Iāve got mine all set for 1 second too. Thatās anywhere from a single light to 27 if I have them all in party mode. Party lights need to flash quickly and 1 second does this reasonably well.
One thing you will need to be careful of if you set a really short loop time (ie 1 second). As I mentioned above, Iāve had mine set to 1 second since I first set it up but I have really only had the lights on for relatively short periods of time and all was good. However, I had some of them on for a long time last night and the automation crashed as I exceeded 10,000 loops. Iām not sure if there is somewhere to increase that limit. If not, then if you have a 1 second delay, you can really only run it for 10,000 seconds which is about 2 3/4 hours. Increase the delay to 2 seconds and you will get 5 1/2 hours. The problem is, at 2 seconds and greater, the lights arenāt really āflashingā. They are just changing colour slowly.
Not sure it is the same problem. Cause it seems to happen every 2-3 minutes.
Iām trying to tweak a little bit the automation in order to see if it improves.
To have flashing lights that are more similar to a disco Iām using the following parameters. My guess is that time_between_changes is somehow too low, and at some point it overloads the requests the light can support. Initially I had 0, now I put 0.7 which seems to be a good compromise.
I think it makes sense that the light itself was getting messed up since the transition time is set to 0.4 seconds, but you had the time between changes set to 0.
This would mean that the transition is still happening when a new change is pushed. My guess us that the light buffers the new incoming requests and only executes them once the transition is done, thus loading up the buffer and then stalling.
There is a warning message at the transition time parameter about this in order to prevent it, but there is no way of coding a hard stop in the blueprint to prevent users from actually setting it up incorrectly.
Iāve upgraded to the new version and it almost works as needed. It does indeed grab a snapshot of how the lights were and returns to it when the trigger is turned off. However, if the lights were off to begin with, it returns to the off state but when you then turn the lights on again they will be in the final colours before the lights were turned off. If the lights are on before the trigger occurs then it works perfectly.
I think what is also needed is ādefault light settingsā that are returned to if the lights were off to begin with. I have implemented this myself with an āOffā automation that sets the lights back to white etc and then turns them off. The code is below.
alias: Party Lights - TV Room - Off
description: āā
trigger:
platform: state
entity_id:
input_boolean.party_lights_tv_room
from: āonā
to: āoffā
condition:
action:
Some smart bulbs support effects like āblinkā, āfadeā, ācolorloopā, etc ā¦
You can now configure these, but support depends on the specific bulb.
This behavior is part of how your lights themselves are configured.
Most smart bulbs will be set to turn on in the last state they were in. See if you can set there āturn onā behavior in the lightās settings.
While I agree in general that this is a good idea, the way you implement it is not very flexible as it would apply all the same settings to the lights in the blueprint.
I will look into giving users the option to choose to return the lights to their previous state or activating a scene. That would allow different settings for each light and even include other entities.
Iāve added the option to configure effects for the lights (your specific lights will need to support it though.) Maybe that will work for you.
Itās pretty easy to make an automation for specifically that though, just loop through blue and red colors with a short delay: