Markers

Serial Send and Receive (Windows)

Sending markers to the EEG can be done using the serial send action.

This action uses an Instance counter between producers and consumers, allowing for multiple simultaneous uses from different timelines on different devices.

The communication is thread safe, meaning it can be used with yields and coroutines.

However, it will result in very poor Measurement Accuracy. As the Windows Scheduler has a 15ms to 8ms (XP, win 7, win8) scheduling delay, depending on your system load, it will have an impact on the trueness standart deviation. Remove unnecessary services and running applications. Change the nice setting of your build to Realtime and you will acheive 90+% fixed trueness.

Marker Send (RTX)

If accuracy with high frequency EEG is needed (ant 10khz or brainproduct), using rtx serial send is a better option.

Sending Markers through RTX needs a Manager running. Use the RTX manager Action to init RTX at the beginning of the experiment and keep it running until the end of your experiment.

The action Marker send is very similar to Serial Send, and uses the same parameters.

Additionnaly,

Delay for : This Enum parameter will delay the sending of the marker by the known (measured) input lag.

Available enums:

    • V-Sync (default)

    • Immediate

    • Monitor1

    • Monitor2

    • HMD1

    • HMD2

    • Eyetracker

    • Audio

    • Glove

    • Extra1

    • Extra2

    • Extra3

    • Extra4

    • Extra5

The List of enums and associated response lags can be configured through the Evrs -> Addon -> RTX panel. Use the BlackBox Toolkit II to measure latencies.

The Enum will then use the corresponding registered parameter.

The total response lag is in milliseconds, and cannot be below 0.00098 s ( 980ns )

Dual marker : toggle range Markers.

In the case of audio, the rotating buffer associated with current running sample size and bitrate will have a very strong effect of latency (from 2-8ms to 10-35ms). Toggling dual marker will send two markers, one at the lowest measured input lag and another at the highest input lag, this will allow for visual aid for interpreting audio input.