Soundy

Any reason CCD wouldn't work, or are you going for a tube-specific effect?

Old B&W Panasonics would certainly work for that. They're probably CCD sensors though; tube types will be rare indeed. Welcome!

Nope, that looks about right for a 16mm lens. See here for examples of different lens focal lengths: http://www.ezcctv.com/cctv-lens-guide.htm 4mm and 6mm lenses are much more common on those types of cameras.

WDR is all a matter of signal processing, so there's nothing about megapixel sensors that prevent it from being done... just a matter of design and cost. Panasonic has megapixel versions of their SuperDynamic cameras... check out the WV-NP502, or the dome version, WV-NW502. I've tested the NP502 and it works great (although I found cranking the ABS and BLC settings worked better than using the SD5, at least in my test setup).

A UPS will still help with power filtering and will protect the DVR from even slight brownouts. For the cost of a basic SOHO UPS (<$100), there's no reason not to.

None that I've come across yet, other than the price - I've seen these types of things go from $75/set to $300 *each* (two required, of course). Certainly not worth installing coax just to use them, but they're great when you have existing coax and running new Cat5 is impractical or impossible.

http://www.geovision.com.tw/english/4_4_main.asp All I got was a Questionnaire? Did you read the TOP OF THE PAGE? It's a Support Request form. The idea is to ask GeoVision for them.

http://www.geovision.com.tw/english/4_4_main.asp

Hmmm... well the only other thing I got is the power run... some of kind of severe noise being induced in that. He's swapped the power supply itself, so that rules that out.

Well that was disappointing, I was expecting some battle action!

I'd say that camera is worth about $30.

That pretty much narrows it down to the cabling run...

I don't have any specific suggestions for a DVR, but if you're using an analog camera with a video server, you'll probably want to make sure the DVR you go with has "hybrid" capability...

The problem with buying a "generic" card is that you don't know for sure what you're going to end up with until you get it. The drivers linked to are generally intended for basic support of certain chipsets; the problem is, many different manufacturers use the same few chipsets, but the cards themselves may vary in design so that they don't work properly with the generic drivers. More common than software not working, is poorly-written software and/or drivers causing system crashes...

From the first post in the thread: Then they need to change it for a better one, or 2 even. Or visible light.

From the first post in the thread:

This is where your experience trumps mine. I look at that jagged edges around that car and I only see interlacing caused by a moving car. Everyone seems to be working on the assumption that this is a moving car, yet nowhere does the OP state that it's moving. Far as I can tell, it's a parked car. This is true; however, the OP also states, "this is what it looks like even when the object is sitting still."

You know, some motion-activated floods could be considered a "safety and liability" feature - not a bunch of bright lights on all the time, but just a couple in key areas that only come on when triggered by the presence of a person. After all, someone walking between the cars at night could conceivably trip and injure themselves, thereby opening you up to a lawsuit. If the HOA denies you this SAFETY measure, it could make them liable as well *ahem*.

I don't know if you'd see a big difference in the CARDS from your end... moreso in the software, and a lot of that is personal preference. I've not used the Aver software, but I have dealt with several GeoVision systems... they work alright, but I'm not a fan of the interface.

Well your first problem appears to be your resolution: the top images are all (approximately) CIF resolution, which is just too small to do anything with given the wide angle shot; you'd have to zoom the camera in tighter. IP will definitely give you more resolution and clarity, but most cameras also need a lot more light - I'd recommend a motion-activated floodlight rather than IR. The bright light snapping on will often scare away someone who's lurking about; at the very least, the natural instinct is to look for the source of the light, which if it's near the camera means you'll get a great, well-lit, straight-on facial shot. BTW, I'd suggest converting images to JPG before posting them - BMP files are huge and take forever to load over a slow network.

We get 75VA and 100VA closed-frame models at ADI/Burtek... but yeah, 40VA is the highest I see for "plug-in" types at our usual retailer.

I don't know about GeoVision specifically, but most PC-based systems I've used, the "available" framerates can be divided between cameras any way you want - for example, you could have two cameras at 10fps and two others at 5fps - anything that adds up to 30. And yes, those rates are fixed - one input set at 5fps will always record at 5fps regardless of the state of the others. Those are all set in the DVR and have nothing to do with the cameras. FPS isn't a factor in getting "crisp, recognizable" images - that's mainly a factor of resolution and shutter speed (and of course, lens quality and how well it's focused). FPS - also called IPS (images per second) just defines how many times it captures the scene. Take a look at this demo to see the difference: http://www.panasonic.com/business/security/demos/PSS-recording-rates.html

Jason, all you need to do is connect the cameras in parallel. Most manufacturers' installation diagrams show them all in a line, with the wire jumping from camera to camera, but they're still wired in parallel. Wiring all of them straight back the DVR accomplishes the same thing, electrically.

Yeah, those are generally really good cameras... I wouldn't expect the camera MODEL to be a problem, but it looks like the specific UNIT may be bad. I've seen interlacing issues with moving objects, and it didn't look like that - the object was "tearing" horizontally, not showin a diagonally-mirrored double-image. I'd almost think that the IR cut filter over the sensor is damaged or partly out of place... if one end was lifted up a little bit, so it's on an angle to the sensor instead of parallel to it, it could cause some refraction and/or reflection that would account for that particular distortion, as well as explain why it's worse to one side of the image.

Damn, that musta been the old Microchannel machines. Did full plug'n'play long before anyone thought of PCI. Some brilliant stuff for its time, MCA. Probably had advantages they still haven't managed to duplicate with PCIe. IBM always makes me shake my head... come up with some great new technology, then find ways to slowly kill it. Was the same with OS/2 when we were promoting and supporting that back in the day. I still have an OS/2 server running my mail here (old Dell PowerEdge dual-PII/233 box with a bunch of SCSI-UW drives), there are concepts in there that Microsoft STILL hasn't figured out in Windows, 15 years later. Big Blue eventually pawned OS/2 maintenance and support off to a third party; it's still floating around out there (as "eComm Station" or something like that - they were aligning it as a server/thin-client system at that point)... IBM has all but orphaned it, but of course, they won't let go of the leash, so nobody else can actually take the technology and RUN with it. Bastards. Anyway... What you might want to do, is look for another PTZ head (on eBay or whatever) that has dead motors but a working controller. Might be able to find something cheap that way. BTW, I'm sending you a PM... don't forget to check it