Simple Automation vs Motion Lighting Apps

[mik3]

I’m trying to measure the difference between speed and function of both apps.

Clearly the motion lighting app allows for more options, but if both options can be achieved using either or, what should be used?

The main reason I ask, simple automation will constantly send actions to the end devices. There is no state checking. I can see this causing a lot of extra hammering on z-wave devices.

How much of a speed gain is there really? Is it worth throwing out state checking?

[bjcowles]

I guess that would be something the user would need to figure out on their own. With so many variables(restrictions, conditions, device choice, mesh traffic, hub load, etc.) it would be impossible to make a blanket statement. Personally, I perceive little to no difference.
FWIW, I did a little bit of basic testing the other day. Each method was run five times, speed results are an average. Motion reported to ON confirmation.

Iris v2 motion sensor and Sylvania/Osram bulb on Hubitat

Simple Automation - 176ms
Motion Lighting - 207ms
Rule Machine (changed trigger w/ on/off conditionals) - 289ms

Iris v2 motion sensor and Sylvania/Osram bulb on SmartThings

SmartThings->Hubitat(via HubConnect)->SmartThings - 348ms
SmartThings SmartLighting(Classic app, all local) - 191ms
SmartThings Automations(Latest app) - 710ms

[mik3]

I've measuring to, and it appears simple automation works faster. I am assuming because I have two different SA rules. One for day, and one for night. So it's less conditions to check.

I may consider making two ML apps, one for day and one for night. This at least would cut down on conditons, but also give me state checking.

[bjcowles]

My general and unscientific observation- the more "stuff" on the app's settings page, the longer it takes to execute.

[mik3]

Yeah, but I think i'm being a little over board. I'm measuring ms here.

Those extra 100-300 ms is another foot step in the room.

Also, checking this over and over again tends to deliver different results. I think it all depends on what other processes are running on the hub.

At the end of the day, one thing I know for sure, using an external automation like NR or HS delivers faster response times then the hub. But is it really worth all the time, effort, money to gain a couple 100 ms.