BreakPoint
Monthly bugs, August 2008
Lets all celebrate the Swiss Independence day: August 1! 😄
This is the continuation of the July 2008 bug thread. The rules are as always:
- Each post here should only contain a brief description of a bug, together with a link to the thread where it is being discussed.
- One post/bug!
- One bug/post!
- No discussion of bugs. All discussions of a bug must take place in the original linked thread.
I just downloaded LV 8.6, as I could not wait to have a look at the new long time awaited 3D graphs primitives, said to run also on MacOS. This is a great new feature.
Of course I found a collection of new bugs (probably known already by the development team since they are pretty obvious, but who knows...)
Additionnaly, I enjoy the idea that this seems to be the first LV 8.6 bug report in this forum.
OK, here we go : (all the bugs have been found on a WindowXP machine)
1 - As shown in the attached picture, a transparent cursor plane hide the 3D surface graph (but shows the axis)
2 - As shown on the same picture, the boundaries of a cursor plane are locked to the axis values defined by the axis autoscale mode. The cursor plane appears much wider than the axis.
3 - While in "perspective" mode, when the 3D indicator size is changed the graph switches to "orthographic" mode. An access to the property window is required to return to "perspective" mode.
4 - after an access to the property window, the graph (perspective mode) in the Render window is streched to fit the window area, but returns back to a normal aspect ratio as soon as the window size is slightly changed.
5 - As shown in the second attached picture, after a slight modification of the window size, the 3D surface is chopped of during a zoom-in.
6 - When the 3D indicator size is streched, the graph is also streched. That gives the graph a poor look during rotation. In the old 3D activeX the aspect ratio of the graph was preserved.
7 - (Not a bug, but a major lack of functionnality) I could not find a way to draw a 3D surface and a line on the same graph. This is easy with the old ActiveX and is a very usefull function when you want to represent the trajectory of an object moving onto a surface.
8 - (Not a bug, but a strong inconvenience) A predefined color spectrum is lacking
9 - (Not a bug, but a minor inconvenience) No logarithmic mode for the axis scales
I better stop now since my wife is desperately asking me to pack my luggage before we leave for NIweek ! See (some of) you soon in Austin !
%20bug1.png)
%20bug%205.png)
Message Edité par chilly charly le 08-03-2008 07:55 AM
Of course I found a collection of new bugs (probably known already by the development team since they are pretty obvious, but who knows...)
Additionnaly, I enjoy the idea that this seems to be the first LV 8.6 bug report in this forum.
OK, here we go : (all the bugs have been found on a WindowXP machine)
1 - As shown in the attached picture, a transparent cursor plane hide the 3D surface graph (but shows the axis)
2 - As shown on the same picture, the boundaries of a cursor plane are locked to the axis values defined by the axis autoscale mode. The cursor plane appears much wider than the axis.
3 - While in "perspective" mode, when the 3D indicator size is changed the graph switches to "orthographic" mode. An access to the property window is required to return to "perspective" mode.
4 - after an access to the property window, the graph (perspective mode) in the Render window is streched to fit the window area, but returns back to a normal aspect ratio as soon as the window size is slightly changed.
5 - As shown in the second attached picture, after a slight modification of the window size, the 3D surface is chopped of during a zoom-in.
6 - When the 3D indicator size is streched, the graph is also streched. That gives the graph a poor look during rotation. In the old 3D activeX the aspect ratio of the graph was preserved.
7 - (Not a bug, but a major lack of functionnality) I could not find a way to draw a 3D surface and a line on the same graph. This is easy with the old ActiveX and is a very usefull function when you want to represent the trajectory of an object moving onto a surface.
8 - (Not a bug, but a strong inconvenience) A predefined color spectrum is lacking
9 - (Not a bug, but a minor inconvenience) No logarithmic mode for the axis scales
I better stop now since my wife is desperately asking me to pack my luggage before we leave for NIweek ! See (some of) you soon in Austin !
Message Edité par chilly charly le 08-03-2008 07:55 AM
There's an apparent bug that got started with 8.5 (8.2 and earlier are OK).
Look at the picture in this thread:
As you can see, the "number to decimal string" shows a red coercion dot.
Wiring an array to any of the integer type "number to (decimal|octal|hex) string" conversions shows a coercion dot (any array input, from I8...U64, FXP, DBL, EXT will do). The coercion dot only disappears if I wire a scalar.
The noninteger type conversions "number to (fractional, exponentional,engineering) string" never show a coercion dot. This seems like the correct behavior.
Here is a more general illustration:
Message Edited by altenbach on 08-07-2008 04:05 PM
Thanx Mathan
So this is the image altenbach had mentioned.....
Message Edited by muks on 08-08-2008 07:40 AM
So this is the image altenbach had mentioned.....
Message Edited by muks on 08-08-2008 07:40 AM
Most VIs from the Math...Geometry palette return stale data (+an error) if the input arrays are empty array.
Correct would be to (1) return an error and (2) empty arrays.
See:
According to some casual testing, all 6 VIs are affected.
Message Edited by altenbach on 08-12-2008 04:14 PM
%20bug1.png){kind=link}
%20bug%205.png){kind=link}
