Skip to content

Log: Turbo-charging Traces (Binary REAL)

Commit: 95ccc87

Overview

Parsing 10,001 floats from an ASCII string is just... painful. It's slow, it's CPU-intensive, and it feels like 1995. We finally moved all major drivers (Keysight, R&S, Anritsu, Rigol) to native binary REAL data transfer.

Technical Lowdown

  • PyVISA Magic: We're now using query_binary_values() instead of query_ascii().
  • Command Overhaul: Standardized commands like :FORM REAL,32 and :FORM:TRAC:DATA REAL across different brands.
  • Speed Win: Data hits your Python script as a ready-to-go list/numpy array without the "comma-split-float-convert" dance.

Why it matters

In high-throughput ATE, every millisecond counts. Moving to binary is the single biggest performance win for SA and VNA trace extraction we've had yet.