-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On 2012-02-16 23:01, Martin Peach wrote:
The [mrpeach/tcp*] classes don't make any assumptions about content, they just output lists of floats as they arrive. It seems more efficient to do that than to output individual floats. Whatever is handling the output of a [tcp*] should be able to decide for itself where the packet boundaries are. I don't assume that a list is necessarily the same length as a packet.
i don't think [mrpeach/tcp*] is doing anything wrong here. on the low-level side, you get buffers of data from the socket. [mrpeach/tcp*] passes these buffers into user-land as they are. great!
theer's a reason why [iemnet/tcp*] does it slightly differently (that is: serialize the data):
- explicitely break the false sense of security of the user, that tcp/ip
could be packet oriented; because from user experience (mainly roman's netpd) it shows that you must take re-fragmentation into account if you want to make a stable system. if you receive the stream as multi-byte chunks the user therefore ought to serialize it manually.
- for performance reasons, it is way more efficient to do the
serialization in C than in Pd.
]btw, you can turn off automatic serialization with [iemnet/tcp*", but i would not advise to do so.
fgjmasdr IOhannes