Using Eye Finder with the FS2331 DDR Probe
The explanation of the procedure for calibrating the probe for optimal read and write state acquisition provides a description of some useful ways you can interpret Eye Finder results. Eye Finder can be very useful in helping characterize DDR busses. You should keep the following in mind as you interpret Eye Finder results:
1.The results displayed by Eye Finder are not quantitative. They should be used as a way to identify signals that deserve closer examination with other calibrated parametric tools (such as an oscilloscope) or to find the optimal sampling time for state analysis. They should not be used as a definitive measurement of DDR signal timing.
2.The absolute accuracy of the position of the data valid window is not specified or guaranteed. Relative positions are more accurate, but skews will exist in the FS2331/analyzer system that are not calibrated out of the Eye Finder measurement and thus will show up in the results display. These skews could total several hundred picoseconds.
3.The width of data valid windows shown in the Eye Finder results display will be less than what you would measure with an oscilloscope. The difference is about one nanosecond. This is due to additional signal jitter within the analyzer as well as an amount subtracted by the Eye Finder software to provide additional timing margin. The data valid window displayed may be as small as
4.Eye Finder speed is a direct function of the density of analysis clocks in the stimulus used while Eye Finder runs. With frequent clocks Eye Finder will usually complete in less than 15 seconds. If Eye Finder is taking longer than two minutes to complete you can do one or more of the following to help it run faster:
∙Turn off the external CPU cache memory
∙Select the "short" option in the "Eye
∙Use a different stimulus. Video files played in Windows Media player have performed well.
29
