FPP weird channel m...
 
Notifications
Clear all

FPP weird channel mapping issue

14 Posts
3 Users
0 Reactions
6,815 Views
Posts: 59
Topic starter
(@elnino)
Trusted Member
Joined: 13 years ago
[#38395]

Hi all, I've just spent the better part of a day working this out and i was wondering if anyone had a reasonable explanation for it.

This applied to both the BBB and Pi 1.5 builds of FPP. This is the first time I have tried E1.31 output from FPP - Fresh, unconfigured builds until now.

I had one E1.31 controller hooked up to 80 WS2812b leds. Universe 1 and using unicast direct to the controller.

I set up the E1.31 output as:

Universe 1, start 1, Size 240

But when testing, with a basic FSEQ file or in the 'testing' mode - Nothing! I tried with da_e131.exe and it works fine! I done some sniffing and I finally worked out that there seemed to be an 'offset' of 256 in the channeling of FPP.

i.e I had to change the configuration to be U1, Start 256, Size 496 and then it worked as expected. I checked over the configurations and I can not see channels 1-255 in use anywhere but I suspect that is what is happening - Can someone explain?

Weirder is that the 'size' seems to represent 'end' as it is exactly 240 more than 256. I've also noticed that FPP seems to round up to the nearest multiple of 3. If i set a channel count of 493, it drops the last pixel, a value of 494 and the last pixel works fine where it should only go red.


13 Replies
Posts: 843
(@captainmurdoch)
Prominent Member
Joined: 14 years ago

It sounds like your sequence file conversion was broken. What sequencer are you using? Did you use xLights to convert to .fseq?

If you had to set a start of 256 on universe 1 then it means that the data in your .fseq doesn't start until channel 256.

In your sequencing software, do you have any other channel outputs defined?


Reply
dpitts
Posts: 1020
(@dpitts)
Noble Member
Joined: 16 years ago

Please post screen shots of Sequencer output setup, controller setup and FPP E131 output setup. There is not a default offset of 256. First report of this.


Reply
Posts: 59
Topic starter
(@elnino)
Trusted Member
Joined: 13 years ago

I have not even got as far as getting a fseq file working. This is just in the display testing section. Or does it somehow read the config from an uploaded FSEQ file for base config?

I added another controller:
30 Pixels attached as universe 2, unicast as the other controller is.

To get it to work:
Start 752, size 90 and it maps fine.

Truly odd behavior.

Edit - On the BBB, i do not have any fseq files uploaded at all.

Edit 2 - Even weirder. So I kept decreasing the Start and size equally. I managed to get to a start of 15 and a size of 255 for the first controller (240 actual outputs needed). Anything lower (on either) and it stops working completely. It looks like one of the values is being misused in the background somewhere.

Then, for the second controller, I could start at 270 and go through to 359 as expected. it seems like the first 'size' field is being treated as 'end' but not subsequent ones.


Reply
dpitts
Posts: 1020
(@dpitts)
Noble Member
Joined: 16 years ago

The screen shots of controller and FPP setup can help us figure out what may be the problem.


Reply
Posts: 59
Topic starter
(@elnino)
Trusted Member
Joined: 13 years ago

Ok, the controllers are my own Arduino ones. They are 100% working with da_e131 as per the screen shots. I have changed the subnet since the original post (trying a few things). I just now done a test with da_e131 with the 2 screenshots attached and can chase all leds off both controllers perfectly.

weird1

weird2

However, in FPP, I have to use this:

weird

Reply
dpitts
Posts: 1020
(@dpitts)
Noble Member
Joined: 16 years ago

You are aware that FPP uses absolute start channels. That means the channels keep incrementing, they do not start back at 1 for each universe. So I would think this would be your FPP setup to make it match the setup in da_e131.

Universe 4 , Start Channel 1, size 240,
Universe 5 , Start Channel 241, Size 90

So turning on channel 1 and 241 would light first light on each controller respectively.


Reply
Posts: 59
Topic starter
(@elnino)
Trusted Member
Joined: 13 years ago

Excactly! But it does not work!

With this:

2015 10 18 20 03 18

The second controller works fine. The first one does not work at all.


Reply
Posts: 59
Topic starter
(@elnino)
Trusted Member
Joined: 13 years ago

A little progress - I changed the config to start at channel one, then as far as i can get is up to size = 126 before it wiggs out and does not work at all. Possibly, the other one works purely because it only has 30 LEDs.


Reply
dpitts
Posts: 1020
(@dpitts)
Noble Member
Joined: 16 years ago

Well even though the E131 module has not changed in 2 years and no one else has mentioned this problem I thought I would look deeper. Weirder things have happened. So I set up my PI to output to my computer with same E131 universe setup.

Fpp E131

I then went to test and output value 119 (0x77) to channel one.

Fpp test

Next I captured universe data with Wireshark. You can clearly see the universe number "4" and the "0x77" in the first channel at offset 126 which is the beginning of E131 data.

Wireshark of FPP

Reply
Page 1 / 2
Share: