GNOME Bugzilla – Bug 355069
Parameters used in chronogram object
Last modified: 2019-03-20 11:27:18 UTC
Hi, As an electronic engineer I think the chronogram object is very useful but I think there is a mistake in the way the parameters are used to draw the shape. When you specify the timing information in the data index (I use the french release and it is named donnée I suppose that in english it is data) you specify the duration of the signal. In the parameter index you specify the rise and fall time of that signal. When the signal is drawing the rising time and falling times are added to the duration. That is the problem, the rising and falling time are degradation of the signal they should be substracted from the duration. Say you define a signal wich is 2 time units high and three time units low. If the rise and fall time are 0 (perfect circuit) the overall period is 5 time units. If the signal is fed by a non perfect circuit the rise and fall time will increase but the total period should not change. Otherwise when you increase/decrease rising/falling time you modify the signal period. ______ ______ | |____________| ______ ______ / \____________/ Rising time should be substracted from up duration and fall time should be substracted from down duration. Regards, Frederic
Lars asked for opinions on the mailing list, so I'm offering mine here. If I understand the submitter right, then his suggestion is correct, but there is a simplification involved. You usually count a signal as being a '1' when it has reached 90% of its full level and as '0' when it has decayed to 10%. Other times are also possible, according to e.g. http://www.aubraux.com/design/rise-fall-time.php, but 90%/10% are the most common IMO. But I could imagine that one sometimes wants to have the times counted from 50% of the full level, so the ideal solution would be some per-edge customization (maybe even with a small line and text saying e.g. "50%"), with a sensible default (90%/10%). Not sure if that's feasible.
I agree with Colin. The simplest will be, as suggested earlier, to include all the rise and fall into the pulse length, but the 10% 90% is good. If the option to make the threshold user selectable does not involve too much code, that would be ideal, with the suggested 10% 90% defaults.
Setting severity enhancement, patches accepted;)
-- GitLab Migration Automatic Message -- This bug has been migrated to GNOME's GitLab instance and has been closed from further activity. You can subscribe and participate further through the new bug through this link to our GitLab instance: https://gitlab.gnome.org/GNOME/dia/issues/177.