Finally, we use this same test device to make additional measurements we feel are of interest to users.
First of all, we measure contrast as a function of brightness. Here again, the method used by manufacturers is clearly not really representative of reality...
Most manufacturers use the FO-FO (Full ON, Full OFF) method:
- Take a white image, measure the light intensity at the center of the display: Ion
- Take a black image, measure the light intensity: Ioff
- The contrast is the ratio Ion/Ioff.
It results in some impressive figures... as impressive as they are unrealistic and useless. We prefer to use the ANSI method:
- On a single image, we display eight white rectangles and eight black rectangles
- We measure the brightness of each black rectangle, then of each white rectangle.
- We calculate the average contrast of the adjacent rectangles.
- We take the average of the contrasts to obtain the ANSI contrast.
This method is more representative of reality, since what we are interested in is the contrast that can be obtained within a given displayed image. Obviously, the resulting contrast values are lower. We've chosen the ANSI method because it corresponds better to how a monitor is actually used.
Another advantage is that it helps you find the ideal point for the best possible contrast. Some monitors' contrast doesn't change with brightness, which allows finer adjustment. Finally, we measure the spatial uniformity of brightness over the panel. Some displays are brighter at the top of the panel than at the bottom, which can be bothersome. So, for a given brightness level (50%, for example), we measure the intensity at 64 areas on the panel to check its uniformity.
Here, for example, is the spatial uniformity of the Prophetview 920 Pro:
The upper left corner of the graph is also the upper left of the panel. The 95-100% area is the brightest one measured. The other values are expressed as a percentage of the maximum brightness. You can see that the Prophetview 920's panel is not perfectly uniform; but it will serve as our reference for future tests.