How does NS handle RS232 feedback?

mpullin

I had a hard time coming up with an accurate title for this thread.

Here's the thing: I'm working with a JVC DLA-RS2 front projector. I can send commands to it, it executes them, like a projector should. The manual is kind of poorly worded but I was able to figure out what to send it to turn it on and off. Cable is correct, baud rate is correct. Yet I very rarely see any information back from that port. Occasionally it will give me an ACK (based on the manual, the ACK is structured correctly, starting with $06) but other than that, nothing. It gives me an ACK maybe once every 25 times I send it a command.

I called JVC and told them I could control it but didn't always get an ACK.
The tech asked me: "Are you using Crestron?"
I said, "No, AMX"
He said, "Here's what you want to do. Go to Google.com, download LookRS232, and connect directly to the back of the projector. If you get responses back, you know there's something wrong with your AMX. If you don't get responses back, then let me know. Although I don't think that's the case."
I said, "Ok I'll get back to you on that"...

...as I was thinking, what a cop-out answer. Assuming I'm not correctly receiving strings because I'm using AMX and not Crestron. Countless times I have been able to correctly read RS232 from a unit with NS, why can't they just trust that I'm getting back what I say I am? Besides, directions from Tech Support of any kind should never involve the phrase "google dot com."

Not sure why, but I installed this crappy program and learned how to use it. To my astonishment, the projector gave me a feedback to every single command. Granted, the feedback given was the same depending on the command that was sent, and was not at all connected to the state of the projector. So the feedback was practically useless to me.

But this experience got me wondering: why wasn't I able to see the responses through NS? Does AMX really handle data coming into an RS232 port slightly differently than the other guys? I would have called the guy back and asked about this, but I already got the information that I wanted and JVC's phone wait times are rather long. So I was wondering... is anyone here able to explain this?

Joe Hebert

How are you monitoring the feedback with NS2? Are you monitoring with SEND_STRING 0s via the Diagnostics window, or monitoring the data via Notifications, or monitoring by watching a buffer variable in debug?

I?ve never experienced the issues that you?ve run into unless I try to do a SEND_STRING 0 that has a NULL in the message. If it does then the Diagnostics window has a brain cramp and drops everything after the NULL. This NULL problem is documented as being fixed in NS2 version 2.6 but it?s not.

The only other thing I can think of that will cause a feedback error is if you?re using a premade cable instead of just pins 2.3, and 5. If the JVC is putting voltage out on pins 4 or 6 then that can cause issues with an AMX comm port since an AMX comm port supports both RS-232 port and RS-485 unlike a typical PC comm port. I have no idea how the dark side handles comms and to be brutally honest I could not care less.

mpullin

I'm looking at Notifications. I checked the cable just now and although I thought we were using a standard cross, I found it was a premade cable. I think the reason it was in there is because a female/female cable is called for and we didn't have any of those custom made. Probably a cable with only pins 2,3,5 would work more consistently.

Thanks for getting me to think about that. It doesn't matter anyway since the feedback for this device sucks, but it helps restore my faith in NS Studio.

vining

I've had a similar problem with an Antex XM piece and a Panasonic 65" Plasma where I could always control it but every once and a while I would stop receiving strings from the device and sure enough once every pin other than 2,3, 5 on the AMX side was snipped the problem went away.