Cortian2

No, your initial post was very helpful thank you, and I didn't realise that you mentioned this at the time. I'm sorry for over-reacting. I think I must have spent too much time on Reddit. S'alright. We all do it. Well, most of us. Ok, I do it You're welcome.

I mentioned using a specific model of Ubiquiti wireless bridge a few posts back. And my reasoning for considering that solution. Did I waste my time writing all that? Two potential issues to consider: 1. Operating environment of the NVR, unless one of the sheds is heated. Reliably. Then I'd still imagine the humidity and dust environment would be sub-optimal. 2. Performance of powerline network adaptors can be highly variable. I would be almost willing to bet the rent that a point-to-point set of 5GHz radios would be better. Again: Not a CCTV expert, or even knowledgeable amateur, by any stretch. But I do know a thing or two about networking tech and practice.

I disagree with the assertion wireless can not be made to work adequately. I think that common wisdom is based upon the fact that most wireless surveillance cameras have crap wireless. This is probably usually exacerbated by the fact that many (most?) WiFi networks are poorly-designed and poorly-implemented, often with substandard, consumer-grade hardware. I would look to using something like Ubiquiti Nanostation loco M5 wireless bridges at each location, with one at the "base" station location. Unless there's earth, stone, metal, water or other RF-blocking/-absorbing things between the "base" station and the bridges, 100 meters shouldn't be a problem. Mind you: I have not tried this! The above suggestion is based upon my knowledge and experience with networking of all sorts over 25+ years in IT. So here's how I'd go about it: I'd purchase one pair of devices and one camera. I'd rig it all up between the "base" and the most challenging (probably, but not necessarily, the most distant) location, and try it. Then go from there. I'm also assuming you're not going to try to run 8MP cameras wireless. That probably would be a fail . Once I get my own surveillance system going, some day I plan to run a 5GHz WiFi bridge circuit to our µBarn, about 100' from the back of the house. I'll hook that to a small PoE Ether switch in the µBarn, put a dual-band WiFi AP up in the peak on the inside and an IP cam or two on the outside. Not because I need to, but because I can

How To Spam A Forum 101 Step 1: Sign up with a bogus account Step 2: Later, sign up with another bogus account Step 3: Post a leading question from one of the two bogus accounts Stop 4: Post an answer from the other bogus account some days later For site administration that's not closely-watching that will appear to look superficially legitimate. (Near as I can tell: Site administration here isn't paying much attention.)

You're welcome. No need to run eight cables to the lounge. Run them to the loft, as you intended, then either: A single cable from the switch in the loft to a switch in the lounge, then a single cable from there to the router A single cable from the switch in the loft to the router, then a single cable from the router to the lounge I prefer option #1, because, as I noted, earlier: My experience is that many consumer-grade routers make poor switches under high demand. And the video feeds from eight cameras I think might qualify as relatively high demand. A small 5-port NetGear ProSafe unmanaged switch is a mere $30 on Amazon in the U.S. But, if you're bound and determined to try the powerline networking, again: Buy a pair of them and test them, first. There's an app called iperf3 that's available for nearly every operating system (Linux, other Unix', MS-Win, Mac OS X, iOS, Android, etc.) that you can use to do a bandwidth test. I use a laptop running Linux for the client because I can subsequently look at the output of the network interface status for a variety of network errors.

In general terms: Figure out how many watts your load consumes and size the UPS so that's about 20% of the load. That will usually get you about 20 minutes of run time. Try to choose a UPS that has an end-user-replaceable battery. They're only good for about 3 years. You should do a run-time calibration (for UPS' that have that option) or run it down to shutdown about once every 3-6 months, otherwise the plates in the battery will plate, reducing capacity, and you won't know when the battery needs replacing.

For devices operating from a fixed location and position: Wired is always better than wireless.

I can't answer any questions about how many cameras you'll be able to view at once, because I'm relatively new to video surveillance gear, myself, and it'll depend upon your software in any event. But, as a retired Systems, Network & TelCom Admin with over 25 years experience building and maintaining networks (LANs, WANs, WLANs) for a living I'm going to suggest cameras <-> switch <-> powerline <-> router <-> powerline <-> NVR is probably not going to result in a happy outcome. First of all, powerline adaptors have highly variable performance, depending upon a variety of unpredictable factors. At the very least I'd obtain a pair and do network bandwidth, latency and error rate tests between Point A and Point B, then Point B and Point C before proceeding. I'd be surprised if you got sufficient low-error-rate bandwidth to support eight video streams. Secondly: It has been my experience that most routers (and I'm making the assumption you probably have either a common consumer-grade or cable company device) tend to be sub-optimal as network switches. Particularly common WiFi routers. In my opinion, based on experience: Switches to switch, routers to route, and WiFi access points to provide wireless access. Avoid WiFi and powerline where wired is feasible. As to WiFi vs. powerline: Whichever works best, which is highly site-dependant.

Generally the way such remote access works is the devices on your personal, private LAN, such as surveillance cameras, are registered with and communicate to a "Mothership." (An Internet-accessible server.) The mobile app then either registers with the device(s) or with the Mothership. When the mobile app then wishes to communicate with the devices on your personal LAN, the Mothership connects the two. That connection may be device(s) <-> Mothership <-> mobile app or the Mothership, after ensuring the proper associations, hands-off the connectivity to device(s) <-> mobile app. Now, security: There are several issues, here: Device security. This is the Internet of Things (IoT) question. Are the devices, themselves, relatively immune to being exploited by Bad Actors? Device <-> Mothership security. Are the devices communicating relatively securely with the Mothership? Mothership security. Is the data on that Internet-accessible server relatively secure? Mobile app security. Is the mobile app relatively secure and relatively immune to be exploited? Mobile app <-> Mothership and/or Mobile app <-> devices security. Are the communications between the three elements in the group relatively secure? Your question may or may not be answerable. It may be answerable only by guesstimate. People would have to know what devices, manufacturers and mobile app you're talking about to have even a hope of answering. The second question you have to ask yourself is this: How much does it matter? The exposure vs. risk question. Sure, nobody likes the idea of being spyed-upon, even if there's nothing interesting to see. But there's a big difference between a jewellery store's or bank's concerns and those of a random private residence. In the latter case: A big difference between a camera in the living quarters and one watching the back yard or porch.