Wednesday, December 18, 2019

Response to all the hyped postings about the Ring "hack story" / Ring review

First off could people stop posting variations of this same story? I must have seen it posted at least dozen times in a couple days. Everyone should have seen it by now. And if they have not they probably will not gain anything from seeing it now. I'm making this cause I'm tired of explaining the story and correcting all the bad replies to all these posts.

Second if you actually read the accounts this hype is based on, it is mainly A guy doing these "hacks" for a podcast. He and his audience thinks it is funny.  Like a radio talk show host making crank phone calls.

Third any script kiddie can do this to accounts on ANY site if login info is reused on multiple sites without 2FA. BTW This is totally diff from brute forcing which even basic VPS served sites block pretty quick these days.

This is the 21st century. Passwords are like door locks. Reusing passwords if like locking your door with an old style lever tumbler lock might be better than leaving it unlocked but I doubt you are still using them on your doors. SMS 2FA is like upgrading to a typical pin tumbler lock. True 2FA with a fob or app is like adding a deadbolt. If you have deadbolts on you doors then you should have 2FA on any accounts you care about someone getting into.

If any of the above is news to you, get studying or hire someone to handle your online interactions NOW! At bare minimum use something like Lastpass to generate unique passwords for each and remember them for you. Personally I would also suggest getting a domain with mail forwarding and using a unique email address with each site as well. That way when you start to see spam come to that address you know the site has either been breached (often this never reported much less makes the news) or they have sold on your info. In the former you need to change your login ASAP. In the later you may want to route that email address trash and stop doing business with that site.

Now as whether you should get a Ring

I bought the original Ring Doorbell back in 2015 and sent it back same day. The video was awful and it took forever to connect.

Recently I got a deal from Amazon for a Ring 2 with Show 5 for less than the normal Ring 2 price so I thought I'd give it another go. Some of the things I ran into:

  • Failed install with flashing light pattern not in list of options. All options basically said rerun setup. Redid install from scratch despite doorbell showing in app since there seemed to be no way to rerun setup on it or remove it. Seemed to sort in end was same cam and old one disappeared.
  • Installed at my desk then moved outside to install on the door frame and could no longer connect to it. Assuming was because it needed to switch to closer AP so told WiFi to do reconnect. It would not reconnect till I powered it off and on again. This is probably because I have multiple APs but I've only seen this before in really old IP cameras where they get stuck on a channel till rebooted. Fortunate this should not be an issue that often.
  • I've been running a ping sorting the WiFi issues
  • Initial tests seem to show the same bad lag as with the original cam. I'll add more details after more tests.
My playlist of sample videos. Note most should have link in the description to a video of the same shot from another camera to compare with.

Seriously if you are just looking for a cam to watch your front door there are a lot of better options out there for less money. The reason to get Ring is the same one I hear the most complains about, the neighbors app. The video doorbell option with replacement guarantee was the main thing I was looking for.  If a doorbell is your main use and you do not need the neighbors app then you might want to look at the Ezviz DB1 It has 3MP rez (half again what Ring has) and can stream to your NVR as well as the cloud and SD card.

Tuesday, December 17, 2019

Pan, Tilt and Zoom worth the money?

In my experience, Pan / Tilts and zoomables as a rule are generally a waste of money unless you are actively monitoring them. After all that was what they were designed for originally. Panning cams were a way to reduce costs back when panning motors were a lot cheaper than cams. These days though with cameras so cheap, even if you are monitoring your cams you generally want as much coverage as you can get with software alerting you to motion or openings where there should be none.

Realistically at home you are not going to be monitoring your cams. Even though I have dedicated monitors so I can see what triggered an alert or what the dogs are going on about, I am still mostly looking at what happened, not what is happening. I have a few PTLs and 1 PTZ. Except for the Wyze Pan I bought to try the tracking, they are all older cams bought before I found I virtually never move them. At the moment I have 2 PTLs are in the main living area (open plan kitchen, dining and living rooms) and I'm thinking of replacing them with wider, higher rez cams and moving the PTLs to the shop where I might want to look about without having to go down there. Mainly to upgrade the shop cams though. I need better coverage in my main living area to track "things" the cat brings in to play with to their hiding places. The PTZ in in the garage I've mounted several places from using it to monitor the gate but decided and Escam QD900 with a 12 mm lens did better. To the north power so I could try and see what the dogs were barking at but it kept going offline. To the garage where I'm replacing it with a GW Security GW5747MIC which covers almost the same area the PTZ can be pointed at almost the same level of detail as the Foscam FI9828P zoomed all the way in. Next stop will probably be in or under the shop though the barn might be a good spot as well to be able to center and zoom on things.

Now I do have a few others that are remotely zoomable but mainly for the autofocus feature they need to support that. If you are looking at close and far targets in the same view, autofocus can make a big difference.

BTW my original pan was a fixed bullet mounted on an antenna rotor. No night vision but it was high tech for a home cam back in the late 1990s.

An example:

DS-2DE4A425IW-DE with sample videos. $489 to $900
Amazon link Newegg link

What you are really paying for is the 25x zoom (120mm lens) and the autotracking. These are designed for use in places like a parking lot after hours were someone monitoring the system can use them to get detail and hopefully track things the monitoring person might not notice. Some high end systems claim to direct PTZs like this based on detected movement form other cameras and sensors. (See below.) But many people seem to think they can use a camera like this instead of covering an area with multiple cameras. Despite what the ads imply I'm still waiting to see a video where a camera actually tracks an object well. Much less deals with multiple moving objects. The wider arc you try and monitor with one of these the worse the results. Note too even if tracking includes zoom, zooming in is optical which limits the field of view and requires time to refocus the image. 

For the equiv money, you could get 4 to 10 cameras in the 4 to 8 MP range setup to cover the the same area at various zoom levels and be pretty much guaranteed to see the object the whole time it would have been visible from the location. And at probably close to the same level of detail while the object is moving. This is closer to the modern WAMI approach to monitoring that assumes multiple objects may need tracked at a time. Granted the NVR required to handle the video will be need to be more powerful but you will get the shot you need and probably save money. Here is an example of tracking a guy looking for stuff to steal on my property after the fact.  While it looks like the kind of videos the tracking cams claim to do it is zoomed and panned in post with views from 5 cameras. For most of it the odds are pretty low a "tracking" cam would have followed him without predictive level AI smart enough to look for him on the other side of an obstruction.

Integrated systems

Now it would be awesome to have some cams like the DS-2DE4A425IW-DE mounted above the roof and tied in to my system to track and zoom in for detail using the other cams and sensors as a guide but that is some pretty high end linkage there to do right. My current system links PIR motion sensors and alarm contact sensors to highlight camera on consoles, mark the video from the linked cams and in some cases even switch the main video feed to the triggered console. This gets me most of the way there and is basically a freebie with my Blue Iris servers being linked to my Homeseer home automation setup. In theory I could integrate some of them to really be worth the cost the system would need tighter integration like the systems that direct PTZ cams to an area based on sensors and triggered cameras which is a costly and is probably best left to pros to get working right.

TODO: find web page again that showed a system with PTZs being triggered by static cams and PIR sensors.

A note about other sensor options

And then there are these

Dahua's 8 x 2MP Multi-sensor Panoramic + PTZ Network Camera which has a 4MP surrounded by 8 2MP cams so the camera has a constant 360 view with a linked PTZ to center and zoom up to 40x on a moving object using their Smart Track linking system. Note this is $7000 camera though there are people claiming on sites like eBay that they will sell you one for as little as $2000. Again this basically gives you a 2 MP panorama with hopefully good detail IF it is tracking the right object. 

Digital Watchdog MEGApix PANO 48MP Outdoor Dome Camera 4 linked 12 MP cams in a single case to give a 180 panorama view. Just $3120

And all these 20MP options.

Second opinions:

Monday, December 9, 2019

Answer to: I just want a decent system

The problem

Decent is pretty subjective. Kind of like asking what is the best way to get to work. You'll get answers from ride a bike to only a Karlmann King will get you there. Not only is it subjective relative to what they need and can afford it is subjective relative what they have actually tried model wise and using the resulting video. As one installer put it "I have to explain to my clients that if you have only ever ridden a horse and someone gives you a small compact car, you are gonna think it's awesome purely because you don't know you can get a Range Rover or a Truck for the same money."

But you still just want something for outdoors

If you just want to get 4 outdoor cams to cover the yard around your rectangular house you could start with this 4 4MP cam bundle from Reolink though I think it was like $50 cheaper last week. Still $341 is not bad to get you started. You can reuse the cams later with about any NVR. They are pretty decent for the money. And the NVR is speced for 8 cams so will probably cover you unless you need more features or go nuts buying cams like me. (60 cams as of this writing on 4 serversNote though you might have a time getting non Reolink cams to work with it.

Better yet if you have a PC not doing much,  get the Reolink RLC-420 cams and a copy of Blue Iris. Note however Blue Iris 5 seems to have developed issues with some cams (including Reolinks) that Blue Iris 4 does not have.

But you still just want something for indoors

Now if you are looking indoors and wireless get some Wyze cams. Pretty much a toss if you want to upgrade later but you are talking about $100 for 4 plus some SD card you can probably reuse if need be. And they come in handy for monitoring things where you want a temp setup or need to get something in place quick. But you read this about the issues with wireless.

On the other hand a bit of research could save you time and money 

The above setups will help you get a feel for the low end. But if you want to do the research up front to get the most bang for your buck you can start reading my post "What is the best camera for X?" to sort what specs you need to be looking for the match what you want to monitor.