Pixel Scaling by Row or Column Removal

Normally when rasterizing a Gerber file, the resolution along X and Y are the same since the output device's pixels are square or round.

GBR_RIP includes the ability to specify different scales in X and Y but this is normally used to correct for very slight distortions due to substrate stretching or other machine imperfections. This means that the difference in X and Y scales is very small. Such scaling is done on the input vector data.

What if our equipment requires a large difference in X and Y scale? For example, suppose each pixel on our machine is twice as wide as it is high? In that case we may need the rasterizer to produce twice as many pixels in Y as in X so that the final output looks correct.

We find that this is most effectively implemented by oversampling and then removing rows or columns of pixels after rasterizing. The command line options dpx: and dpy: activate the row/column removal.

Row or Column Removal

If columns are removed, the image is "squashed" along X; if rows are removed the image is squashed along Y. Due to the nature of the TIFF file, it is much faster to remove rows than columns. Various settings are shown below on the "happy" face image. [hface.gbr 10KB RS247X file.]