DV Info Net - deinterlacing effect

I'll take a stab at this. It doesn't matter whether the material was shot in 30i,30p, 24p or anything else. When it is ready to be viewed on an NTSC television it is converted to 30i which means that the frame, whatever its source, will be split into two fields which will be recorded and displayed subsequently. If the source video was 30p that means that a single picture was taken every 30th of a second. This picture will be separated into high and low fields when displayed on a TV set. You still get the whole picture - it's just presented in two parts written 1 60th of a second apart. If the source video was 30i then the image was captured as two fields captured 1/60 of a second apart. Thus a moving object will appear in two different places in a 30i image corresponding to the amount it moved in 1/60th of a second. If you display this image progressively, i.e. present both frames at once, then the two displaced images will be seen and this is most annoying. Thus the NLE and photo editors have a deinterleave function which will remove either the high or low field leaving every other line blank. These software packages usually fill the blank lines by interpolation between the retained lines. There is no need to deinterleave in any image that was captured progressively (30p, 24p) because both fields come from the same exposure. To deinterleave would accomplish nothing but throwing away half the vertical resolution of the image.
[Note: 24p with 2:3:3:2 puldown will have 1 frame in five with the two fields from separate images. With 2:3 pulldown 2 frames in five will have fields from different images.]