Incompetent graphs
-
Dimitrios Kanellopoulos Community Managerlast edited by Dimitrios Kanellopoulos 3 May 2019, 22:23 5 Mar 2019, 22:22
@remib Gavarnie looks fantastic!
I cannot share more on this subject, also different thread is needed. A quick example would be that a lot of Ambits 3/ Traverse have issues come connectivity issues from our feedback so #1 prio is to make the basics right for them before we jump to maps etc, where more people end up having connectivity issues trying to use features they would like to.
-
@Dimitrios-Kanellopoulos said in Incompetent graphs:
above is FullScreen FYI
cool, goes into right direction, suggestion: can we normalize the x-scale to full/rounded integer instead current granularity of e.g. (your example here) 23:11 as half/50% of the total training duration? This didn’t add any value to an end user glacing on a 5 inch screen and creates unnecessary noise on the x-axes labels. Rather would prefer normalized value on x (aka comparable buckets of the plot) like 5/10/15min or 2/4/6/8min that is easier for comparing and didn’t add this noisy level on the graphs with the 4 digits , if I want second-to-second accurate data I would go anyway into the detail map view
Polar flow or movescount web/old mobile app does this as well *see below some examples)
current SA (lot of digits on axes, not normalized):
movescount web (rounded full integers, normalized):
old movescount mobile app (normalized):
Polar flow (normalized):
-
@Tobias-F Great idea about the normalized number on x axis. I wanted to suggest the same this morning but didn’t have time. In addition to removing clutter that would also make it easy to estimate values in between by glancing at the graph. For example you know instantly that there is 7:30 between 5:00 and 10:00 but the same isn’t true for arbitrary time labels. It would be great if numbers on y axis were normalized too.
-
@silentvoyager said in Incompetent graphs:
make it easy to estimate values in between by glancing at the graph
fully agree!
-
Yeah good points here. Ill pass those to the developers asap. Thanks guys. Good morning from here.
-
@Dimitrios-Kanellopoulos said in Incompetent graphs:
Yeah good points here. Ill pass those to the developers asap. Thanks guys. Good morning from here.
Take note of those last things for your QS too, good notes for both QS and SA.
-
Today I saw one fairly high profile athlete posting a picture generated by SA - one that shows elevation profile on top of a photograph. That was shared with a pretty wide audience of trail and mountain runners - more than 4000 people. And while the picture looks pretty cool and features Suunto logo, it does show the same incompetence. The profile is from out and back Grand Canyon R2R2R run and therefore it should be symmetrical. But on the picture the descent from the north rim looks quite a bit longer than the ascent. That wouldn’t make any sense to anyone who’d pay attention. Please fix this sooner.
-
@silentvoyager You are talking about time vs distance axis right?
-
@silentvoyager I like to come back to the initial thing and wanted to ask, if you had some breaks at about 1:35, 2:15 and around 3:25? I have seen a similar issue at the Polar Vantage, it seems that the time spent in the breaks are influencing the lap overal time (becomes longer) the lap data is the expanded before representation to the longer lap time but squeezed later in the real active lap time. Someone understood this (except me)?
-
@johann-fuehrer yes that is a known issue.
-
@johann-fuehrer On iOS fixed on 1.9.0 and higher, on Android should come with the same iOS lap table that we are building.