MaxIcon

If these are passive POE injectors/splitters, you'll need a power supply for them as well. Do you have a link to them?

I've never used one. I don't believe it does hand-shaking, so the 48Vdc is always on. That's not usually a problem, but could be if you get a short somewhere. This particular model is 8 port, 60W, so 7.5W/port average if all are used. You'll want to check the power rating of the models you want to make sure it'll support them. I don't like running gear like this at its maximum rating due to the risk of Chinese power supply failure, and would consider getting a higher power supply if I wanted to run near 60W.

One issue with a wide angle camera for close-up shots is that it sets exposure based on the whole image. Many cams will let you define the exposure area with BLC, but Hiks are not very good at that, as they don't let you define a specific BLC area. The other thing to consider is whether you'll want to identify anyone further back, which will depend on the camera angle and mount. Pixels per foot drop off rapidly with wide angle, but if your viewing area is limited, that won't matter. This is always a trade-off.

Don't know about the Swann firmware, but the newer versions of the Hik firmware support profiles, allowing you to customize a variety of settings between day and night mode. The main problem is that you can't have it automatically switch profiles based on the day/night switching; you have to create a schedule for it. Hard to say why they did that.

The other consideration with the Hik is that it doesn't set a fixed exposure, just a maximum exposure, so if things are too bright, it still can run shorter exposures to compensate. If your SDI is setting a fixed exposure, you don't have that option, but if you can set the max exposure and leave the minimum exposure much faster, it should take care of it.

For lenses, you can either specify a longer or variable lens when you buy the cam, or buy one from m12lenses.com, being sure to look at their megapixel 1/3" lenses. It's a bit of a pain to change a lens on these cams, but inexpensive. http://www.m12lenses.com/category-s/64.htm For settings, you can run profiles that automatically change at given times if your firmware is recent. I don't remember when they introduced profiles, but you'll see it on the image settings page if you have them. This lets you set day and night profiles with custom settings for the best images.

For your other cams, you'd need an 8 port POE switch to power them. This switch would need to be connected to the NVR so it can access the extra cams, so would need a 9th uplink port if you wanted to run 8 cams from it. Without the uplink port, you could only run 7 cams on it. One possible benefit of this is that you could locate it close to where that group of 8 cams is and run shorter cables from the switch to the cams, with a single cable going back to the NVR, but it depends on your layout. Other options are passive POE injectors like this, which go between your non-powered ports and the cams. This is more cluttered, as you'd need 8 more cables from the NVR to the injector, but is less expensive than an 8 POE port switch: http://www.amazon.com/WS-POE-8-48v60w-passive-Ethernet-Injector-cameras/dp/B0086SQDMM Wireless will be a problem with that many cams, and is not as reliable as wired. Since you need power to the cams anyway, best bet is to go all POE for reliable performance.

You need to set the cams to the same subnet as the modem - 192.168.1.x. Best bet for this is to use Hik's SADP software, or the Swann equivalent, which should be able to find your cams and let you reset the IP address to match your subnet.

What's important is to to decide what you're trying to capture and test for that. Your image is a typical Hik low-light image, where the area illuminated by the IR is clear but the IR coverage is smaller than the field of view, the IR reflective plate is washed out, and the dark areas have a lot of noise. Try walking around the car, especially in the areas where the vandals would act, and see if you can get a good image. Your IR spot will always point at the center of the FOV, so you'll get a decent image there and noisy images everywhere else. You also have a very wide FOV compared to the car; if you want to catch people on the lawn, that's useful, but if you want to focus on the driveway, you might want to get a longer lens. This would also have the benefits of more pixels per foot, and the IR would cover more of the FOV. Because of the wide FOV, you'll have few pixels per foot once they move back a little, and you'll get nothing but a blob at night. My guess is that someone standing at the passenger's door or the trunk will be unidentifiable. It's not usually practical to capture nighttime activity and license plates with the same camera, so if you want plates, you'll need to read up on the other plate posts. This plate reflects so much IR that it's washed out, but exposing for a good plate read will make everything else dark and useless, as noted above. If I wanted to watch the cars for vandalism, here's what I'd do: - Get a longer lens that will cover the driveway without so much yard. - Add an extra wide-angle IR illuminator or a couple of narrower ones that illuminate the areas that the camera's IR doesn't, especially further away. - Do some walkthroughs to see if it captures your motion, if you're using MD, and to make sure you can get an identifiable image in the places you need it.

That capture was 100 ppf, which is pretty ideal at night if the lighting is good and the camera is pointing directly at the plate. As the angle gets greater, like if you're capturing on a through street with the cameras mounted back from the curb (like on a house), there's a trade-off between distance and plate angle. On my Hik 2032 setup with a 25mm lens, mounted at the corner of the house, cars are about 80' away before I can get a good angle on the plate. Any closer, and I get a good view of the side of the vehicle, but the plate is at too much of an angle. At this range, the camera's IR is pretty useless, especially on this 2032, which has lost some of the IR LEDs. As a result, I can get good readable images in the daytime, but at night, there's no reliable legibility of the plate. The other problem I see is that cars are going by too fast at night for the auto exposure to correct. By using fixed exposures, it improves things, but the only way I get readable plates at night is if another car is coming and lights up the plate of the first car. Different cars have different lighting; some don't light the plate at all. So, if you have a good choke point, where you can get a reliable, head-on, full frame image, you can make it work. Other situations, like a house with a big yard set back from a through street, can be more challenging with inexpensive cams using M12 lenses. For the OP, having dedicated cams at each entrance/exit with a good FOV should provide a usable solution.

I have 1 spare cat5e cable ran out the barn now. That one cable could run video for 4 cameras if I used a 4 port switch? How does that work for viewing them if only one cat5e goes to the nvr from the barn. All the camera on the house would have there own cable. Do you loose anything by doing that way? Yes, that one spare cable could handle all 4 cams and more, no problem. Think of an IP cam like any network connection, not like a camera connection; all the cams are sending their video as network packets all the time. The switch and cable don't care how many devices are talking, as long as the bandwidth is sufficient. You can have a switch full of devices connected back to your main network by only one cable.

All the cams I've tried have allowed formatting the cards in place, but firmware can be unpredictable. The big names should have no problem with this. To do any kind of corrupt/damaged file recovery, you'd want to remove the card and plug it into a PC. Some are easier than others to get at; most of mine are quite a pain to remove. I don't know about the capacity issues. I quit using SD cards some years ago because of how much trouble they were (though multiple cams still have them installed), and support has progressed since then.

There are 2 key issues with LP cams - field of view/pixels per foot, and lighting. To get a readable plate, you need enough pixels per foot. This is usually done with a longer zoom lens. Resolution combined with field of view need to give you 80 ppf, give or take, and more is better. As a result, a plate cam will have a narrow field of view, so you need both a wide area cam to watch what's going on and a narrow area cam to capture the plate. Lighting is the other tricky part, and will depend a lot on what cam you choose. Most inexpensive cams will do fine during the day, but may struggle at night, depending on the local lighting. Capturing front vs rear plates at night involves different levels of lighting (headlights on or off, brake lights on or off, is the plate lit, is your IR strong enough to reach the plates, and so on). There are lots of solutions to this, usually involving money and/or special setups. Best bet is to set up a dedicated camera with a lens that gives you 80-100 ppf in the area you expect to capture a plate and see how the results come out. Then you'll know what you need to do to improve it.

I like my BI box to run between 60% and 70% at idle, usually closer to 60%. You don't want it to hit 100% when multiple cams are recording since that can cause video dropouts, and viewing from a remote client will add some CPU load too. I used to run at 15 fps, but dropped everything to 10 fps as I added more cams to keep the CPU down. Anything that BI does to the video adds to the CPU load; rotating video, timestamps, whatever. If you use direct to disk, you can't do any of that, and the CPU will be lower. The cam looks good for that price range, but most cams look good during the daytime. The Hik 2CD2xxx series don't have great night performance, and you have to pay a fair bit more for a cam to improve that.

With 4 cams 250 feet away, you could either run 4 x 250' cables from the cams to the NVR or main POE switch, or you could run a local 4 port switch powering those cams, then run a single cable from that switch back to the NVR. I like to have Gb uplinks, but it's not necessary for just a few cams unless there's a lot of other network traffic. Even running at 8 Mbps, 4 cams would only be 32 Mbps, so a 100Mb switch would hold up fine. For starting from scratch, it depends on your budget. If you want to keep prices reasonable, it's hard to go wrong with the Hik 3MP bullets and 3MP turrets. I'd look at getting a few of each and trying them out, and that will tell you if you need more capability, as well as what kind of lenses you'll need for the coverage you want. If you want to spend more, the sky's the limit.

Here's a link to a post with the newer version. Not sure if it still works, though: viewtopic.php?p=253713#p253713 From that post: Long story short as in my case with login 'admin' and password 'admin' my telnet login is like this: login: admin password: 7ujMko0admin So the prefix '7ujMko0' is the important part here. Older ones use xc3511

If you do have file corruption problems for whatever reason, those can slowly fill up the card until there's little room left for recording, and the card will need to be formatted. I had this happen on one cam, and you always discover this at the critical time that you need the backups recordings! So, if you use SD cards, it's not a bad idea to check them or format them regularly.

Hik, and most of the mid-range manufacturers, are not real strong on documentation, especially about keeping it up to date as versions change. Luckily, community support is strong for the popular brands.

This may be true, but IP cams are the 800 pound gorilla of modern surveillance, and provide the most flexibility going forward. Lots of choices to upgrade cams and NVRs, replace failed components, add multiple recorders for the same signals, use various mobile apps, etc, and lots of pricing flexibility. Unless I had a specific reason to use an alternate technology (existing cable, minimal lag, whatever), IP seems like the best bet for new installations.

I believe they do this to keep the cost and noise down, since 16 port POE running at full power is quite a lot of watts for an NVR size appliance and would require loud fans, but that's just a guess. Like you say, it doesn't make sense, though remote POE switches can be useful if a bunch of your cams are in an area away from the NVR, since you'd only need one cable from the 2nd switch to the NVR instead of 8 cables from the cams to the NVR. Here are a few ways to get power to the cams plugged into non-POE ports: - DC power, either at the camera or using inexpensive passive POE injector/extractors. This isn't so good for long runs due to the power loss from resistance at 12Vdc, and makes for more cluttered cabling and connections. - 9 port POE switch providing power to the other 8 cams and attached to the NVR with the 9th port. Likewise, multiple switches can be used depending on the cam layouts. - Passive multi-port POE injector like this: http://www.amazon.com/WS-POE-8-48v60w-passive-Ethernet-Injector-cameras/dp/B0086SQDMM

That looks like a good in-between solution, especially if it's priced attractively in NZ. I haven't seen that one before. The BB unit is a more general purpose network tester, and you can get it with a tracer module that comes in handy, but it's a bit pricey if you don't do this a lot.

Looks good. I haven't updated to V4 yet, but it's looking pretty stable. The image is pretty washed out where the sun is bright on the road, but that's a pretty small part of the FOV, and your detail between light and shadow looks good. I like to have a longer motion detect break time on critical cams like this. When the motion is briefly stopped, like when the mailman is behind the truck, it looks like it times out and stops, picking up again when he's back in the FOV. The sensitivity is good, so you catch everything, but if someone stood still while doing something without a lot of motion, you'd lose that part of it. This is a trade-off with separating out MD events and having more dead air when, say, a car goes by. Alternately, increasing your pre-trigger frames will catch more activity before the MD kicks in, avoiding the moving object suddenly appearing in the video.

I know some people here have experience with them, but you might be best off at one of the dashcam forums. Here's the big one -http://dashcamtalk.com/ If you google dashboard cam forum, you'll see others.

There are 3 basic ways to test POE power at the camera end: - Cheap simple power tester ($30), just shows the presence of power but no detail, doesn't work with the camera plugged in, best bet if you want simple and cheap: http://www.amazon.com/gp/product/B00AWRUK0G - Better, more expensive tester ($130), not only tests for POE power but will connect inline with the cam and show how much power it uses, as well as troubleshooting some network/cabling issues: http://www.amazon.com/Byte-Brothers-POE1000IL-LAN-Tester/dp/B0015BMXFI/ref=pd_cp_pc_0 - DIY voltage tester (few bucks), make an ethernet breakout box so you can read voltages on the various wires, requires a meter and knowing what you're doing.

Well, it may not seem like it to everyone, but we're already in a golden age of surveillance cams. The capabilities of $100 cams now are far greater than $1000 cams 5 years ago, and progress was much slower then. The same is true of NVRs. I'm sure they'll keep improving, with the best improvements being very expensive before they trickle down to the midrange like Dahua and Hik, but low-light performance has always cost more.