|
Nov 2018
2:45pm, 13 Nov 2018
17,991 posts
|
flanker
Pretty much the same for my F3HR. It's good for lifetime tracking, but not much use for vigorous activity. If you are running with it, it needs to be on tight, and even then position, skin colour, 'hairyness', etc all can affect accuracy.
|
|
Nov 2018
2:54pm, 13 Nov 2018
65,320 posts
|
Gobi
Currently testing a forerunner 35
Usual good for life tracking
1 run easy done and data accuracy was ok
Shall do a speed session and report back
|
|
Nov 2018
7:32pm, 13 Nov 2018
13,624 posts
|
Bazoaxe
The optical on my new 735 seems pretty good. It appears higher than I would have thought overnight, measures standard running pretty close to what I expect, but under records any faster runnning
|
|
Nov 2018
8:33pm, 13 Nov 2018
13,464 posts
|
Chrisull
"under records any faster runnning" - that seems to be the default for any optical monitor.
|
|
Nov 2018
8:55pm, 13 Nov 2018
9,980 posts
|
Badger
Well, any optical monitor that's mounted in the back of a watch, anyway. (No, I did not run a 5k PB in zone 2...).
|
|
Nov 2018
7:50am, 14 Nov 2018
29,535 posts
|
HappyG(rrr)
Badger, see that jump just after half way? Is the reading "more correct" after the jump, would you say then (if you were flat out all the way in a 5K)? Mine does that. It goes along in the 150s/160s, then suddenly jumps up to 180s and stays there. I don't know which is right, but it means at least half is wrong, and none of it is helpful! Worse than useless, for me.  G
|
|
Nov 2018
8:45am, 14 Nov 2018
6,033 posts
|
larkim
Listening to the parkrun FWT podcast this morning, the chair of the parkrun research board happened to rattle off some info about wrist-based HRMs. As well as saying that broadly +/- 3 or 4 beats was about as accurate as he expected them to be, he did say that the jumps around (as shown in the graph above) are down to the algorithm's in the watch trying to deal with the data they are receiving, and they may flip between different solutions to the mathematical problem they are trying to solve - i.e. when they get messy data, how to report it. I suppose it would be trivial to assume that they simply read each "blip" they detect, but I suspect the real processing is more complex than that dealing with stronger or weaker impulses as the body / wrist moves around etc.
|
|
Nov 2018
9:03am, 14 Nov 2018
29,536 posts
|
HappyG(rrr)
Makes sense larkim. But also makes the effectively useless, if the sort of graph Badger shows (that I also get) is common.
I have tried both wrists, tightened right up etc. I don't know what else I can do? I'm going to try Garmin support next. Even though I don't expect them to be able to fix it, they need to know that for a proportion of users that it's useless. Shame because I like the idea of having HR data for all runs as well as for all day comparison. But just now, it tells me nothing. Or worse it tells me lies! G
|
|
Nov 2018
9:24am, 14 Nov 2018
17,993 posts
|
flanker
if it's jumping to a value of around 180 is it picking up cadence or static from clothing? I've got a couple of race l/s shirts that guarantee to throw the reading right off.
|
|
Nov 2018
9:37am, 14 Nov 2018
9,981 posts
|
Badger
I should explain the graphs. Two watches, one using a chest strap and one my Scosche Rhythm+. The line at the bottom is the difference. The jump is real, and it is me putting in a sprint to drive my HR up. The difference is pretty small, and some of it is the traces not being perfectly synchronised because I didn't start the watches at exactly the same time.
|
