I think one of the biggest problems is that I am now retro-fitting MP to my program, although I have replaced all template code and everything is written to be very modular.
Still that said, its difficult to get everything setup right.
Your reply is very useful, I appreciate the time you invested into this post. Thanks fastlane.
The comment you make about the term 'pointer' is very important, and I think this is mostly where I am falling down. Simply I am not sure what the SEND_*_TO function is expecting. I have taken a look at the latest AUM workshop, which has helped.
Still I have managed to get 'locally' several clients to connect, and for the server to update the positions of individual entities based on the local camera position from each client.
The server is able to seperate each client, based on the ID variable being passed from the server to the client, the wing and a prayer element comes here I think.
I am not getting a true handshake between client and server, and instead I am relying on operations taking a maximum of 2 seconds to complete, and then pushing ahead, presuming the information has been exchanged correctly.
So locally, with faster response times, this is maybe not an issue, however over the net 2 seconds might just not be enough time. More's the point, this is really bad code and both server and client should wait for each other until the exchange is complete, or indeed failed.
This is something I really need to re-write immediatly.
Your bang on about diagramming the code, laying out the exchange process is something I intend to invest considerable time in. This is something we are very thorough on at work and often the engineers are in meeting rooms re-drawing these connection diagrams.
Still, with my entire toolset to migrate to a client/server environment, my first step was simply to establish connection and simple avatar. This will allow the team to continue working with the tool and indeed share/discuss their work in a live environment.
With this in place I will then need to rip this baby apart and build her back up again
Not the first or last time I will need to do this! lol - ripping out all the templates was fun
It sounds like you have already tackled many of the tough challenges ahead, for example, we will def want to include mySQL, we already have our servers in place and have committed to this route. (TBH Much based on your own posts about this over the past few years, as well as common trends and most importantly, accessibility.)
It's a great relief to know that someone has already walked this path and barring, d_play issues ;), I believe that we can achieve our ultimate goal.
Thanks for all your help and support fastlane.
