We've received a few recent reports of export speed being slow when exporting raw data to a csv file. Unfortunately, exporting to a csv file has its limits, and the time it takes to process the data depends on the following:
- Number of digital transitions in your data
- Number of analog samples in your data
- Your PC specifications
We might be able to squeeze out a bit more performance in terms of speed, though due to the csv file format itself, the speed increase may be in the order of a few 10s of percent. We wouldn't be able to drastically increase the speed by 10x.
Besides the csv format being quite inefficient, one area that might be causing a bottleneck in speed is the target location for the export file. For example, saving to a network drive or an external usb drive will cause limitations in export speed.
We plan to improve the performance of our software, but in the meantime, here are some workarounds.
- Export only a section of the data
- Export only a few channels
- Export to binary rather than CSV. We also have a sample python script to read binary data (provided in the link below, at the bottom of the article).
{% content-ref url="../faq/technical-faq/binary-export-format-logic-2.md" %} binary-export-format-logic-2.md {% endcontent-ref %}
Since export performance is related to how many digital transitions occur in your data, you can check this by using our built-in Clock Stats measurement extension. Install the extension (it should be installed by default) and SHIFT+drag across your entire range of digital data to count the number of rising edges and falling edges in your capture.
An alternative approach (much faster and produces a smaller file) would be to export to a binary file. More details on the binary file format can be found in the link below.
{% content-ref url="../faq/technical-faq/binary-export-format-logic-2.md" %} binary-export-format-logic-2.md {% endcontent-ref %}
