The summary pretty much says it. I don't have a position marker using Marble 2.01 from Suse Tumbleweed rpm's, plasma 5.8.0, frameworks 5.26.0, QT 5.7.0. This situation has obtained for a couple of months of relatively frequent Tumbleweed updates identified this way... 2016-07-28 15:57:23|install|marble-kde|16.04.3-1.1|x86_64| 2016-08-10 21:57:20|install|marble-kde|16.04.3-2.1|x86_64| 2016-09-03 10:11:38|install|marble-kde|16.08.0-1.1|x86_64| 2016-09-09 19:22:44|install|marble-kde|16.08.0-1.2|x86_64| 2016-09-16 15:57:36|install|marble-kde|16.08.1-1.1|x86_64| 2016-10-12 23:10:56|install|marble-kde|16.08.1-1.2|x86_64| It possibly started with 16.08. Possible relevance--I am getting gps from an Android phone using application 'GPS2NET', and on the Linux box,' adb forward tcp:6000 tcp:6000', followed by 'gpsd -n tcp://localhost:6000'. xgps doesn't show a satellite selected, and doesn't show EPX, EPY, EPS, EPD. The lack isn't new, though. Emitted sentences: GPGSA, GNGSA, BDGSA, GPRMC, GPGSV, GLGSV, GPGGA, GPVTG, GNGNS That's all for now. Can anyone confirm the lack of a position marker? Reproducible: Always By the way, tracks became narrow white lines, practically invisible, at the same time. Is that expected? Invisible lines aren't very useful.
I just noticed the position marker is visible at zoom levels 5 and below. It is also visible at zoom level 6 if positioned well below the crosshairs; that is, below about the 3/8th of the screen height.
This bug also affects me. The position indicator does not render correctly when the map is zoomed in more than a few ticks. It does not matter if the position is sourced from GPS, Qt Positioning Location, or Current Route. Both marble (KDE) and marble-qt are affected. While the marble package from Ubuntu 16.04 (deb version 4:15.12.3-0ubuntu2) was OK, the marble package from Ubuntu 17.10 (deb version 4:17.04.3-0ubuntu1) is bugged. I have bisected the history and found the offending commit: d03d2483bae, "Fix bug that the position marker was showing up on the..." Reverting this commit on top of v17.11.80 and recompiling fixes the bug. I live in the northwestern hemisphere: i.e., positive latitudes and negative longitudes. It could be that the changes introduced in the above commit do some math which doesn't quite work out world-wide.
Changing to "confirmed" as bug affects more than one user. A fix is proposed in https://phabricator.kde.org/D11853