tmorse305 Posted August 14, 2019 Posted August 14, 2019 I've notice recently that there is a ~15 second delay between an ISY variable changing and the Alexa voice response. It use to be just a few seconds at most. I'm aware of the 30 second debounce specification but I'm waiting minutes between attempts and it's still the same 15 second delay. Is anyone else seeing this? Did Amazon make a change? Thanks, Tim
larryllix Posted August 14, 2019 Posted August 14, 2019 Careful with the Routines. I have noticed Amazon likes to slip a programmed delay in any Off detection Routines every time you open them for editing. I have been caught a few times on that Gottcha' 1
tmorse305 Posted August 14, 2019 Author Posted August 14, 2019 Thanks larryllix, The routine that is delayed is a Device/Motion Sensor category in Alexa. The delay is on detection, I didn't see any delays, but I also didn't see a way to add a delay. If you wanted a delay is it added as an Action, similar to adding a wait in an ISY program? Tim
larryllix Posted August 14, 2019 Posted August 14, 2019 The delay is found in the Alexa App on a mobile device only, under routines, in the particular "When" section if you select "Not Detected". For Alexa perceived "motion detectors", it states "When xxx detects no motion for ZZ minutes" in very small font.
tmorse305 Posted August 14, 2019 Author Posted August 14, 2019 Got it, thanks. That isn't related to the problem I'm seeing. Are you seeing a delay in your Alexa responses to ISY variable changes on the "detect" side of the actions?
larryllix Posted August 15, 2019 Posted August 15, 2019 11 hours ago, tmorse305 said: Got it, thanks. That isn't related to the problem I'm seeing. Are you seeing a delay in your Alexa responses to ISY variable changes on the "detect" side of the actions? Not that I have noticed. Alexa delays do come and go at times. My vocals announcements re hard to tell as I only announce my fryer is done and two garage doors are open or both closed. Not much time comparison there to be noticed.
tmorse305 Posted December 13, 2019 Author Posted December 13, 2019 I finally figured out the cause of the delay so I thought I would share it in case anyone else is experiencing this problem. I am using a state variable configured as a Alexa Sensor in the Portal. I have the sensor trigger value set to 2. I am using the same variable for program control so there is a brief moment when the programs are running that the value is set to 1. It would appear that when the value changes from 0 to 1, the Portal communicates to Alexa that there is No Motion. Then when the value changes to 2 the Portal communicates that there is motion. The problem is that Amazon imposes a 30 sec debounce rule, so the first communication triggers the debounce and delays the spoken message. I'm not sure if this is intentional behavior or a bug but heads up to anyone that is using variables in this way.
larryllix Posted December 13, 2019 Posted December 13, 2019 I have found MS simulation via ISY Portal to not be reliable to trigger routines.AFAICT so far, the first trigger works almost immediately but further attempts appear to be locked out at some level of the process.Sent using Tapatalk
simplextech Posted December 13, 2019 Posted December 13, 2019 25 minutes ago, larryllix said: I have found MS simulation via ISY Portal to not be reliable to trigger routines. AFAICT so far, the first trigger works almost immediately but further attempts appear to be locked out at some level of the process. Sent using Tapatalk There's a rate limit imposed by Amazon on sensor triggers. I think it's 30 seconds in between. 1
tmorse305 Posted December 13, 2019 Author Posted December 13, 2019 I wonder if your routines are being effected by the 30 second debounce. The other issue is that if you are in the middle of a debounce delay and the trigger condition goes away, then the routine will never run. 1
larryllix Posted December 13, 2019 Posted December 13, 2019 Ahhhhhh.. so you mean we should hold a logic 1 for more than 30 seconds before asserting a logic 0?That could explain some of random failures I don't hear from my Echoes.Sent using Tapatalk
tmorse305 Posted December 13, 2019 Author Posted December 13, 2019 1 hour ago, larryllix said: Ahhhhhh.. so you mean we should hold a logic 1 for more than 30 seconds before asserting a logic 0? That could explain some of random failures I don't hear from my Echoes. Sent using Tapatalk Yes, otherwise there is a change that the routine won't trigger.
Recommended Posts