Cursor Misalignment in Intensity Graph (LV6.1)

[oldshaghat]

Howdy.

I have an Intensity Graph, with a 2D array of data (about 300x300) and a cursor. The cursor is locked to the plot, which for intensity graphs, means that the cursor is drawn at the lower left corner of the pixel that is being addressed.

Most of the time this works fine. However, sometimes, the cursor is drawn as being located at one pixel, but the values reported for the cursor are in fact a neighbor (usually the one to the right). The zoom level of the graph is such that a given pixel is larger than a screen pixel (about 3-5 times), so I wouldn't think that it was an aliasing problem, but the problem does go away if I zoom in a little further.

Has anyone else seen this? Does anyone have a fix?

[chilly_charly]

The problem does not seem to happen with LV 7.1. But may be I didn't use the properr size. Could you post a demo vi ?

Why don't you unlock the cursor ? That would be less confusing for a standard user, anyway ...

[oldshaghat]

I don't have a repeatable set of conditions that lead to this problem, so I don't have a demo vi that exhibits it.

As to unlocking the cursor, I think that in an intensity plot that is more confusing; the location and value reported for a cursor position are shifted half a pixel down and half left with respect to the way the plot is filled in. In other words, say we have a 10x10 plot shown in 100x100 pixel window. Each intensity value is 10 pixels wide. The axis values line up on the boundaries of these pixels. When the cursor is within half a unit (5 pixels) of the crossing point of these axes, that is, near the intersection of boundaries of four adjacent pixels, the value reported is the pixel location and value in the upper right of these four pixels. This means that with an unlocked cursor, I can be positioning it over (visually) the lower left pixel, but selecting the upper right. In fact 3/4 of any given pixel's visual representation is actually going to select some other pixel. So I've locked the cursor to remove that uncertainty; now it is supposed to be always in the lower left corner of the pixel selected.

[chilly_charly]

You made me nervous ! 😉
I have verified the cursor behaviour. I do not obtain the result that you have described. The "pixel" intensity correspond always to the cursor position, and there is no shift in X and Y.
So, unlocking the cursor is clearly the right option.