Logic 2 - Ideas and Feature Requests

t = 0 should always refer to the trigger event, and not the first sample saved. In other words, the sample time should be allowed to be negative in cases where the sample written to file occurred before the trigger event.
The GUI shows negative time for samples occurring before the trigger, but the data file starts the data at t = 0.... (not coherent)

In fact, the trigger time could also be placed in the binary file (following is an example of a proposed header file):

Header:
char[8] id;
int32 version;
int32 type;
double begin_time;
double end_time;
double trigger_time;
double sample_freq;
int channels_in_file;
For each channel in file:
char[16] channel_name; // Provide up to 16 bytes for signal name for each channel in file
int32 channel_index; // Bit location of channel data in samples
int32 initial_state

Then for the data:
For each sample, is there is only 1 signal in file:
double sample_time; // indicates signal toggle time
For each sample (multiple signals in file)
double sample_time; // indicates signal toggle time
byte|short|int channel_data; // Size dependent on # of channels in file (8, 16, 32(?))