You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Jotting down an idea I had after thinking about #172
At the moment the EVR receives timestamps by accumulating 1 events and 0 events into a shift register, followed by a reset event that declares the contents of the shift register to be the timestamp.
We could do something similar for PandA sync.
If we use the top 3 bits of the sequence number as a 1 event, 0 event, and reset event, then we do the same shifting on the receive side then we can send the timestamp. We could also make this future extensible by saying the first 32-bits of the shift register are the timestamp but we can use the rest of the bits (as we can send 1 bit every 48ns there could potentially be 20 million of them) for some future purpose.
The text was updated successfully, but these errors were encountered:
Jotting down an idea I had after thinking about #172
At the moment the EVR receives timestamps by accumulating
1
events and0
events into a shift register, followed by areset
event that declares the contents of the shift register to be the timestamp.We could do something similar for PandA sync.
If we use the top 3 bits of the sequence number as a
1
event,0
event, andreset
event, then we do the same shifting on the receive side then we can send the timestamp. We could also make this future extensible by saying the first 32-bits of the shift register are the timestamp but we can use the rest of the bits (as we can send 1 bit every 48ns there could potentially be 20 million of them) for some future purpose.The text was updated successfully, but these errors were encountered: