The format is quite simple. Each file has from 1 to 16 cahnnels in it (channel corresponds to an ADC input). In the process of recording the ADC converts voltages at analog inputs into 16-bit numbers (actual ADC resolution is 12-bit, meaning that the upper 4 bits are not used). These numbers are retrieved from the ADC by the data acquisition program and are stored on the disk in the order of the channel (input) number: i.e 16-bit word for the channel 1 is followed by a 16-bit word for the channel 2, etc. The process of sampling of input channels is repeated at the ADC sampling rate (typically from 10 to 30 kHz).
File layout for 3 samples of 4 channels looks like:
| Times | 1 | 2 | 3 | |||||||||||||||||||||
| Channels | 1 | 2 | 3 | 4 | 1 | 2 | 3 | 4 | 1 | 2 | 3 | 4 | ||||||||||||
| ADC Words | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | ||||||||||||
| Bytes | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 |
here bytes 13 and 14 represent a 16-bit sample of voltage at channel 3 taken at time 2, the size of the file is 24 bytes.
There is no information in the file on the number of channels recorded, or about the sampling rate or about the stimulus. All this information is stored in the main database and it can be retieved by the filename. Which means that renaming a datafile is definitely a bad idea.