re: General security camera discussion (was "Unifi Protect security cameras")

quote:

---

That's much fast notification than the sub right?

---

Maybe not. I just checked timestamps on the emails I get from UINotify vs the reddit sub, and it looks like they are pretty close together at the start of the drop but by the end of the batch the UINotify emails have gotten behind. They both started around 6:03AM but the sub finished at maybe 6:15 and I didn't get all the emails until 6:25. Looks like UINotify's paid plan gets the emails before the free plan, but I'm not paying that and then have to get up early every morning to see what Santa brings.

quote:

---

paying

---

Yea no, the only thing I'm paying for is gear

If smart detections and a good UI are weighted heavily, then I don't think anything comes close to Protect at this time. If the budget is weighted more then I would go with the reolink. They do offer good quality for the price, but I can't speak to the app and UI as I have not tried it.

I would avoid WiFi cameras, and especially battery powered ones, at all cost. They just miss too many events, and pulling out the ladder to charge them all the time gets old fast.

quote:

---

Wireless - EufyCam 2 Pro Kit

---

I just boxed mine up to return. I liked everything about it except it would not detect human motion more than 20' away which was a deal killer for me.

Another new camera in EA, this time a complete 180 from the last one. While the G5 Bullet is priced shockingly low at $99 (at least while in EA), this one AI DSLR is priced at $2,499.

So far I have thought of zero use-cases for this camera.

They have finally put out a new Flex, skipping right over G4 and going to G5.

Same form factor as the G3 and seems to be the same size, so hopefully all the G3 mounts and skins will work with it.

Double the pixels (1080->4MP/2K), wider FoV (80deg->102deg, which for a general purpose camera greater than 90 degrees is great), and I have to assume it will do smart detections. Price jump from $79 to $129, which is a bit more than I'd like but not awful I guess.

After a rash of car break ins in our neighborhood over the weekend (Sat Night/Sun AM), including our across the street neighbor's car getting stolen, I'm looking to upgrade pieces and parts of my reolink system. I'll say that our current system "saved" our family. After checking car doors on the entire street, they got to our driveway, saw our cameras and took of RUNNING (literally) up the street. Our 4 and 5 mp cameras are about 2-3 years old (I installed them during Covid).

The way the thefts unfolded was interesting. The thieves parked in the one spot on our cul-de-sac where I couldn't get a clear view of their vehicle (far away and dark). They drove down the street and parked in such a way that their headlights blinded my camera the entire time. When they left, the never pulled into the cul-de-sac, they simply backed up keeping their headlights pointed at my cameras. While good in the daytime, they just don't have the dynamic range to deal with that at night time.

But all was not lost. The wives in the neighborhood have a ongoing group-me text to keep track of the kids and what not. Using the time stamp from my NVR, several folks were able to check their ring doorbell cams and the like and we got a few grainy views of the vehicle as it left. Well, when I checked my cams again yesterday AM, the vehicle returned. This time I was able to get a good view as in entered and left the cul-de-sac. I spent about an hour at the police station yesterday AM letting them download and review my video(s).

Because of all of this, yesterday I upgraded the main camera that looks down our street. I installed one their 4k cameras with "onboard person and vehicle detection" (RLC-811A). It was basically a plug and play exchange. However, after getting a few bugs worked out with the NVR (had to upgrade the firmware), I've discovered that the person and vehicle detection doesn't work with my NVR.

So, I'm in a bit of a dilemma. Do I just keep what I have and live without the extra bells and whistles like I have the past couple of years, or do I spring for a new NVR that gives me P&V detection? I'm currently maxed out using 8 of the 8 channels on my NVR. While that's fine, I'm also feeling limited by the 2TB storage space (about 4 days worth of recordings.) Do I spring for the 16 channel version? (There are 1 or 2 spots where I feel I could add additional cameras. Should upgrade a hard drive storage space? Jump into blue iris? FWIW I like the reolink ecosystem and have basically had zero problems with it since I installed it.

Let me have it.

quote:

---

Do I just keep what I have and live without the extra bells and whistles like I have the past couple of years, or do I spring for a new NVR that gives me P&V detection?

---

Well, what's P&V detection worth to you, and how does that compare to the cost to upgrade the NVR?

quote:

---

I'm currently maxed out using 8 of the 8 channels on my NVR. While that's fine, I'm also feeling limited by the 2TB storage space (about 4 days worth of recordings.) Do I spring for the 16 channel version? (There are 1 or 2 spots where I feel I could add additional cameras

---

Sounds like you're looking for more reasons to justify the new NVR.

quote:

---

Jump into blue iris?

---

A lot of people like it obviously but I don't think I could wash away the stench of having to use Windows.


So your dilemma here is very common in the "traditional" NVR world. You have a fixed number of camera channels and a fixed number of hard drive bays, and the features are often fixed as well (not much reason they can't add P&V detection to your NVR other than they just don't want to, the compute work is done by the camera so the NVR just has to understand the event, new firmware could add the feature). I've mentioned a few times in this thread that I have an NVR software project sketched out, but of course this is a huge undertaking. That said I have found some good open source pieces to that puzzle, hopefully soon I can just put everything together.

Anyway, the main feature I'd implement would be to make the NVR hardware "stackable", kind of like Ubiquiti's. You can start with one machine that can handle 6 or 8 cameras, then when you need more cameras or storage just add a second machine. And a third, fourth, etc., essentially forming an NVR cluster. And they would be cheap hardware like Raspberry Pi level, the hard drive would be worth more than the rest of it. So upgrades are painless and seamless, no tossing of old equipment or swapping hard drives around. You just keep adding these cheap boxes for more cameras, storage, and redundancy. And you could scatter them in different rooms of the house so a thief won't know how to take all your footage.

I think someone should make a NAS that works this way too.

quote:

---

That's brilliant actually.

---

I thought so.

I've had some talks with the developer of Moonfire NVR. Moonfire does an excellent job of recording and playing back video, but the front end is not really polished and doesn't do any motion detection or anything. It's strictly a recorder, but that's fine for my project because I would focus on the front end UI and get as many events from the camera event streams as possible.

Another advantage to the multi-box setup is you can add boxes with different features as necessary. Like maybe your cameras don't do object detection, so you could add it by getting a box with a GPU or maybe an NPU.

The ultimate goal if I ever get there would be to develop open source cameras that run the NVR software so you don't need a central recorder at all, the cameras are the NVR cluster. This would be extremely flexible. For example, the RPi has both ethernet and wifi, so if that is what your camera is built on you could add a PoE module to it and use that for power and data, and then it has wifi built in which could be used for direct camera to camera communication. This could be used to bridge gaps to detached buildings, or you could have battery backups inside the cameras so that they could keep clustering via wifi and sharing data for redundancy if the power goes out or even if it's yanked off the wall.

quote:

---

I'll say that our current system "saved" our family. After checking car doors on the entire street, they got to our driveway, saw our cameras and took of RUNNING (literally) up the street

---

I struggle with marginal upgrades. Night time filming is tough. If you had 10x the cameras at 10x the cost, would you necessarily have made an ID that led to an arrest? Probably not guaranteed.


I like my detections and timeline scrolling with Unifi. They’re very nifty. But it’s not like I can trigger some sort of automation/push notification for a grey ‘97 Taurus showing up.



I still want my wife to let me put an $1800 PTZ camera with 22x optical zoom on my house. She’s still balking.