Hi TDK/InvenSense Team,
I am an embedded systems developer and a long-time user of various InvenSense IMU products. I am writing to seek an official clarification regarding the definition and hardware architecture of the classic MPU-6050/6500 series, in order to resolve a technical dispute within our local developer community.
The core of the dispute is whether "MPU" is merely a brand trademark or if it signifies a specific architectural difference compared to standard IMUs. One argument suggests that because all IMUs have "digital filtering" (DLPF), the MPU's internal processing is functionally identical to any other IMU.
However, based on the MPU-6050 Product Specification (Rev 3.4), I have observed technical features that seem to distinguish it from a "regular" IMU:
Trademark Definition: Page 52 officially lists "Motion Processing Unitâ„¢" as a registered trademark.
DMP Engine: Section 5.4 describes the internal DMP (Digital Motion Processor) as an engine that supports 3D MotionProcessing algorithms.
Architectural Intent: The documentation states that the DMP is designed to "offload" intensive Motion Processing computation requirements from the system processor.
Could you please provide clarification on the following:
Was the "MPU" name officially chosen to highlight this integrated hardware processing capability to distinguish it from traditional/standard IMUs?
Is the DMP a dedicated hardware engine specifically designed for Sensor Fusion (e.g., Quaternion output), rather than just basic digital low-pass filtering (DLPF)?
I hold this series in high regard as a pioneer in motion tracking. Any clarification would be invaluable for our community.