Any tricks to increase accuracy of current tempo while running?

Ruben GB said:

Is it possible to turn on / off calibration during a training?

Yes, although I never tried it, I see no reason why it should not be possible, since the settings menu is accessible even during an activity. I do not think it is necessary though. The algorithm sees when the GPS signal is accurate, and when not, so it should be enough spending over 2 hrs under good conditions (not necessarily in a single run).  

Hi again

Finally I have bought the HRM Pro + belt

It is still calibrating, around 1h:30 usage. I hope by the end of the week it will have enough data to be calibrated. My current configuration is Auto calibration ON / Pace Indoors / Distance Indoors.

Some questions regarding this issue

1. Is there any way to know when the 2hr calibration threshold of "good data" is reached?

2.  Outdoors it seems that still only GPS is used. I see the same pace issues in the same places where I normally had had it. When will it start to use the HRM Pro data in places with bad GPS signal? Must I change Auto Calibration to OFF to get it?

3. Currently I am running in my "standard" pace between 5.30 and 6.30 pro KM, in flat terrain and also some short (500m) step climbs and that will be my start calibration. If later I want to include the calibration for my max pace (around 4.00 pro KM during 200-300m for interval training for example), and I switch on calibration again,  will it add this to my previous calibration or will it overwrite it?

Thank very much in advance for your help

Ruben GB said:

1. Is there any way to know when the 2hr calibration threshold of "good data" is reached?

Try a short run using the Indoor Run or Treadmill activity, and see how well it works

Ruben GB said:

  Outdoors it seems that still only GPS is used

No, outdoors, the combination of GPS and accelerometer (internal or the one in the HRM) is always used

Ruben GB said:

Auto Calibration to OFF to get it?

You can keep the auto-calibration on. You can also turn it off, if you are happy with the accuracy.

Ruben GB said:

will it add this to my previous calibration or will it overwrite it?

There is only one calibration, but the stride length is being dynamically adjusted, using the cadence, hence it should work for a wider range of paces. I recommend calibrating at the standard pace, though.

Thank you again for your response

Surely I still haven't reached the calibration 2+ hours, but by the end of the week I can make your suggested test in a known distance outdoors but recording it as an indoors activity. I will also check using only HRM for pace outdoors (distance will be always GPS as I am pretty happy with the results) and see what happens. 

Regarding the usage of HRM Plus outdoors with my current configuration, perhaps the problem is that I still haven't reached the 2 hours calibration. Today I saw the same issues as always in the same places (trees + buildings with 5-6 floors near). Paces of 8:00 / 9:00 km when in reality I was running during my warm up 6:45-7:15 in flat terrain. I Will check it also later when I have more calibration data.

Regards,

Ruben GB said:

2.  Outdoors it seems that still only GPS is used. I see the same pace issues in the same places where I normally had had it. When will it start to use the HRM Pro data in places with bad GPS signal? Must I change Auto Calibration to OFF to get it?

I use F6 + HRM Pro for multiple years. I use your settings and the pace is very variable during a run with quite constant pace (eg. super flat surface). I am not convinced that with your settings HRM is used for pace estimation. It is clearly stated in the settings of the watch - "pace - indoor only". I found no mention in the documentation, that HRM is used for pace. If yes, it still puts lot of weight to GPS and with F6 you will never get precise enough GPS readings in areas where som signal reflections could occur. HRM is used for other useful statistics such as ballance, vertical oscillation etc. You can try to set the mode for pace to "always", but you cannot use the autocalibration.

1000000000 said:

I found no mention in the documentation, that HRM is used for pace.

Yes, it is. I wrote a Connect IQ app showing me the GPS-only pace, as well as the adjusted pace, using the (internal or external) accelerometer. It clearly shows that the accelerometer is being used also during GPS runs, whenever the signal is not sufficiently strong and accurate, and that happens even with the setting "indoors only".

You may be also able to test it yourself too even without the CIQ app - run the same outdoor distance with and without the HRM (using the setting "indoors only"), and see whether the distance is really identical. If you are lucky, and the accelerometer of the standalone watch is well calibrated, the pace and distance will be identical. However, in most cases the distance and pace will differ. In my case the standalone watch is almost always 10% off the true distance. And when using the HRM (with the option "indoors only"), the distance is within 1% of the true distance. That clearly proves the HRM accelerometer is used even outdoors, despite using the setting "indoors only".

trux said:

Yes, it is. I wrote a Connect IQ app showing me the GPS-only pace, as well as the adjusted pace, using the (internal or external) accelerometer.

It uses accelerometer of the HRM or accelerometer of the watch? I see no difference in pace variability during a constant rune using HRM or watch only.

1000000000 said:

I see no difference in pace variability during a constant rune using HRM or watch only.

HRM moves practically in a straight line following the track, without changing the orientation (more or less practically a 1D motion), unlike the watch constantly moving in all possible directions, as your arm swings around (3D motion). That makes it much more difficult for the standalone watch to get the correct estimation. And it is even worse on watch models without the gyro sensor. Believe me, there is a huge difference between the HRM and the standalone watch.

trux said:

Believe me, there is a huge difference between the HRM and the standalone watch.

Not for my case. You are, if I am understand you correctly, put a lot of weight on bias of overall distance (with watch you stated that you are always 10 % off some sort of reference distance). I do not have any issue with bias, my issue and most likely the issue of Ruben is quite high variance of estimates of actual pace. 

Indeed, estimation based only on accelerometer with device that moves in many directions needs the filtering of gravity acceleration. But that is another topic.