Hi, Just a general question, .Net 3.5 has lots of enhancements on secure and transacted webservices ? Does anyone here have experience of this ? does it add overhead ? Can these webservices (in a LAN environment and assumed stateless / lightweight) be relied upon to be used in realtime (by realtime I mean response of 1 second or less ?). I realise that this cannot be guaranteed as it depends heavilly on what the service is doing and what network it is deployed upon etc etc etc, but assuming there are no issues here, what I am simply asking is, is the benefit of 'knowing' your webservice will return impaired by any overhead that prevent such a webservice from being generally reliable enough to be used in a real time application ?

If not, are there any other recommened architectures / methodologies I can take a look at ? (remoting for example) or should I simply stick with a tcp socket and build-my-own ??

Thanks

Chub.