Been going over this Ad Nauseum for a week.
FPP to e682 single controller for a megatree.
FPP appears to be working now and its been set aside as it seems the issue is with the e682.
While I troubleshoot, the e682 is being tested using a direct connection via Vixen3(no fancy or additional setup to configuration here)
Layout:
1-14 e682 connections in use for the tree
15th connection in use for the tree topper
16th connection not in use.
each connection controls 1 strand separated into 2 strings with 32 pixel nodes each. This creates the 28 string, 32 node tree
e682 configured for each plug as 64 nodes, ZigZaged after 32nd node
-tests show a correct display setup using on-board test patterns.
Now the issue.
String2 copies String1 in every aspect.
example; a slow chase lighting each node individually to confirm each pixel works and is mapped and configured correctly will result in a chase where String1, pixel1 AND String2, pixel1 will light both strings will slowly chase upwards together.
it displays as node1 & 33, node2 & 34, node3 & 35, etc. This happens for each strand on each plug.
if I change Vixen to chase only String2 then nothing shows.
Checking all 2688 channel outputs show none are duplicated or slaved.
I have replaced the e682 3 times now so it cannot be a bad board or bad flash of the eeprom. It's somewhere in how I have set up the zigzag.
Suggestions?
If they are working as expected with the e628 only , could it be how you have the strings/element patched in Vixen ?
angus40 wrote:If they are working as expected with the e628 only , could it be how you have the strings/element patched in Vixen ?
with the current test patterns in the e682 firmware there is no way to confirm individual node/pixel function as tests are en-mass across all connections.
So far all my tests have shown that selecting ZigZag at node32 of a 64node string results in both halves of the string acting like a cloned pair.
String1 is cloned on String2 and they both respond the exact same.
Aurbo99 wrote:with the current test patterns in the e682 firmware there is no way to confirm individual node/pixel function as tests are en-mass across all connections.
I don't think that is entirely true. I'm running firmware ver 4.081 though maybe something has changed if you're running different vers. I just ran a test on my controller.
I took a string of 50 nodes and connected to port 1-1 set for 50 nodes, no zig-zag.
Tests 4, 5 & 6 turn on all the nodes dim (red, green or blue) and then brightens each node in channel order.
With no zig-zag, node 1 flashes bright, then 2, all the way through to node 50.
If I set zig-zag to 25, the string flashes node 1, 2... to 25, then node 50, 49 down to 26.
This is a way you can verify the e682 is set correctly.
*note: this test goes through ALL the nodes starting at 1 in port 1-1. If you only have strings connected to ports in 3-1 for example, you have to wait for the flash to get through all of ports 1 and 2 first. It might look like it isn't doing anything, but just wait.
Cmndr Brain wrote:I don't think that is entirely true. I'm running firmware ver 4.081 though maybe something has changed if you're running different vers. I just ran a test on my controller.
I took a string of 50 nodes and connected to port 1-1 set for 50 nodes, no zig-zag.
Tests 4, 5 & 6 turn on all the nodes dim (red, green or blue) and then brightens each node in channel order.
With no zig-zag, node 1 flashes bright, then 2, all the way through to node 50.
If I set zig-zag to 25, the string flashes node 1, 2... to 25, then node 50, 49 down to 26.
This is a way you can verify the e682 is set correctly.
*note: this test goes through ALL the nodes starting at 1 in port 1-1. If you only have strings connected to ports in 3-1 for example, you have to wait for the flash to get through all of ports 1 and 2 first. It might look like it isn't doing anything, but just wait.
I'm using firmware 5.xxx beta
I removed the zigzag configuration on the e682, used Vixen to reverse map the 33-64 nodes on each string, this appear to work
