All-in-One Scene: How to do it Better

[cag014 731]

7 hours ago, Rover said:

Scenario:

An AOQ trueAct line has to be triggered by 2 remote controllers of type sceneActivation (WALLC-S) and by 2 Fibaro Keyfobs and by one of type centralSceneSupport (Remotec ZRC90) and by your nice AOQ_REMOTE QA.

How to implement this without copying the (long) trueAct line?

Using a HC3 scene to toggle a global variable is one solution, but maybe you know a more elegant way.

The best option is:

1. Create jM line with all necessary trueAct

2. Use runAction in other lines to execute trueAct on specific line

In this way you need to maintain only one trueAct table in the future.

For example:

jM={

{0,"`keyFob1`",{state="sceneActivation=5",trueAct={{....},{....},{...}}}},

{0,"`keyFob2`",{state="sceneActivation=5",trueAct={"`keyFob1`>1","runAction","trueAct"}}},

{0,"`WALLC`",{state="sceneActivation=5",trueAct={"`keyFob1`>1","runAction","trueAct"}}},

}

Please pay attention to device in trueAct (marked bold), it has ID and jM line number to indicate exactly the line.

I strongly recommend to place lines which will be used for runAction as a first lines in jM, so if in the future you'll add/delete lines, those lines will stay at the same place.

Please see runAction command's options in user's guide. In general you can run any table and specific command(s) in the table.

Edited September 18, 2020 by cag014

[Rover 47]

Please login or register to see this spoiler.

What does "in set" means?

[cag014 731]

On 9/17/2020 at 4:07 PM, Rover said:

Please login or register to see this spoiler.

May I suggest to use built in verification option:

For example:

{0,"`DeurenRamenStatus`",{state="value=false",trueAct={"all","armAlarm!{$Beneden:value=armed|$Boven:value=armed}",""}}},

The verification foramt is exactly the same as in conditions.

Now, AOQ will verify the states several times within ~7 seconds  and will send the orginal command again after 2-3 seconds.

If verification failed errAct{} executed, so please add

errAct={"Arming error","sendEmail,2","Niet alles armed"}.

If errAct{} not defined error will be shown.

The line with errAct

{0,"`DeurenRamenStatus`",{state="value=false",trueAct={"all","armAlarm!{$Beneden:value=armed|$Boven:value=armed}",""}, errAct={"Arming error","sendEmail,2","Niet alles armed"}}},

 

errAct{} table accepts all standard commands.

 

BTW: when verification failed the status of failed condition colored red, so you can see what failed.

Edited September 18, 2020 by cag014

[cag014 731]

1 hour ago, Rover said:

Please login or register to see this spoiler.

Oops.. sorry... my debug line.

[cag014 731]

@gggizmo

Attached version has no delays at all on primary HC3. (tested and verified on mine system)

On slaves maximum possible delay is less than 700 milliseconds.

It works excellent on new HC3 version without any extra load on CPU.

If you have time, please try it and see what happens to CPU load, some kind of benchmark between the versions.

 

 

Please login or register to see this attachment.

Edited September 18, 2020 by cag014

[gggizmo 6]

7 hours ago, cag014 said:

@gggizmo

Attached version has no delays at all on primary HC3. (tested and verified on mine system)

On slaves maximum possible delay is less than 700 milliseconds.

It works excellent on new HC3 version without any extra load on CPU.

If you have time, please try it and see what happens to CPU load, some kind of benchmark between the versions.

 

 

Please login or register to see this attachment.

Sorry to disappoint you. Results from my tests are:

increased delay on triggered zwave devices actions 2 to 6sec, and same cpu load compared to 2.3 or 2.4.

 

In mixed configuration there's a little more cpu load but may be it's due to specific high activities in during the test timeslot

[gggizmo 6]

9 hours ago, cag014 said:

@gggizmo

Attached version has no delays at all on primary HC3. (tested and verified on mine system)

On slaves maximum possible delay is less than 700 milliseconds.

It works excellent on new HC3 version without any extra load on CPU.

If you have time, please try it and see what happens to CPU load, some kind of benchmark between the versions.

 

 

Please login or register to see this attachment.

Just a question: do you use home remote? Do you force screen on all the time on homeremote devices?

[Rover 47]

Please login or register to see this spoiler.

Sometimes the devicename has been shown ?, sometimes the deviceID...

[Rover 47]

@cag014

In the Alarm Zone editing screen I only see the HC2 devices that are part of HC2Proxy, but no <device>'hc2 devices.

Do I need the Alarm Zone concept of HC3 with AOQ or is making use of device lists quite as good?

Please login or register to see this spoiler.

 

[cag014 731]

6 hours ago, gggizmo said:

Just a question: do you use home remote? Do you force screen on all the time on homeremote devices?

What do you mean by home remote? Like tablets?

In general I have  2 android tablets constantly connected to HC2 (screen is off and turned on when motion detected in the room)

4 keyFobs (two for HC2 and two for HC3)

1 Fibaro swipe for HC2

1 Zwave.me wall controller for HC2

 

But I believe this is not exactly answers to your question!

[gggizmo 6]

2 minutes ago, cag014 said:

What do you mean by home remote? Like tablets?

In general I have  2 android tablets constantly connected to HC2 (screen is off and turned on when motion detected in the room)

4 keyFobs (two for HC2 and two for HC3)

1 Fibaro swipe for HC2

1 Zwave.me wall controller for HC2

 

But I believe this is not exactly answers to your question!

I don't know why but i was sure you mentioned tablet with homeremote app ( in the past i used imperihome..rip). And i think that some load on hc3 is due to constant refresh activities from tablets with homeremote app.

So we have very similiar implementation: 4 tablet waked by motion, 4 zvawe.me wall controller, 2 zwave.me keyfob...no swipe.

 

Today i tried to swap hc3 ( moving actual configuration from old to new one) to test if i owe a defective unit...no way same behavior, so i think that my home is hc2 friendly...but hates the new fibaro hc3:-)

[cag014 731]

34 minutes ago, gggizmo said:

I don't know why but i was sure you mentioned tablet with homeremote app ( in the past i used imperihome..rip). And i think that some load on hc3 is due to constant refresh activities from tablets with homeremote app.

So we have very similiar implementation: 4 tablet waked by motion, 4 zvawe.me wall controller, 2 zwave.me keyfob...no swipe.

 

Today i tried to swap hc3 ( moving actual configuration from old to new one) to test if i owe a defective unit...no way same behavior, so i think that my home is hc2 friendly...but hates the new fibaro hc3:-)

I really don't understand how come you have delays on HC3?

I'm monitoring delay for last 24 hours and don't see any delays on Z-wave, although there are delays on UI performances.

[cag014 731]

4 hours ago, Rover said:

@cag014

In the Alarm Zone editing screen I only see the HC2 devices that are part of HC2Proxy, but no <device>'hc2 devices.

Do I need the Alarm Zone concept of HC3 with AOQ or is making use of device lists quite as good?

Please login or register to see this spoiler.

Since you're using jgab's HC2Proxy you should have devices on your HC3 which emulate HC2 devices, correct? That's what I understand by briefly look on the code.

So just use this devices in AOQ as HC3 devices and you're set.

Question: Is the HC2 emulated devices on HC3 works w/o delays. I mean when sensor breached how fast you see that on HC3?

By the way I have found HC2Proxy to complicates just for emulation and you can emulate HC2 device in different way on HC3, but this is another story...

[Rover 47]

15 minutes ago, cag014 said:

Since you're using jgab's HC2Proxy you should have devices on your HC3 which emulate HC2 devices, correct? That's what I understand by briefly look on the code.

So just use this devices in AOQ as HC3 devices and you're set.

Question: Is the HC2 emulated devices on HC3 works w/o delays. I mean when sensor breached how fast you see that on HC3?

By the way I have found HC2Proxy to complicates just for emulation and you can emulate HC2 device in different way on HC3, but this is another story...

The whole operation to program all HC2 code in AOQ is to migrate everything from HC2 to HC3.

While all functionality should be keeping alive with this operation I do not migrate all devices from HC2 to HC3 before all functionality is working with AOQ.

Therefor HC2 devices should be operated from AOQ.

I am very happy with the recent possibilities of <device>'hc2 and direct use of remote controllers and even a iOS QA in AOQ.

HC2Proxy is a mapping of HC2 devices to HC3 devices with new device ID's for the HC2 devices. I have used it for a number of devices of different type in order to realize the construction in AOQ, but to map all my devices is not the intention, while it is a lot of work and with every change in a child device the device ID changes too.

So I hoped with the possibility of using remote controllers in AOQ to fade out HC2Proxy, in order to diminish complexity of having two device ID's for the same HC2 devices.

Now I am in the phase of implementing alarm scenario's and see that using Alarm Zone editing <device>'hc2 devices are not recognized. For real operation all HC2 devices should make part of the Alarm Zones.

Or, I ask the same question: Do I need the Alarm Zone concept of HC3 with AOQ or is making use of device lists (and not using Alarm Zones) quite as good?

Or is it possible to make Alarm Zone usable with <device>'hc2 devices?

Answers on this questions is a prerequisite to be able to continue developing further with AOQ.

 

Your question: Is the HC2 emulated devices on HC3 works w/o delays. I mean when sensor breached how fast you see that on HC3?

I see no possibility of measuring that now, because I only use <device>'hc2 devices, except for a few sensor devices in the Alarm Zone construct.

But in the period I have used HC2Proxy devices for steering lights with a motion sensor no too long delays could be noticed.

[cag014 731]

@Rover

I do understand what is your goal, but I think this is exactly the same procedure that you'll do with other devices, changing the IDs in jS2n.

You can use HC2Proxy only for sensors you need in alarm zones and define them as part of HC3 devices.

After migration you need to change theirs IDs in jS2n (as all other migrated devices), remove HC2Proxy sensors and that's all, isn't?

For example:

Use HC2Proxy to map HC2 kitchen sensor in HC3, assuming the new ID in HC3 is 145

jS2n={kitchenMS=145}

After migration the sensor has ID of 200, so

jS2n={kitchenMS=200}

While in any case you'll need to go through alarm zones to define migrated sensors.

 

Edited September 19, 2020 by cag014

[gggizmo 6]

1 hour ago, cag014 said:

I really don't understand how come you have delays on HC3?

I'm monitoring delay for last 24 hours and don't see any delays on Z-wave, although there are delays on UI performances.

Really, i don't know but i don't have much more  to try. 

After total reconfiguration of every single devices, swap of hc3 appliance, i'm really near to give up.

to avoid any family complain, today i started my test at 4:30...and i wasn't able to have less than 1,5 sec delay with a single jm line. That's why i swap hc3. I was sure to have a defective unit, but it was a wrong assumption. And i can't figure out why i have no delay at all with more than 40 block scenes.

How many zwave devices have you paired with your hc3?

[cag014 731]

23 minutes ago, gggizmo said:

Really, i don't know but i don't have much more  to try. 

After total reconfiguration of every single devices, swap of hc3 appliance, i'm really near to give up.

to avoid any family complain, today i started my test at 4:30...and i wasn't able to have less than 1,5 sec delay with a single jm line. That's why i swap hc3. I was sure to have a defective unit, but it was a wrong assumption. And i can't figure out why i have no delay at all with more than 40 block scenes.

How many zwave devices have you paired with your hc3?

Unfortunately I have only 5 devices...

AOQ uses same refresh-state triggering engine as block scenes.

Interesting, could you try it with one jM line only?

Anyone else has same delays ?

Edited September 19, 2020 by cag014

[gggizmo 6]

26 minutes ago, cag014 said:

Unfortunately I have only 5 devices...

Interesting, could you try it with one jM line only?

Anyone else has same delays ?

It' s exactly what i did today,. First tried 4 jm lines then 3,two, one..minimum delay 1,5 sec. And the single jm line was to drive a light and a motion sensor located at less tha 3 mt from hc3 appliance

Edited September 19, 2020 by gggizmo

[Rover 47]

1 hour ago, cag014 said:

@Rover

I do understand what is your goal, but I think this is exactly the same procedure that you'll do with other devices, changing the IDs in jS2n.

You can use HC2Proxy only for sensors you need in alarm zones and define them as part of HC3 devices.

After migration you need to change theirs IDs in jS2n (as all other migrated devices), remove HC2Proxy sensors and that's all, isn't?

For example:

Use HC2Proxy to map HC2 kitchen sensor in HC3, assuming the new ID in HC3 is 145

jS2n={kitchenMS=145}

After migration the sensor has ID of 200, so

jS2n={kitchenMS=200}

While in any case you'll need to go through alarm zones to define migrated sensors.

 

So it is not possible to make Alarm Zone usable with <device>'hc2 devices?

[cag014 731]

14 minutes ago, Rover said:

So it is not possible to make Alarm Zone usable with <device>'hc2 devices?

No.

usage <device>'hc2 means to get data directly from HC2.

alarm zones of HC3 don't see that device and the device could not be included.