Soundy

You can, and it will.... how bad will depend on a number of factors. Right off, you're cutting the signal level in half, so the picture will get a little darker. If whatever you're connecting to the other side of the Tee doesn't put a proper load on the line, it could get VERY dark and lose a lot of contrast. Best way to do it is with active line splitters, but it won't hurt to simply test it first with some Tees.

GEM also has one that does power and ethernet over a coax line: http://gemelec.com/index.php?page=shop.product_details&flypage=shop.flypage&product_id=704&category_id=68&manufacturer_id=0&option=com_virtuemart&Itemid=1

Not all AC waveforms are sinusoidal.

Sorry, poorly worded... "AC-like" is what I was going for, since a video signal LOOKS like an AC waveform.

You SHOULD feel dumb! It takes me at least four glasses to become that forgetful!

I can't speak SPECIFICALLY for the VBM model (we've always used the VCM variant), but I should think so. Consider that it otherwise would require separate power and BNC connections, there should be LOTS of room for an RJ45 (or however it terminates; I've never used the version with the built-in balun). Those are the maximum resolutions you'll get with any analog system, so no, it's not "over-buying". Keep in mind that even if it were, you may want/need to upgrade the DVR some day, so there's no point in under-spec'ing the cameras now. Something like this? Most brands (GEM, GVI, MuxLab, NVT, etc.) have similar models...

Have you looked at using a NAS array with supported IP cameras and encoders? Check the likes of QNAP and Synology - many of their systems support direct writing by IP cameras, with management via web interface. QNAP, I know for sure, has versions of most of their models that are designed specifically for surveillance use; I believe Synology does as well, and AFAIK both are based on a form of Linux/BSD/*nix. For analog cameras, you'd just need to use a supported encoder. Both have demos available on their webstites.

Looking through other recent threads, I looked at this CNB camera ( VBM-24VF ) http://cnbusa.com/en/html/product/product.php?inc=spe&seqx_prod=1077#p_v1 I like it's variable focal length - would let us dial-in just how wide of a view we need. One thing that's not clear to me is just what angle it can point to... Could I mount it with the dome pointing perpendicular to the wall? And have the camera's view skimming the wall? Those are fantastic cameras, and yes, you can mount them to the wall - you can aim them pretty much anywhere within a full hemisphere, including looking straight back down the wall. That's an option on some models - without it, you can still use UTP, but you need to use a separate video balun. Not difficult to do, although either way, you WILL need a balun to connect the UTP to the DVR. This does bring up one point though: while the Lorex cameras are powered by the DVR through their proprietary cables, you will need to supply power to any other cameras yourself. The CNBs are dual-voltage, meaning they'll work with 12VDC or 24VAC (or just about anything in between, although those two are the "standards" in CCTV). It appears on most of the specs - for example on the CNB above it says: Video Output Level 1.0 Vp-p (75Ω, Composite) That just means the AC video signal measures 1V peak-to-peak, which is really pretty much universal for analog video (they specify p-p because AC voltage is most commonly rated and measured at its RMS (root-mean-square) level).

Best cameras *for the money* right now are probably the CNB Mona Lisa line. 600TVL, outstanding low-light capability, true day/night, no IR needed in most cases. You won't find them for $20, though. Don't expect much of anything for a measly $20. You think your cheap cameras look "pixelated" now, wait until you compare them to *quality* cameras.

This will be your biggest issue. Everything else should be (relatively) easily doable, but I can't think of any CCTV products that will do this. The problem is, there's no way to do it without either stretching or cropping most or all of the images, and doing that just isn't a desired thing in surveillance installations. You can do 2x2 easily, or 3x3, or 4x4... A six-camera layout is easy enough, and not uncommon - imagine a 3x3 grid, where the top-left four boxes are joined back into one, so you have one large image, with two small one to its right, and three more small ones across the bottom.

With any sort of "lossy" audio or video compression, there's a tradeoff between file size and media quality. A higher level of compression will result in smaller files (and thus lower bandwidth needed for network streaming - important if you're viewing over the internet), but it also means lower relative quality. Lower compression will give better end quality, but requires larger files and greater bandwidth for streaming. H.264 is a relatively new compression format that CAN produce substantially smaller files with minimal loss of quality. However, it does typically require substantial processing power for both encoding and decoding (some manufacturers' implementations are worse than others), and it's not necessarily the best suited for CCTV use. At a full 30fps, the difference can be very noticeable, but CCTV often records at lower framerates, and you don't find as great an advantage as the framerate drops. ?CIF (Common Interchange Format) specs 352x240 video resolution. 4CIF is four times that - 704x480. D1 is essentially the same as 4CIF. So yes, it's four times the resolution. It doesn't directly relate. You're never wasting your money going with higher-resolution recording. Consider if you ever want/need to replace those cameras, you'll probably end up with something that's 480TVL or better. Even at 330TVL, you're still losing substantial resolution by recording at CIF. As with compression, the trade-off is file size. Higher resolution = better detail = larger files.

This is the biggest issue with auto-tracking: unless it's driven by some VERY advanced analytics, with a LOT of programming by the user on how that particular user wants to prioritize things, it really has no way of knowing what object the user would really want it to track. Frankly, I think auto-tracking is largely a gimmick and I suspect unless you dump six figures into it, the customer would end up being disappointed at best, frustrated at worst, with the results. A better idea would be a panoramic high-megapixel camera that simply records everything around it and lets you zoom in after the fact, wherever you want.

No no no, no zeevee. From what I can find on it, it's designed to send HDTV (component/DVI/HDMI/whatever) over a single coax? This is not required with an IP camera. Plug camera into network. Plug PC into network. Install and setup NVR software. Tell NVR software the IP address of the camera. That's all there is to it.

This is the real trick, as a lot of these sellers will advertise "SONY CAMERAS" in big letters... in reality, these are usually cheaply-built cameras using a very common Sony CCD. The sellers in these cases are trying to sell a "big" recognized brand name. And no offense to SEANHAWG, but...

^+2

You CAN get devices that will receive the IP camera signal and output analog video... but what's the point of that? You take all that hi-res goodness and crush it down to 720x480 again. The whole point of IP cameras is that you DON'T NEED a capture card at all - the DVR/NVR simply receives the IP camera video over the network, in its full transmitted quality. This is the case with ANY IP cameras, not just Mobotix. Mobitix are just like any other IP cameras, with some extra-spiffy bells and whistles. That depends on the software. Some of them take over the entire desktop. Others don't. Here's the thing: recording multiple video channels is a system-intensive task. IN GENERAL, YOU ALWAYS WANT A DEDICATED MACHINE FOR IT, and it has nothing to do with whether the display fills the desktop. The video processing can adversely affect other tasks, causing the machine to run slowly and respond poorly; also, running other tasks can adversely affect the video recording, causing dropped frames and other glitches. We build and install DVR and NVR systems all the time using Shuttle mini-cube cases; these generally hold two drives so we use a smaller (80GB or so) drive for a dedicated system drive, and a larger (500GB-1TB) drive for data. If you're doing all-IP though, and don't need the room for a capture card, you can even use a laptop or netbook. Are you talking about an individual spot-monitor output, separate from the regular desktop monitor? Some cards can do this; some can't. However, if you're running all-IP, it would be kind of excessive to have a full DVR card just to use the output (although I have seen it done). However, there are other ways to do it as well. There are software packages that can show one or more IP cameras in one or more dedicated full-screen displays using multiple VGA video heads - like having a multi-monitor setup, where one is your desktop and the others are strictly cameras.

I doubt it as well, but since he's stated the budget is fixed, I'm trying to reduce the requirements So would I Keep in mind that "IR range" itself is a nebulous concept, especially when you throw in the varied ways manufacturers measure that quantity. Whether it's a USABLE level of illumination will depend on the camera as well - your typical cheapie built-in-IR camera has horrid low-light response to begin with, hence the need to add the IR. Use a camera like a Mona Lisa, that has true day/night function and excellent low-light capability, and you need a lot less IR to get a usable result at a given range. I didn't know we were! I'm not saying this WILL work... I'm just saying that within the cost constraints, it's a lot MORE LIKELY to give acceptable results than most other options. Low-cost cameras with good low-light capability, means lower IR requirements, which at least HELPS to reduce the overall budget hit. Not saying there's not more reduction needed, just providing a stepping stone. I'll leave it to someone else to do the actual calculations of IR output vs. beam spread vs. illumination range vs. camera sensitivity, I'm just putting ideas out there As far as the power requirements, keep in mind the OP has stated that the power source is a separate budget, and we don't know what that budget is or how it will be addressed - it could turn out that there's lots of power available to run plenty of illuminators, and as has already been approached, there are ways to reduce the power requirements there as well (PIR triggers for the IR, etc.). BTW, you noted further above, the need for additional relays because of the low current capacity of a PIR's output relay... now without going down to the garage to actually GET a PIR and check the spec sheets I'd suggest there are still ways around the issue. One could spec PIRs that HAVE beefier relays. One could also switch the 120VAC supply to the IRs' power adapters, since the current will be 1/10th of what it is at 12V (and these type of relays' contacts are typically rated at 250VAC or better - naturally one would have to check the specs of the actual units used). Or, again, if the power budget is sufficient... just leave the IR running all night and turning on and off via their own light sensors. You may not be able to, if there's no ambient light...

viewtopic.php?f=5&t=21679 viewtopic.php?f=5&t=20982 And especially: viewtopic.php?f=5&t=20048

Well first, his camera count was six... second, depending on the layout, you wouldn't necessarily need a separate illuminator for each camera - one could possibly flood an area covered by two cameras.

Not really - basic PIRs are a dime a dozen... "pet-immune" types maybe a quarter a dozen. See, I was thinking with good low-light day/night cams like the Mona Lisas, you don't need the tight, long beam - just a "flood light" type coverage should add enough for the CNBs to work with. Hmmm, very good point. At a full 80+ degree 2.8mm shot, anything 100' away will be a speck anyway.

You can hook it up that way, but keep in mind that you'll probably get a "halo" effect as the IR reflects off the inside of the dome. Cameras that come with the IR usually have a black rubber or foam "gasket" that fits around the lens and snugs up against the dome to block that reflection.

^Huh??????

A thought on the camera issue: one of the most popular cameras hereabouts right now are the CNB Mona Lisa series, as they're excellent low-light day/night cameras for the price. Regular retail, four of them would take up your whole budget, but they have been seen on eBay for well under $200. Not that I normally advocate shopping fleaBay for "mission critical" gear, if you can find them there for a good price.... The low-light performance of these cameras would give you good results with even minimal ambient light, and if need be, you could add a couple of cheaper IR flood illuminators, as the cameras should be a lot more sensitive to lesser amounts of IR as well. Put the IR on a light sensor so it only turns on when it gets REALLY dark, and you should be okay.

What type of LCDs are you using - standard aspect, widescreen? Standard computer or TV displays, or dedicated CCTV type? What are you driving them with? Single live camera, multiplexer, computer output? The biggest problem I usually see is that most monitors available now are 16:9 or 16:10 widescreen, and most do a terrible job of stretching a 4:3 image to fill it.

Actually, we prefer long, wordy noob posts - it's the short "what kind of camera do I need for my house" kinds that are maddening. You've actually given us something to work with, rather than us having to ask a ton of questions before we can even START getting anywhere! The PTZs are your biggest budget hit here; two *decent* PTZs alone could take up the entire budget. Replace those with fixed cameras, and you shouldn't have a problem getting the system itself in on budget. Does this budget include the power source, though, or just the CCTV components? Depending on where you live and the amount of sunlight you get in the winter months, you may need extra batteries to keep things going. Be sure to spec deep-cycle batteries. Besides the up-front cost, remember that PTZs will be a constant (relatively) heavy drain on the power if you have them running tours (and if you don't have them running tours, there's no point to using PTZs). How about the internet connection for the remote viewing? If there's cellular service, you'll need a 3G stick or something similar (depending on what your wireless provider has); don't forget to factor in the monthly data plan there. If not, you'll need satellite, and two-way satellite can be REALLY spendy, especially if you want decent upstream bandwidth for streaming video. If there's no power local, I assume there's little to no ambient lighting, so IR will almost be a necessity... however that alone is going to draw a lot of your power if you want it to work to 100'. That means higher capacity needed for batteries and charging panels. Don't worry, it's not as dire or daunting as it sounds But all of this needs to be taken into account (and probably more I haven't thought of yet). Power will be your main concern, keeping in mind that winter months will mean greatly reduced recharge capacity.