This is usually caused by clocking issues, drive issues or a combination of the two. A few things to check:

On a MADI system:
• Is the framerate set in the Hammerfall DSP app TCO page set to the same framerate as the Renderer and Pro Tools?
• Is wordclock being distributed or daisy chained? Daisy chained WC can be problematic.
• In the Hammerfall DSP app TCO page is the incoming WC frequency showing any excessive jitter.
• Is the Sync HD clocked to the same WC or to Vid Ref?
• Is the drive full? Has it been forced optimized/defragmented?

On PC Systems:
• Check the Windows Control Panel Performance tab - Advanced – to verify it is set optimize “programs” and not “background services”
• Verify that the renderer is the only application running and check task manager to make sure no unnecessary background processes are running and detect PC vs Renderer CPU spikes
• Verify that only the supplied media SSD is being used as the record destination

On a Dante System:
• Is there only one external master set in the Clock page of the Dante Controller with all other devices slave?
• Are the SyncHD and MTRX both getting distributed wordclock?
• Are all the devices in the Clock page of the Dante Controller showing green and stable?
• Is the one external master the MTRX?

On both (and also applicable to running the Dolby Atmos Production Suite):
• Disable re-renders and headphone processing to reduce CPU usage during mastering.

On the Dolby Atmos Production Suite using the Dolby Audio Bridge please reference:

On the DAW:
• What is the source of the timecode? Sync HD or LTC in an audio track? LTC Generator?
• What framerate? We have seen anecdotal issues with 23.98 from Sync HD that have been alleviated by using LTC in an audio track.
• Disable any plug-in automation that can affect automatic delay compensation. This can include automation that varies over-sampling rates.