This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

llpm work flow

hi support team,

in ncs llpm example, we tested it and find out one problem, why 4bytes of timer data would spend more than 1ms? please see picture 1.

further more,

1. bt_latency_request would only be send out for 2ms interval, is it correct?

2. if i change data to 20 bytes, it would be more than 1500us, why it costs so much time. please see picture 2.

     the reason for this question is only 16 bytes added, only 66us should be added based on 2M phy.

3. for write response, i think this is only ack from slave, because callback handler param is from write param, please see picture3

Regards,

William.

Parents Reply
  • I can see it's not entirely consistent:

    So, sometimes the application is marginally fast enough to handle the write request before next connection event, and sometimes it is not. Looking at the delta time is says 745-746us, so if the application is not able to handle the write request fast enough, then the write response packet will be delayed with one connection event (+1ms in this case). I assume the longer packet the write request is, the less is the delta time, and the application will have even shorter time of handling the write request. In general write requests are slow, it is faster to use write command or just send notifications.

    Kenneth

Children
No Data
Related