Unmanaged 1Gbit Switch:
→ grandMA3 console with 2 universes of Art-Net output (Universe 0 & 1)
→ notebook with DMX Workshop
→ RDM10 Port A → Output 1
RDM10 Firmware: 3.0.0
grandMA3 Software Version: 2.3.1.1
Output 1 is a DMX line with 8x ETC SolaFrame Theatre
TL;DR: When I activate RDM globally on the MA3 console, the Art-Net universe of Port 1 changes to a random number and the DMX output STOPS.
Steps:
turn off RDM globally in grandMA3 console
RDM10 factory reset
Load Netron preset 7 (ArtNet 2.X.X.X)
RDM10 display shows:
Input A: ArtNet 1
Input B: ArtNet 1
Port LEDs flashes white when DMX line is connected or when DMX Workshop trigger “Discover RDM”.
RDM10 reports 8 RDM devices to DMX Workshop.
When I activate RDM in MA3 console:
RDM10 Port LEDs flashes white
Fixtures respond with very high latency to DMX changes (multiple seconds)
After some seconds:
LED A Off
LED B flashes White ( = RDM pattern)
all Port LEDs blink blue (= no signal pattern)
No DMX Output from RDM10!
RDM10 display shows:
Input A: CUE (there a no cues recorded after factory reset, right?)
Input B: Artnet1
DMX Workshop and MA3 reporting Universe of Port1 (A) switched to random Art-Net universe like 96 or 37. RDM10 OSD and Website report Port 1: Universe 1
This error can be reproduced whenever I connect the console with activated RDM and Netron preset 7.
I would be very interested in testing any beta firmware in our production environment. For now the RDM10 is not needed for our set-up, but I would like to see this bug fixed and then recommend further Netron purchases.
I could also try to downgrade the RDM10 to firmware 2.X or test again this setup with a grandMA2 console.
Thank you very much for your detailed report of this issue. Sorry to hear you have encountered this.
I have run some tests with an RDM10 running firmware 3.0.0. I don’t currently have an MA3, so have run some tests using ONYX, and ZerOS, which has a fairly aggressive RDM implementation. So far, I have not encountered the issue you are seeing, where RDM discovery appears to cause the RDM10’s universes to change. Therefore, it seems as though the issue is being caused by something specific the MA3 is doing. We will need to perform some tests using this MA3 version to troubleshoot further.
In the meantime, if you would like to roll back your RDM10’s firmware to v2.9.5, this can be done using the “Recovery” button in the Netron Central Utility app for Windows. If you would like instructions on rolling back the firmware, please let me know. As I am currently unable to replicate the issue, I of course cannot guarantee whether v2.9.5 will exhibit different behaviour.
Thank you once again for reporting this. Please let us know if you have any questions.
thanks for the quick reply! It’s great that customer feedback is taken so seriously here!
I suspect there are several issues:
A) The random change in the universe does indeed seem to be a bug in the RDM10, since it only ever affects Port 1. When I use Port 2 in the exact same setup, there’s also a lot of RDM traffic from MA3, but after disabling that in the console, the DMX signal is stable again and the fixtures are controllable.
B) It looks like RDM via Art-Net isn’t quite ready for production in the MA3 console yet. I might take a closer look at the network traffic and also test with version grandMA2.
I tested again with grandMA2 (exactly Mode2 on the same grandMA3 console) and found the same behavior. I used a direct ethernet connection between console and RDM10, no network hardware and again 8 fixtures on the DMX line.
With activated RDM on the console i can see the devices appear and then they disappear again after the Art-Net universes has randomly change on the RDM10.
We have been testing this today with grandMA3 Software Version: 2.3.1.1, and have so far not been able to replicate the issue. It is possible that the issue is being caused by the way MA is handling these specific fixtures. We will therefore need to investigate further.
To aid us in troubleshooting this, please could you make sure that all MA-Net traffic is disabled on the MA’s network interface connected to the RDM10. With MA-Net fully disabled, does the issue still persist?
The MA3 node is always connected to a different network interface then the RDM10. But first things first: I had some fun while testing this and I tried to rule out some potential error causes.
The first setup was without grandMA3 node, without network switches. Only one fixture LED Fresnel connected. That worked as expected and I could even “identify” via RDM. From this known good state I continued testing.
Same setup but with long DMX cables (>80m) and 7 equal fixtures resulted in the universe changed bug. So indeed something is weird with my local setup.
Same showfile with production setup (network switch with VLANs, MA3 node on different network interface, long cables, 6 universes):
One * LED Fresnel conected to RDM10 works well with RDM and DMX.
1 GLP X5 alone does not work. RDM10 has lost its connection a few seconds after turning on RDM on the MA3 console. I can reset via restart of RDM10 or via toggle Art-Net output in the console.
When I connect all 13 fixtures in the DMX line, I see the universe change bug about 4 seconds after RDM is activated in the MA3 console.
So, the implementation in Netron can work with MA3, but not wit certain fixtures? Any idea, why the console floods the Art-Net interface with RDM packets?
Again the changed Art-Net universe is not visible in the OSD nor RDM10 website. Only in the Art-Net menu in the MA3.
Thank you very much for the additional information.
The Obsidian team are at USITT this week, so apologies for our delay in troubleshooting this our side.
With all 13 fixtures connected, please close MA, and then run DMX Workshop. Is DMX Workshop able to reliably discover all of these fixtures connected to the RDM10? Can you reliably Identify them using DMX Workshop? Do you see any strange behaviour with the RDM10’s universe assignment when using DMX Workshop?
I tried it again without the grandMA3 console and I see some strange behaviour. I used only a Laptop with DMX Workshop Version 7.50, a Netgear Switch and the RDM10.
One LED Fresnel Fixtures works fine, I can Identify the fixture and the RDM10 stops the white RDM blinking after a while. Two fixtures also work. If there are more than two fixtures connected via DMX, the RDM detection becomes unstable.
When I connect all 13 fixtures, the RDM10 flashes very long in white and detects not all fixtures in the Webinterface. DMX Workshop detects (mostly, but not always) 13 fixture, but doesn’t display the data for most of of them. About every 10th time I tried, I was able to reproduce the universe change after clicking “Discover RDM”. See attached photo showing the difference between software OSD and DMX Workshop.
Unfortunately, I haven’t had a chance to try RDM with these fixtures directly via DMX (so without the Art-Net layer) yet. But it seems that the devices we use aren’t exactly simple
I have 16 fixtures connected to the RDM10, each with multiple personalities, plus sensors. After multiple presses of “Discover RDM” in DMX Workshop, the RDM10 remains on Art-Net universe 0.
Out of interest, is your DMX line terminated? Do you see any difference in behaviour if this line is terminated?
Next month, I will be attending PLASA “PlugFest”, where manufacturers meet to test RDM interoperability. I will therefore test my RDM10 with as many different fixtures as possible, in an attempt to replicate the issue. The manufacturer of DMX Workshop will also be there, so I will also be able to liaise with them.
RDM10 Preset 1 works fine with RDM. So a basic DMX cable connection is working and MA3 can identify all four fixtures.
RDM10 Preset 7 (Art-Net) triggers the bug described in the first message. After activating RDM in the MA3 software, it takes 2 to 20 seconds for the universe to change. The changed universe is shown in MA3. I can reproduce this process.
Without MA3 console connected and only DMX Workshop, I can use RDM without problems and identify the fixtures, even though there is a slightly bigger latency.
So, in my opinion, I’ve ruled out: specific fixture problem (3 manufacturers tested), DMX termination (no difference), corrupted MA3 show file, faulty RDM implementation (direct DMX cable works).
Anything I can do on my end to help you further?
(I have a short video of the bug, maybe I could send this via DM or e-mail to you?)
