Soundy

If all he wants is broad over-views of the work areas, then putting the cameras up that high is probably even preferable, as it will give a wider shot and reduce the chance of actually seeing any identifiable faces, in case anyone has any concerns about that. The only potential issue would be the girders partially blocking the views, so mounting the cameras to the girders , or even just at their level, may be desirable. There are a couple ways to do this. One, you could use "beam clamps" (often used by sprinkler guys) to clamp to the girders. They typically look like this: . You'd attach the camera mount via the bolt-hole on the bottom. The other common option to use a piece of pipe to extend the camera down from the Q-deck - I've commonly used an octagonal junction box and a cover plate with a threaded hole, and a length of iron pipe, into a dome enclosure.

I dunno about Geo, but on the Vigil systems with the XECAP cards (one switched-analog and one virtualized MUX output), the MUX output can display whatever IP cams are supported. Since the output is generated by the system, it can be fed whatever the software can draw.

We have a PC-based (Windows XP) DVR in a site that seems to regularly be "missing" video from some cameras. Ongoing troubleshooting has narrowed it down to those cameras conveniently being turned off somehow at "convenient" times, or someone tampering with the DVR. We're hoping at some point soon to add a hidden camera to watch the DVR, but in the meantime, I'm looking for some sort of covert keylogger/activity logger, preferably freeware or open-source, that I can install on the machine before I take it back to them (I've had a loaner in while testing their original machine, and they're reporting the same symptoms with it). Has anyone done anything like this before, and does anyone have any suggestions for what to use?

Just one other note: the main PHYSICAL difference between RG-59 and RG-6 is the size (thickness) of the center dielectric - it's larger with RG-6, in most cases requiring a different connector than RG-59. This is true for most crimp-on connectors (including compression-fit types), and since the overall cable diameter is correspondingly larger as well, will also affect most twist-on designs.

I've used a LinkSys SFE-1000P for two different sites now, and been pretty happy with them. They have eight 10/100 PoE ports and two fiber/GbE (gigabit ethernet) ports - cameras go on the PoE ports, DVR and NAS RAID rack go on the gigabit ports, and everyone gets along nicely! They have management capability as well, but I haven't yet had need for it. Sure, they're just two different types of wire. Current PoE supply spec is, I believe, 48VDC at up to 15.4W, as MYA notes; PoE devices should be able to operate on anything over 12-15VDC, so line loss shouldn't be an issue except over a VERY long distance... and remember that ethernet over Cat-5e is limited to 100m anyway. Cameras should not be burning out unless: a) the camera doesn't adhere to the spec; b) the source doesn't adhere to the spec; c) there's a serious wiring problem. I've had a couple cameras die on PoE, and it was specifically because of a wiring problem.

Why not just use the remote client software? Much more convenient for browsing/searching recorded video. Besides, you don't want to be "changing" the original files if you ever plan to use them for evidence of any sort.

Actually, there's analog in a DVI output as well, on separate pins... hence your basic DVI-VGA adapters that most DVI-equipped cards come with.

It would help if you had a link to the specific cables you're talking about, but at a guess, I'd say the cable is intended for component-video output - separate red, green and blue signals - and won't work unless your TV also accepts component input. What DVR card are you using? Doesn't it have an RCA composite-video output?

The problem with cheap cameras like that is... well, the image quality is cheap too, especially in low light. Don't expect to get any sort of usable picture at night unless you're parked in a very-well-lit area. Even right under a street light probably won't be enough for evidence-quality ID, although it might give you an idea who it is if it's someone you know.

Don't expect great quality going from PC display down to TV, regardless of which way you do it. My PC's ATI All-in-Wonder card, with built-in tuner, has composite and S-video outputs, and at XGA resolutions only large text is really readable on my 14" monitor (an old Commodore 1701!) Doing a dual-monitor setup in this way can be tricky, as well, because with most cards, Windows won't activate or acknowledge the second output unless the TV is connected and turned on at startup. DVI is "digital video interface" and is the new standard over VGA - frankly, if your computer monitor supports DVI (as most LCDs do now), you should be using that instead of the VGA out, as it will generally give you a crisper, cleaner picture.

There are a number of ways to do it, depending on the DVR and the TV and desired final result. Most newer TVs (flat panels particularly) have VGA, DVI or HDMI inputs that can be fed directly from your DVR PC. You could also use something like an AverKey to convert a VGA display to S-video or composite. Most PC DVR cards I've seen have an analog-out that will feed one camera (usually selectable in software) via composite, that could be get into a regular A/V input on your TV. Modulators that will put a composite or s-video signal onto channel 3 or 4 are cheap and easy to come by, but beyond that, you're probably looking at several hundred dollars for something designed specifically for CCTV to modulate it onto a higher channel, if you want to go that route.

That opinion is garbage. Why would a "professional" want to limit himself like that? Standalone systems are great for what they are, but they're generally more limited than PC systems - for starters, they're nowhere near as expandable: I have a PC-based DVR in one site with almost 10TB of storage, with three internal 1TB drives and a network-attached RAID with eight more 1TB drives in a RAID-5 configuration for another 6.5TB available. I haven't yet run across a standalone that supports external storage in any form, be it USB, Firewire, or NAS. I've also not yet run into one that supports more than two internal drives, and most are very picky about specific brands and models. And the first time you have to enter camera and system names by spinning a jog wheel and tapping on a couple buttons, you'll be cursing the lack of a keyboard... There's nothing inherently WRONG with a standalone DVR (at least not as a concept; I've run across some pretty poorly-designed and shoddily-built ones), they fill a specific need, primarily that of low cost. There's a small crossover range, but the two really fit different market segments.

f-stops are a rarely-used and even less-understood term when it comes to CCTV, I find... few guys even consider the actual maximum aperture of a lens, or understand the relationship of the f-stop numbers, despite the fact that most lenses list at least the max aperture. Either you use an auto-iris lens and don't think about it, or adjust the manual iris until the picture looks right - I guess there's not really a lot of call for it in CCTV. Coming from 25-year photography background, I guess I just have more experience with the subject

I think more then twice Right ? Not PRECISELY twice, but fairly close. Without doing the actual math, consider that f/1.4 is exactly one stop "faster" (ie. twice the light) as f/2.0... and you can extrapolate that f/1.2 will be approximately one stop faster than f/1.8. If it wasn't past my bedtime, I'd do the actual math for you, but at this time of the night, you're on your own

There's little point in most instances with going to any kind of super-fancy microphone; full-range 20-20k response simply isn't a necessity. I've done it with simple computer mics, cheap Radio Shack dynamics, etc. The main thing you want to do is make sure you get a dynamic and not a condenser mic, as a condenser requires a power supply, which is usually in the form of a battery that has to be replaced on occaision (usually immediately AFTER someone discovers - the hard way - that the system has been recording silence for two months).

The 3700 isn't really great for low-light, no. Check out the Pelco CC3751H2 for that. A zoom of that range probably has a relatively small maximum aperture, which will limit the amount of light it will let through; looking at Pelco's spec pages, it's not entirely clear, but it's probably no better than f/1.8 Try using a shorter-range zoom or even a fixed-focal lens of f/1.2 or better (smaller number = larger opening). f/1.2 allows twice the light of f/1.8.

Was there a brand, model name or serial number on the DVR itself? You can usually look up the manufacturer, and sometimes the specific model, of such cards by finding the FCC ID number (should be silkscreened on the PCB) and looking it up at http://www.fcc.gov A photo or two of the card might help us identify it as well.

Pretty good article... the only thing that I don't like about it is the term "megapixel lens" in general. It's a misleading, inaccurate, and inappropriate term that unfortunately has come into such common use that even the manufacturers are using it. For one thing, it gives people the incorrect idea that lenses "have pixels"... kind of like in my car-audio days, when I'd regularly hear guys asking how many watts a sub "has", and assuming that speakers with higher wattage ratings would always "hit harder". Second, it gives the idea that "regular" CCTV lenses simply will not work on megapixel cameras, which is also misleading and inaccurate. They may not be the idea choice, as your article details, but the myth needs to be dispelled that megapixel cameras REQUIRE "special" lenses. Part of the problem with marketing terms like this, especially this early in the technology's development, is that it can be too-easily tossed about with no real numbers to back it up. Any manufacturer can claim their lens is a "megapixel lens" with little to back that claim up, and with no distinction whether it's suitable for 1MP, 3MP, 5MP or higher. Sure, the good ones will include spec sheets with that data... but most won't, and most people won't know how to read it anyway, and you end up with issues like in this thread where people can't understand why they went for a higher-resolution camera (5MP in this case) when they can't see any improvement over a cheaper one (3MP)... I mean, it's a "megapixel lens", right? Anyway, that's my rant... other than propagating a bad marketing term, it's a good article It goes hand-in-hand with my other rant, people who refer to megapixel cameras as "High Definition" or "HDTV" cameras, trying to cash in on THAT unrelated marketing term. Not only is a poor attempt to use an incorrect but popular buzzword, but it does the cameras a disservice, as anything over 2MP is far higher resolution than HDTV.

"PoE" = Power over Ethernet... regular analog cameras != Ethernet. Also, while a "quad-balun" may be convenient for four cameras, it's not necessary - with only two cameras, a pair of spring- or screw-terminal baluns (such as these) will work fine and probably cost less.

In addition to the low light hampering autofocus, keep in mind that as the aperture opens up in lower light, the depth of field shortens, so things closer and farther than the point you're focused on will go out of focus.

Definitely sounds like you're getting to much voltage drop. Unless there's something seriously faulty with the power supply, 1A should be way more than enough current (heck 300-500mA is probably enough), so it's not that. I'm guessing this is a regulated supply as well, as an unregulated "12VDC" unit would probably be producing anywhere from 14 to 18V at the output, and the drop over a longer run wouldn't be so problematic. If it's regulated and actually putting out a solid 12V though, it may be dropping to 10V or less by the time it gets to the camera. Again, a multimeter is your friend - check the voltage at the supply and at the camera and see how much you're losing.

See if it's printed on the wire itself - might be 22, 20, 18, etc. What is it sold as - speaker wire, zip wire, etc.? See http://en.wikipedia.org/wiki/American_wire_gauge for more info.

What size (gauge) wire are you using to extend the power? I'm betting it's too small and the voltage is dropping too much by the time it reaches the camera. Try metering the voltage at the camera.

70+VA is a lot of current for running over two pair in a Cat-5 (that's where the real bottleneck is for the current. I'd definitely recommend a separate 18-16ga. power wire for that kind of juice.

That won't change the software. Drivers merely provide "translation" between the hardware and the software (or in most modern cases, the operating system itself). The software that does the actual recording, still has to be able to handle 32 channels of video. If you look at the software, it has options to configure 16 cameras, and to view 16 cameras. Plugging another card in won't (necessarily) make it magically show an extra 16 cameras (disclaimer: some software WILL do this, but as I already said, I'm not familiar with the GeoVision system *specifically*). That I don't know. It may TECHNICALLY work, but there may be licensing issues involved if you want to be LEGAL. I'd suggest waiting until someone responds with a definitive answer whether this will actually work with the GV systems... otherwise you may end up with a card that won't do what you want.