is it David?  Just curious based on the email address),
>
> One quick question.  Is the filtering of the saved power data applied
> to power data coming from ANY of the ANT+ compatible power meters, or
> was it only applied to certain models?  Based on the reports of the
> Garmin being able to determine what power meter it's talking to (as
> evidenced by different setup screens available for different models) I
> was wondering if it was handling the data streams for various power
> meters differently.
>
> The only reason I ask is that (as can be seen in this thread and the
> posts between Dave Harris and I) when comparing a PT output to a CinQo/
> Garmin output, I'm not seeing the same sort of differences that Dave
> is seeing comparing PT vs. PT/Garmin recordings.
>
> No matter what though...as we've emphasized already, I'd highly
> encourage you to change the firmware so that NO filtering is done on
> the incoming data for ANY power meter.  In fact, for certain power
> meters that actually transmit data at faster rates than 1s (such as
> the CinQo with it's 4hz data rate) I'd encourage you to either record
> ALL the data, or at least save at each 1s point an average of the 4
> data points coming in each second.
>
> Thanks for listening!
>
> On Dec 18, 6:30 am, Nolan <***@garmin.com> wrote:
>
>
>
> > How much variability is really "real" in power data.  Does a users
> > pedaling efficiency cause a large fluctuation in the data?  If I ride
> > on a trainer at a constant rpm how much variance should there be in
> > the data?   I am new to power and have not analyzed the myriad of
> > files and data that all the people on this list do.   From a user
> > standpoint it looks like the meter is lying to me if I can pedal on a
> > trainer at a constant rpm and get 20 watt differences in the data each
> > second.   So is this why a display filter is used to placate the user
> > that the data is more constant but still record all the differences in
> > the actual saved data?   If all filtering is removed from the saved
> > data does an Edge need to  have a display filter to compensate for the
> > user experience?
>
> > On Dec 17, 3:08 pm, Tom_Anhalt <***@yahoo.com> wrote:
>
> > > On Dec 17, 7:46 am, Nolan <***@garmin.com> wrote:
>
> > > > There is a problem in the current Edge release related to computing
> > > > the distance and speed data from the new Powertap hub.  We know about
> > > > the issue and have a correction.  It should be released soon as we
> > > > continue to verify the release.  It is true that there was a small
> > > > filter on all the power data.  During early testing we would see
> > > > obvious spikes in power.  This made the Edge appear unreliable even
> > > > though the data was coming from the meters.  So a small filter was
> > > > added to reduce obvious invalid wattage values that were coming from
> > > > the meters.  We are evaluating if this filter needs to be present for
> > > > all power meters.
>
> > > Thanks for the input Nolan!  I'd like to add to the request that NO
> > > filtering be done to the data coming from the ANT+ compatible power
> > > meters.
>
> > > However, could you guys PLEASE put a filter on the GPS-based speed, or
> > > somehow otherwise fix the "speed spikes up/down" recorded when using
> > > 1s sampling?  Now THOSE are some "obviously invalid" speed values ;-)- Hide quoted text -
>
> - Show quoted text -
--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[Tom_Anhalt]

On Dec 19, 5:32 am, Nolan <***@garmin.com> wrote:
> The filtering is currently done on all power meters.  Any differences
> would be based on the nature of the data.  It is not a time filter but
> a sample filter.  As such the time based systems would be more
> affected by the filter than an event system (based on cadence).
>

Thanks! But, I think I'm being a little "extra-dense" this
today...can you describe the difference between those two types of
filters a bit more? I'm just a mechanical engineer and I'm not up on
all the signal processing stuff...
--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[MC Double E]

It sounds like he is saying that it's not a filter in the sense that
it has a defined impulse response that is applied to all data. It's a
sample by sample decision whether to accept or reject?

On Dec 19, 4:59 pm, Tom_Anhalt <***@yahoo.com> wrote:
> On Dec 19, 5:32 am, Nolan <***@garmin.com> wrote:
>
> > The filtering is currently done on all power meters.  Any differences
> > would be based on the nature of the data.  It is not a time filter but
> > a sample filter.  As such the time based systems would be more
> > affected by the filter than an event system (based on cadence).
>
> Thanks!  But, I think I'm being a little "extra-dense" this
> today...can you describe the difference between those two types of
> filters a bit more?  I'm just a mechanical engineer and I'm not up on
> all the signal processing stuff...
--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[Zach Rogers]

Not the original poster, but I think I understand.

The time based systems, e.g. powertap, can very well have single sample
spikes in the data by their very nature.

The 1 pedal cycle based systems would tend to have less/no single sample
spikes.

This difference is simply based on the fact that the Powertap could
conceivably record a record that is mainly comprised of the power phase of
your pedal stroke, followed by a record that is mostly the top or bottom of
your pedal stroke. Even when you are pedaling nice "circles" you will push
down on the pedal a lot better than you can pull fore/aft. A system that
records one record with each cycle of the pedals is already averaging out
the pedal stroke portion of the datastream and thus won't give you the power
stroke, TDC, power stroke, TDC, ... type datastream.

So if you were to record the raw data stream from both types of systems and
the Garmin filtered data stream from both types systems, the effect of the
filtering would be more obvious on the PT data.

Now, obviously more is going on here then simply smoothing out 1sample
spikes, but it makes a good example of how applying the same filter to both
types of data can have different results.

-Zach

On Fri, Dec 19, 2008 at 4:59 PM, Tom_Anhalt <***@yahoo.com> wrote:

>
>
>
> On Dec 19, 5:32 am, Nolan <***@garmin.com> wrote:
> > The filtering is currently done on all power meters. Any differences
> > would be based on the nature of the data. It is not a time filter but
> > a sample filter. As such the time based systems would be more
> > affected by the filter than an event system (based on cadence).
> >
>
> Thanks! But, I think I'm being a little "extra-dense" this
> today...can you describe the difference between those two types of
> filters a bit more? I'm just a mechanical engineer and I'm not up on
> all the signal processing stuff...
> >
>

--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[jason osborne]

Really it's probably just a variant on a low-pass filter allowing the system
to attenuate signals that are "unrealistic" based on the sample set of
maximum power values recorded during the ride, plus some tolerance.

And yes, for "rev" based systems the range on the spikes will be smaller
than the "time" based systems.

Either way - filter bad, data good.

Let me filter later if I want, the PM does enough time based averaging in
simply collecting the data for this trackie anyway. :)


Jason.


On Fri, Dec 19, 2008 at 2:35 PM, Zach Rogers <***@gmail.com> wrote:

> Not the original poster, but I think I understand.
>
> The time based systems, e.g. powertap, can very well have single sample
> spikes in the data by their very nature.
>
> The 1 pedal cycle based systems would tend to have less/no single sample
> spikes.
>
> This difference is simply based on the fact that the Powertap could
> conceivably record a record that is mainly comprised of the power phase of
> your pedal stroke, followed by a record that is mostly the top or bottom of
> your pedal stroke. Even when you are pedaling nice "circles" you will push
> down on the pedal a lot better than you can pull fore/aft. A system that
> records one record with each cycle of the pedals is already averaging out
> the pedal stroke portion of the datastream and thus won't give you the power
> stroke, TDC, power stroke, TDC, ... type datastream.
>
> So if you were to record the raw data stream from both types of systems and
> the Garmin filtered data stream from both types systems, the effect of the
> filtering would be more obvious on the PT data.
>
> Now, obviously more is going on here then simply smoothing out 1sample
> spikes, but it makes a good example of how applying the same filter to both
> types of data can have different results.
>
> -Zach
>
>
> On Fri, Dec 19, 2008 at 4:59 PM, Tom_Anhalt <***@yahoo.com> wrote:
>
>>
>>
>>
>> On Dec 19, 5:32 am, Nolan <***@garmin.com> wrote:
>> > The filtering is currently done on all power meters. Any differences
>> > would be based on the nature of the data. It is not a time filter but
>> > a sample filter. As such the time based systems would be more
>> > affected by the filter than an event system (based on cadence).
>> >
>>
>> Thanks! But, I think I'm being a little "extra-dense" this
>> today...can you describe the difference between those two types of
>> filters a bit more? I'm just a mechanical engineer and I'm not up on
>> all the signal processing stuff...
>>
>>
>
> >
>

--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[Greg Steele]

you should have never gotten ride of the PCIV and it's 0.1s recording :P

g

jason osborne wrote:

> Let me filter later if I want, the PM does enough time based averaging
> in simply collecting the data for this trackie anyway. :)
>
>
> Jason.


--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[jason osborne]

Never had one.

Besides, I doubt I'd be happy anyway as, I'd still have to rely on an
assumed "constant" inter-rev pedal velocity.

Jason.

On Fri, Dec 19, 2008 at 3:06 PM, Greg Steele <***@wattagetraining.com>wrote:

>
> you should have never gotten ride of the PCIV and it's 0.1s recording :P
>
> g
>
> jason osborne wrote:
>
> > Let me filter later if I want, the PM does enough time based averaging
> > in simply collecting the data for this trackie anyway. :)
> >
> >
> > Jason.
>
>
> >
>

--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[AndyF]

On Dec 18, 10:41 am, Tom_Anhalt <***@yahoo.com> wrote:
> On Dec 18, 6:30 am, Nolan <***@garmin.com> wrote:
>
> > How much variability is really "real" in power data.  Does a users
> > pedaling efficiency cause a large fluctuation in the data?  If I ride
> > on a trainer at a constant rpm how much variance should there be in
> > the data?   I am new to power and have not analyzed the myriad of
> > files and data that all the people on this list do.   From a user
> > standpoint it looks like the meter is lying to me if I can pedal on a
> > trainer at a constant rpm and get 20 watt differences in the data each
> > second.   So is this why a display filter is used to placate the user
> > that the data is more constant but still record all the differences in
> > the actual saved data?   If all filtering is removed from the saved
> > data does an Edge need to  have a display filter to compensate for the
> > user experience?
>
> As others have said so eloquently (Hi Dave Harris!) the variability in
> the power data is REAL, especially coming from the PT due to the fixed
> time base calculation, as opposed to the calculation being performed
> on an integer pedal cycle basis (i.e. SRM, CinQo, or even Polar
> <sorta>).

I've seen so much discussion on this forum about this. My experience
is derived from building automotive and karting dynamometers.

>From an electrical systems perspective, it is best to filter prior to
sampling to avoid aliasing. The requirement stems from the Nyquist
Sampling Theorem ( ref: ) which states that:

"An analog signal that is bandlimited to B Hz can be completely
reconstructed from samples spaced closer than 1/(2B) seconds apart.
"

What if this condition is not met? Then you get aliasing. When a
magic condition is reached ( noise frequency component is very close
to 1/2 sampling frequency ) then you get really ugly stuff
happening. I think someone called it "precession". I think it's
more precisely called "beating".

More background from Wikipedia about aliasing:

"For a sinusoidal component of exactly half the sampling frequency,
the component will in general alias to another sinusoid of the same
frequency, but with a different phase and amplitude."

To prevent or reduce aliasing, two things can be done:
1. Increase the sampling rate, to above twice some or all of the
frequencies that are aliasing.
2. Introduce an anti-aliasing filter or make the anti-aliasing
filter more stringent.
"


There you have it -- either sample more often or pre-filter the analog
signal _prior_ _to_ _sampling_.




>
> So...in short...filtering of the power data prior to recording?
> Bad.  :-(
>
> User selectable "smoothing" (i.e. simple moving average) of the
> DISPLAYED power (e.g. no smoothing, 2s, 5s, 10s, 30s, etc. averages)?
> Good :-)
--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[Trimix]

I am also riding with powertap cpu and 705 side by side. The distance
rode is way off which I mention before and I see they are working on
it. I have also notice the powertap cpu reacts faster than the
705.Basically the power number are pretty close between the two. If
anyone working on this issue wants data showing the two units working
at the same time they can email me directly.

Ren

On Dec 17, 4:08 pm, Tom_Anhalt <***@yahoo.com> wrote:
> On Dec 17, 7:46 am, Nolan <***@garmin.com> wrote:
>
> > There is a problem in the current Edge release related to computing
> > the distance and speed data from the new Powertap hub.  We know about
> > the issue and have a correction.  It should be released soon as we
> > continue to verify the release.  It is true that there was a small
> > filter on all the power data.  During early testing we would see
> > obvious spikes in power.  This made the Edge appear unreliable even
> > though the data was coming from the meters.  So a small filter was
> > added to reduce obvious invalid wattage values that were coming from
> > the meters.  We are evaluating if this filter needs to be present for
> > all power meters.
>
> Thanks for the input Nolan!  I'd like to add to the request that NO
> filtering be done to the data coming from the ANT+ compatible power
> meters.
>
> However, could you guys PLEASE put a filter on the GPS-based speed, or
> somehow otherwise fix the "speed spikes up/down" recorded when using
> 1s sampling?  Now THOSE are some "obviously invalid" speed values ;-)
--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[J***@aol.com]

I use a Quarq CinQo with an Edge 705 (firmware release 2.5). I also have a
PT 2.4 that I use with the PT CPU. I often use them both on the same
workout, in order to compare the data. It appears that the Edge 705 smooths the
data, since many of the peaks that appear in the PT data, do not appear in the
CinQo data as reported by the Edge 705. Data from longer/steadier intervals
is essentially the same from both units. But, sprints, as reported by the
Edge 705, have the peaks lopped off. My guess is that the CinQo is reporting
the data, but the Edge is not giving output that matches the input. It would
be great to have the Edge 705 data output match the input from the power
meter, without smoothing or filtering of any kind. It appears that in Level 5+
efforts, the Edge 705 is hiding my watts. :)

Thanks, Jim


In a message dated 12/17/2008 10:46:45 A.M. Eastern Standard Time,
***@garmin.com writes:


There is a problem in the current Edge release related to computing
the distance and speed data from the new Powertap hub. We know about
the issue and have a correction. It should be released soon as we
continue to verify the release. It is true that there was a small
filter on all the power data. During early testing we would see
obvious spikes in power. This made the Edge appear unreliable even
though the data was coming from the meters. So a small filter was
added to reduce obvious invalid wattage values that were coming from
the meters. We are evaluating if this filter needs to be present for
all power meters.




On Dec 16, 3:04 pm, Trimix <***@robinhill.com> wrote:
> Yesterday I ran the PT and the 705 while riding a computrainer. I
> turned off the sat function on the 705. I rode a course on the
> computrainer and I notice the speed a on the 705 flashed in and out.
> The miles ridden was also way off. In 20 mile at least 9 miles. If the
> garmin can pull from the hub why is it so off on the 705.
>
> ren
>
> On Dec 16, 12:08 pm, hmspain <***@gmail.com> wrote:
>
>
>
> > I just checked my Garmin 705, and it allows you to "see" the heart/
> > power/cadence (heart from the heart strap, and power/cadence from the
> > hub). The question is, how do you instruct the Garmin 705 to pull the
> > speed from the hub rather than depend on this calculation on the
> > Garmin 705?
>
> > On Dec 16, 8:01 am, hmspain <***@gmail.com> wrote:
>
> > > Once you update your PT hub to ANT+, assuming you have a Garmin 705,
> > > you will have a couple choices to make.
>
> > > 1) Pull the speed from the Garmin (like you have always done), or pull
> > > the speed from the PT hub (along with the power).
>
> > > 2) Pull the cadence from the Garmin cadence device (like you have
> > > always done), or pull the cadence from the hub.
>
> > > Based on the previous discussion, I think the decision is clear, pull
> > > the speed/cadence (and power of course) from the hub. Let the Garmin
> > > do what it does best, that is record the track. No need for another
> > > battery powered device on the bike (the Garmin cadence device), just
> > > pull this data from the hub.
>
> > > BTW, when I was still running the non ANT+ hub, I would import the 705
> > > data into SportTracks and then overlay the PT data :-).
>
> > > On Dec 16, 7:24 am, Howard <***@atlanticbb.net> wrote:
>
> > > > Thanks for all the discussion. What I plan to do while WKO+ is
> > > > working on an update is to ride with both the powertap head unit and
> > > > the edge. I'll download the edge to sportstrack and the powertap
cpu
> > > > to WKO+.
>
> > > > Howard- Hide quoted text -
>
> - Show quoted text -


**************One site keeps you connected to all your email: AOL Mail,
Gmail, and Yahoo Mail. Try it now.
(http://www.aol.com/?optin=new-dp&icid=aolcom40vanity&ncid=emlcntaolcom00000025)

--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[Tom_Anhalt]

On Dec 17, 10:26 am, ***@aol.com wrote:
> I use a Quarq CinQo with an Edge 705 (firmware release 2.5).  I also  have a
> PT 2.4 that I use with the PT CPU.  I often use them both on the  same
> workout, in order to compare the data.  It appears that the Edge 705  smooths the
> data, since many of the peaks that appear in the PT data, do not  appear in the
> CinQo data as reported by the Edge 705.  Data from  longer/steadier intervals
> is essentially the same from both units.  But,  sprints, as reported by the
> Edge 705, have the peaks lopped off.  My guess  is that the CinQo is reporting
> the data, but the Edge is not giving output that  matches the input.  It would
> be great to have the Edge 705 data output  match the input from the power
> meter, without smoothing or filtering of any  kind.  It appears that in Level 5+
> efforts, the Edge 705  is hiding my watts.  :)
>

Umm...that all depends on what you consider is the "proper" method for
calculating/recording power values.

The CinQo is transmitting the average power over a COMPLETE pedal
cycle while the PT is merely calculating/recording the average power
over a FIXED 1.26s time interval, number of complete pedal cycles be
damned. This means that one would expect, especially in a sprint
situation, the PT to report slightly higher "peak" values (i.e. one
record values) than a CinQo, regardless of whether or not the 705 was
applying any smoothing.

I've been running a similar combo for awhile (CinQo/Garmin and PT Pro
wired) and have found that even though the "peak" value in a sprint
may be reported higher with the PT, the 5s average sprint powers
reported are nearly identical. That's the "danger" of looking at peak
1 record power values from a PT...they can be "inflated".

Do a search of the list archives under "precession" and "aliasing"
with regards to PT data and you'll quickly get up to speed ;-)
--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[Dave Harris]

> -----Original Message-----
> From: ***@googlegroups.com [mailto:***@googlegroups.com] On
> Behalf Of Tom_Anhalt
> Sent: Wednesday, December 17, 2008 2:02 PM
> To: Wattage
> Subject: [Wattage] Re: Edge 705 and ant+ Powertap
>
>
>
>
> On Dec 17, 10:26 am, ***@aol.com wrote:
> > I use a Quarq CinQo with an Edge 705 (firmware release 2.5).  I also
>  have a
> > PT 2.4 that I use with the PT CPU.  I often use them both on the
>  same
> > workout, in order to compare the data.  It appears that the Edge 705
>  smooths the
> > data, since many of the peaks that appear in the PT data, do not
>  appear in the
> > CinQo data as reported by the Edge 705.  Data from  longer/steadier
> intervals
> > is essentially the same from both units.  But,  sprints, as reported
> by the
> > Edge 705, have the peaks lopped off.  My guess  is that the CinQo is
> reporting
> > the data, but the Edge is not giving output that  matches the input.
>  It would
> > be great to have the Edge 705 data output  match the input from the
> power
> > meter, without smoothing or filtering of any  kind.  It appears that
> in Level 5+
> > efforts, the Edge 705  is hiding my watts.  :)
> >
>
> Umm...that all depends on what you consider is the "proper" method for
> calculating/recording power values.
>

Hold on there Tom. The way the 705 is currently filtering overall power &
work (kJ) is reduced when compared to a direct PT download. JD is spot on -
the 705 does hide the harder watts. Regardless of your argument for
instantaneous 1s power being accurate or not, it all comes out the wash -
unless it gets a 705 pre-wash.

I haven't done a sprint workout with both going, but that would be
interesting indeed...although I know enough to know what it would look like.

It sounds like the Garmin folks are on it though and we have good things to
look forward to.


--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[Tom_Anhalt]

On Dec 17, 1:21 pm, "Dave Harris" <***@gmail.com> wrote:

>
> Hold on there Tom.  The way the 705 is currently filtering overall power &
> work (kJ) is reduced when compared to a direct PT download.  JD is spot on -
> the 705 does hide the harder watts.  Regardless of your argument for
> instantaneous 1s power being accurate or not, it all comes out the wash -
> unless it gets a 705 pre-wash.
>
> I haven't done a sprint workout with both going, but that would be
> interesting indeed...although I know enough to know what it would look like.
>
> It sounds like the Garmin folks are on it though and we have good things to
> look forward to.

I'm beginning to think the "filtering" may be more of an issue with
the more "stochastic" nature of the PT data stream than the more "well-
behaved" nature of the CinQo data stream. Or...possibly the filtering
isn't applied to the data coming from the CinQo? (Nolan's post above
isn't exactly clear on that) If not, how do explain the following
plots that were the result of plotting the MMPs from both meters
(CinQo/Garmin and PT Pro) over the course of >10 separate rides,
including sprint workouts?

http://www.imagefilez.com/out.php/i311096_CinQoVsPT.jpg

http://www.imagefilez.com/out.php/i311541_PTCinQoTukey.jpg

http://www.imagefilez.com/out.php/i311542_PTCinQoTukey.jpg

On top of that, as I reported before, the calcuated TSS values for
these rides were ALWAYS within a handful of points of each other, and
typically only varied by 1 or 2 points. Although I didn't
specifically look at the total kJs, based on the TSS scores being so
close, I'd hazard a guess that they weren't significantly different
either. I'll tell you what...I'll look them up tonight and post it
later.

It would be interesting to find out if the "filtering" was only
applied to PT data in the Garmin...hmmm....
--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[Dave Harris]

> -----Original Message-----
> From: ***@googlegroups.com [mailto:***@googlegroups.com] On
> Behalf Of Tom_Anhalt
> Sent: Wednesday, December 17, 2008 2:57 PM
> To: Wattage
> Subject: [Wattage] Re: Edge 705 and ant+ Powertap
>
>
>
>
> On Dec 17, 1:21 pm, "Dave Harris" <***@gmail.com> wrote:
>
> >
> > Hold on there Tom.  The way the 705 is currently filtering overall
> power &
> > work (kJ) is reduced when compared to a direct PT download.  JD is
> spot on -
> > the 705 does hide the harder watts.  Regardless of your argument for
> > instantaneous 1s power being accurate or not, it all comes out the
> wash -
> > unless it gets a 705 pre-wash.
> >
> > I haven't done a sprint workout with both going, but that would be
> > interesting indeed...although I know enough to know what it would
> look like.
> >
> > It sounds like the Garmin folks are on it though and we have good
> things to
> > look forward to.
>
> I'm beginning to think the "filtering" may be more of an issue with
> the more "stochastic" nature of the PT data stream than the more "well-
> behaved" nature of the CinQo data stream. Or...possibly the filtering
> isn't applied to the data coming from the CinQo? (Nolan's post above
> isn't exactly clear on that) If not, how do explain the following
> plots that were the result of plotting the MMPs from both meters
> (CinQo/Garmin and PT Pro) over the course of >10 separate rides,
> including sprint workouts?
>
> http://www.imagefilez.com/out.php/i311096_CinQoVsPT.jpg
>
> http://www.imagefilez.com/out.php/i311541_PTCinQoTukey.jpg
>
> http://www.imagefilez.com/out.php/i311542_PTCinQoTukey.jpg
>
> On top of that, as I reported before, the calcuated TSS values for
> these rides were ALWAYS within a handful of points of each other, and
> typically only varied by 1 or 2 points. Although I didn't
> specifically look at the total kJs, based on the TSS scores being so
> close, I'd hazard a guess that they weren't significantly different
> either. I'll tell you what...I'll look them up tonight and post it
> later.
>
> It would be interesting to find out if the "filtering" was only
> applied to PT data in the Garmin...hmmm....


I’m not sure I understand what you are trying to s how in your plots. The
first one I couldn't access, the others show significant differences - but
this is PT vs Cinqo, right?

The point here is that to get to the jist of this thread you'll need to have
2 different receivers recording data from a single power meter, where one of
those receivers is the 705. Throw in 2 power meters and all bets are off.


--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[Greg Steele]

Dave,

I just thought of something though... Doesn't WKO+ manipulate the data
as well?

In that powertaps record @ 1.26s per record and the data (at least as
displayed in wko+) is on 1s per record?

And could this not also be an artifact of that?

Drop me the three (wko+ of each and native pt) if you would as I would
like to poke at them....

g


Dave Harris wrote:
>
>
>> -----Original Message-----
>> From: ***@googlegroups.com [mailto:***@googlegroups.com] On
>> Behalf Of Tom_Anhalt
>> Sent: Wednesday, December 17, 2008 2:57 PM
>> To: Wattage
>> Subject: [Wattage] Re: Edge 705 and ant+ Powertap
>>
>>
>>
>>
>> On Dec 17, 1:21 pm, "Dave Harris" <***@gmail.com> wrote:
>>
>>> Hold on there Tom. The way the 705 is currently filtering overall
>> power &
>>> work (kJ) is reduced when compared to a direct PT download. JD is
>> spot on -
>>> the 705 does hide the harder watts. Regardless of your argument for
>>> instantaneous 1s power being accurate or not, it all comes out the
>> wash -
>>> unless it gets a 705 pre-wash.
>>>
>>> I haven't done a sprint workout with both going, but that would be
>>> interesting indeed...although I know enough to know what it would
>> look like.
>>> It sounds like the Garmin folks are on it though and we have good
>> things to
>>> look forward to.
>> I'm beginning to think the "filtering" may be more of an issue with
>> the more "stochastic" nature of the PT data stream than the more "well-
>> behaved" nature of the CinQo data stream. Or...possibly the filtering
>> isn't applied to the data coming from the CinQo? (Nolan's post above
>> isn't exactly clear on that) If not, how do explain the following
>> plots that were the result of plotting the MMPs from both meters
>> (CinQo/Garmin and PT Pro) over the course of >10 separate rides,
>> including sprint workouts?
>>
>> http://www.imagefilez.com/out.php/i311096_CinQoVsPT.jpg
>>
>> http://www.imagefilez.com/out.php/i311541_PTCinQoTukey.jpg
>>
>> http://www.imagefilez.com/out.php/i311542_PTCinQoTukey.jpg
>>
>> On top of that, as I reported before, the calcuated TSS values for
>> these rides were ALWAYS within a handful of points of each other, and
>> typically only varied by 1 or 2 points. Although I didn't
>> specifically look at the total kJs, based on the TSS scores being so
>> close, I'd hazard a guess that they weren't significantly different
>> either. I'll tell you what...I'll look them up tonight and post it
>> later.
>>
>> It would be interesting to find out if the "filtering" was only
>> applied to PT data in the Garmin...hmmm....
>
>
> I’m not sure I understand what you are trying to s how in your plots. The
> first one I couldn't access, the others show significant differences - but
> this is PT vs Cinqo, right?
>
> The point here is that to get to the jist of this thread you'll need to have
> 2 different receivers recording data from a single power meter, where one of
> those receivers is the 705. Throw in 2 power meters and all bets are off.
>
>
> >
>

--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[Alex Simmons]

On Dec 18, 11:27 am, Greg Steele <***@wattagetraining.com> wrote:
> Dave,
>
> I just thought of something though... Doesn't WKO+ manipulate the data
> as well?
>
> In that powertaps record @ 1.26s per record and the data (at least as
> displayed in wko+) is on 1s per record?

Not in my WKO+ it doesn't. PT data is shown in 1.26 second intervals,
or at whatever interval frequency the PT CPU stored the data at.
--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---

[Dave Harris]

I'm butting up against an ugly work deadline so this will be short...

You are comparing 2 PMs. That is fraught with error. I'd expect the Cinqo
to read higher than the PT - the PT is known to be stingier than any other
PM due to driveline friction. Based on your description it seems the 705 is
scrubbing an amount roughly equal to driveline losses.

Honestly I could be way off - and I don't really care. The 705 filters -
and loses - data in my system. The Garmin rep said as much on wattage
today.

On this part:

> Doesn't the
> above HARD DATA imply that the 705 is doing something different with
> the data stream coming from the PT than it does with the CinQo?

Not at all - wouldn't you expect the CinQo to read higher? Where did the
driveline losses go?

One thing I'll say as it really makes the Cinqo look good.

I think this horse is flogged!

> -----Original Message-----
> From: Tom_Anhalt [mailto:***@yahoo.com]
> Sent: Wednesday, December 17, 2008 5:47 PM
> To: Dave Harris
> Subject: Re: Edge 705 and ant+ Powertap
>
> Dave, I tried sending this to the list, but google seems to be having
> fits right now...
>
> On Dec 17, 2:09 pm, "Dave Harris" <***@gmail.com> wrote:
>
>
> >
> > I’m not sure I understand what you are trying to s how in your plots.
> The
> > first one I couldn't access, the others show significant differences
> - but
> > this is PT vs Cinqo, right?
>
> Try looking at it here:
> http://img367.imageshack.us/my.php?image=cinqovsptgq1.jpg
>
> That's a plot of CinQo/Garmin MMPs (for the "stock" WKO+ durations)
> vs. PT MMPs for >10 rides. The slope of the line is .9968 (i.e. on
> average, the readout for the CinQo/G was 0.3% lower than the PT) with
> an R^2 = .999
>
> That says that for the MMPs that make up the MMP chart in WKO+, on
> average the power values would vary by only 0.3% with a fairly high
> "goodness" of fit.
>
> The "Tukey Mean-Difference" plots show whether or not the variation is
> a function of mean power on an absolute or percentage basis. On a
> percentage basis, the difference actually NARROWS at the higher power
> levels
>
> >
> > The point here is that to get to the jist of this thread you'll need
> to have
> > 2 different receivers recording data from a single power meter, where
> one of
> > those receivers is the 705. Throw in 2 power meters and all bets are
> off.
>
> Right...you're comparing the recordings of 2 separate head units from
> a single hub. But, I'm showing that using the same 2 head units
> (sorta...the PT you have has been changed to be at 1s intervals) BUT
> with the 705 recording the power values from a DIFFERENT power meter,
> the "agreement" is BETTER than your case. Don't you find that odd?
>
> Are you implying that adding in the extra "uncertainty" of the
> separate power meter makes the errors LESS on average?? Doesn't the
> above HARD DATA imply that the 705 is doing something different with
> the data stream coming from the PT than it does with the CinQo?


--~--~---------~--~----~------------~-------~--~----~
You received this email because you're subscribed to the Google [Wattage]
group at http://groups-beta.google.com/group/wattage?hl=en.

To UNSUBSCRIBE, send a blank email to:
mailto:wattage-***@googlegroups.com
-~----------~----~----~----~------~----~------~--~---