Results 1 to 2 of 2

Thread: "Super" reflex

  1. #1
    Automated Home Sr Member
    Join Date
    Oct 2009

    Default "Super" reflex

    Hi Karam, I think you may have mentioned sometime ago that you had given some thought to a "firmware" module, where more advanced reflex actions can be stored. My main issue is that I am running WinXP on an EPIA machine and given that there is no more security updates for WinXP, I would have to upgrade my OS and motherboard. As always, I would prefer Cortex running on linux but I guess this will never happen? On a day-to-day function, how much MORE does Cortex actually do over and above reflex? Probably a huge amount, but just wonder if all the logic things can be offloaded? Perhaps a raspberry pi solution using its GPIO pins?

  2. #2
    Automated Home Legend Karam's Avatar
    Join Date
    Mar 2005


    Our use of the terms Cortex and Reflex is probably not far off the analogy mark. There is a huge difference. Switching a light on by pressing a button isn't simply a matter of the button sending a signal to the light actuator, infact it goes through a whole load of decision making which takes into account all sorts of possibilities and the wider state of the whole system - even whether the button that you pressed should presently be interpreted as a light switch in the first place (think of a DFP button being used within a menu at that time for instance). Of course ultimately it is all logic but to create the kind of high level of abstraction that is required and the complex integration means you need a matching coding environment i.e some high level language with object oriented features. Of course you can get such an environment on a linux platform but you have to remember that when we first started out there was nowhere near the level of tools and support as there was/is for Windows. We didn't particularly want to write all the wider features such as telephony, speech processing, video processing, graphing etc from scratch. If we'd just set out to deliver yet another logic/schedulling/macro type of system then possibly linux would have been ok even then. Regarding security, I'm not an expert on such matters these days so I can't make a comment about linux vs Windows. I know that Windows has long been lambasted for security holes but other than hearing such things I do not know the exact details of why a hacker could not get through a linux based system any more than a Windows one. May not be as easy but probably just depends on the prize.

    As a general audience comment: To some extent it depends on how you deploy our system as to whether you appreciate the difference that Cortex makes. In other words if your installation is mainly a collection of controls but not capable of full automation then it probably won't be obvious why Cortex is so different, but if your installation is a 'full' deployment then the automation aspect becomes much more apparent and the differences between Cortex and Reflex and in fact Cortex and any other kind of home automation software become much easier to understand.

    But certainly I can understand that it could still be attractive to use our networked hardware with a souped up Reflex type of control. It would be possible of course if you wanted to put the effort in to do it. Certainly a Raspberry Pi might be a good platform and you don't need to use GPIO pins - just go through USB to the PCU or via ethernet (when we get our finger out with the PCE..). It isn't entirely plain sailing though even at the lower levels because although its easy enough to send a single command to a module on a sedentary network its quite another thing handling the cacophony of data flying around on a dynamic network, queuing up and handling incoming and outgoing data, dealing with any communication errors and so on. Not particularly specific to our system - just perhaps more apparent due to its extent.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts