Skip to main content

Benchmark Comparison

Every conducted and radiated scan is repeated on a benchmark: a shipping, CE-marked marine display of similar class. Without calibrated instruments, the benchmark is what turns a trace into a judgement, because it shows how a product that passed formal testing looks on the same bench.

Like with like​

The comparison is valid only if everything except the device is identical:

  • the same supply, being the battery;
  • the same LISN and the same terminations;
  • the same antenna position and distance;
  • the same cable routing; and
  • the same analyser settings.

The benchmark runs at its brightest and busiest setting, which is the closest it has to the board's mode C. The benchmark's own operator manual sets how it is powered and connected, and its declaration of conformity names the standards it was tested against.

Reading the difference​

The benchmark is a different product, often a smaller display with a different backlight. A level difference between the two traces therefore shows where the board is noisy relative to a product that passed. It does not show by how much the board would fail. A write-up states the comparison that way, and never as a margin to a limit line.

A peak that appears on the board and not on the benchmark is the most useful result, because it points to something the benchmark's designers solved and the board has not. A peak that both show at a similar level is usually a source common to the product class, such as a backlight converter, and is a lower priority.