porg
Active Member
I reworked the help screen a bit, is this better?
- It's good that both sides have their heading now, more clarity.
I suggest to name the headings "Info panels" & "Commands".
- The grey box behind "Info panels" is irritating. If a box there, then also one behind "Commands". At best no box at all. As you have headings on both sides now, the content segregation is now anyhow clear.
- But again, this differing notation "(...)" is irritating again. Better would be to use:
<left-arrow-symbol> Previous panel
- <right-arrow-symbol> Next panel
- left/right previous/next panel
Great!I managed to fix the timeshift issue without using any more resources
The point here is, that the keypress action must have a feedback in the GUI, so we must anyhow at least inform with "Logging stopped, file closed" and then hide it after $info-timeout secs, thus I thought why not add the specific info right away with it. If you remember the "…hhmm…" part (only 4 digits) then you most likely can identify your session, the date is likely the same for all of your successive tests.The only thing I'm not gonna do is to display the log filename when you stop logging
If you can manage to include that gnuplot into the package, and that sysinfo properly interacts with it, great!Server based plotting service would be possible but quite a bit of work to make it work reliably. It might be easier to package gnuplot in a PND and add some basic zenity stuff to select a logfile to plot.
My thought behind server-side plotting was to establish a standard graph look, so that this can work as a standard for Pandora users comparing their specs, avoiding to get plots from dozen different softwares with hundred of different graph styles.
By supplying sysinfo with gnuplot and your script interface between them, this comparability is achieved! Users will generate their plots with the same software & settings, and can upload those pix right away. Wonderful!
UPDATE: Keep in mind to create the plot templates in a way that they are suited for comparison. For the Y-axis use 0 as the bottom point and the possible design maximum value (+ some extra, should some units maybe top that) as the top point. Then you can put graph pix side-a-side (or overlay them with reduced opacity), and get a reliable comparison. For the x-axis (time) I think it is not really possible to use a single standard, as some observations may be for minutes, some for hours, some for days. In comparison situations one must (mentally) stretch them accordingly. But if the Y-axis is suited for comparison, then, that's a big point!
Last edited by a moderator: