Codemasters TeamHi @Hoobiechoobie ! In the past I asked for a way to properly detect and handle "flashbacks". With the currently available data this is not possible. The solutions we discussed in the past were:
Here the relevant links to the discussion:
Thanks.
@Kafumanto wrote:Hi @Hoobiechoobie ! In the past I asked for a way to properly detect and handle "flashbacks". With the currently available data this is not possible. The solutions we discussed in the past were:
- to make the field "m_frameIdentifier" continuous (i.e. to update it always with increasing values, so that we can use it together with "m_sessionTime" to detect a flashback);
- add a new "continuous" field for this use-case (e.g. a "total elapsed number of frames" field).
I am not sure what m_packetVersion in header is used for as all the values I captured so far come with value "1" only.
Could it be re-used for flashbacks, to differentiate old frame IDs from new timeline?
(bear with me, pls, I am new to UDP capture and process so sorry for talking gibberish)
Codemasters Team
@P4st3l1ak wrote:
I am not sure what m_packetVersion in header is used for as all the values I captured so far come with value "1" only.
Could it be re-used for flashbacks, to differentiate old frame IDs from new timeline?
(bear with me, pls, I am new to UDP capture and process so sorry for talking gibberish)
m_packetVersion was originally intended to be used in cases where we changed the packet structure after launch and app developers could use this value to work out which version of the packet to decode. However, I'm not sure that we've actually used this. Instead, we lock the packet structures at launch and avoid making breaking changes to the data format to ensure that the apps remain compatible during the game's lifecycle. In the case of flashbacks, the new variable will hopefully do what is required.
