yea same here ..
im having now issue but i am getting the same count of sequence errors minus 60 per socket...
Socket s1 Socket s2 Socket s3 Socket s4 Socket s5 Socket s6
Universe Number 1 2 3 4 5 6
Packets Received 677,668 677,668 677,669 677,669 677,671 677,671
Sequence Errors 677,408 677,408 677,409 677,409 677,410 677,410
Barnabybear wrote:If your using the J1sys addin to send to your E68x - if you send 10 packets you will get 9 errors or 1000 packets will get 999 errors. It's a know issue that has no effect on your show. There is a post started by me somewhere - I will look it later.
The issue with the j1sys E1.31 plug-in was (actually "is" - because I have not seen a new one yet) that it would send a sequential sequence number for every packet sent - but it should send a sequential sequence number for each DMX *stream* not just the encapsulated E1.31 packet. This ensures that the receiving device can count on the sequence number for packets destined for it. It was a harmless failure but something that should be fixed so as not to confuse debugging efforts. Not sure how Joe implemented it in his code however.
HLS implements E131 sequence numbers .....
Each E131 Universe tracks its own sequence number which is incremented with each packet sent.
Another possibility is a packet is lost or overwritten in the Ethernet receiving module.
E682 controllers use a Ethernet module that actually receives the packets and presents them to the Prop (main cpu on E682). If the module cannot transfer a packet to the Prop before it is overwritten - then the E682 would see a packet with a skipped sequence number.
Does sequence number dropping occur when the E682 is only receiving 1 universe (low throughput requirement) or when running a show and pushing all 4 or 6 universes (hi throughput).
Show time (all universes being transmitted) is the most stressful on the communication processing module and well as the Prop.
Joe
Had a bunch of sequence errors until I spent a little more time trying to understand. I had to fix my universe numbers and then assign the starting addresses to the cluster correctly. Then it magically worked.
You may get some very strange results if you assign the same universe number to more than one socket.
