Soundy

Did you replace the lens along with the camera, or move the original lens to the new camera? If the latter, is it an auto-iris lens? It's not uncommon for the iris driver to fail and the iris simply stays closed...

^Not to mention, WiFi is limited to, what, 300' outdoor line-of-sight?

Have you tried switching the polarity on the baluns?

Personally, I'd probably go with a separate power supply for each floor, to reduce the losses (especially for the furthest floor from the DVR). Since the cable is likely to route through or near each floor's electrical or phone room, that gives you a good spot to locate each floor's supply. While one uninterrupted run from camera to DVR is generally preferred, of course, having a separate PSU per floor negates running siamese cable for the entire length of each run, and probably means a junction box for the video feeds in the same vicinity as the PSU. Frankly, what I'd probably do in this instance, is use Cat-5e and baluns all the way. You can run one Cat-5 to each camera, terminate to RJ-45s, and use baluns like these to easily split out the video and power. You can then use four of these and run all 16 video feeds over four Cat-5e from each floor to the DVR locations, then use four of these (per DVR) to split that back out into the 16 video signals into the DVR. That's the "clean" way... or you could just take the video pair from each camera and splice it to a pair of the floor-to-floor "drop" runs (use a punchdown block to allow for better future flexibility), and split it out at the DVR same as above, or just using four individual baluns.

Why not?

That's what I'm saying, what's the point of recording 40 seconds of nothing happening leading up to something triggering the motion? In my experience, 3-5s is usually sufficient, and 10s is the MOST I've ever needed to set.

Hey bud, clear your browser cache... Do that every day A little paranoid are we? Actually I can see that being good, but man would I hate having to enter all those user/passwords every day. I must go to 9 or 10 sites a day that require login. Those are stored in cookies, not cache.

Cluttered cache isn't the cause of "500 Server Error" messages and various intermittent MySQL errors. Unfortunately, if it is a damaged/corrupted database, a scan-and-repair is the only real fix... just because it stopped happening doesn't mean it's fixed itself, it just means you're not running into the damaged table(s). But like I say, it's an easy fix from a command line, if anyone in charge wants to know...

I've noticed, in addition to the slowdown, some various database and server errors. Having run phpBB-based websites and other MySQL-based sites, I'd hazard a guess that the site's phpBB database is corrupted, which would easily account for both issues. A database scan-and-repair will probably fix it... it's really easy with MySQL, especially if one has command-line access

My coworker bought an Archos too, and the thing is amazing. Another bonus with it, the screen is very high resolution, probably twice that of most cheap portable DVD players, so fine-tuning camera focus is even easier. His has built-in WiFi as well, so if I plug a cheap wireless router into the network with IP cameras, it can link to them through that too Oh, and the other advantage to having an Archos or a DVD player or anything else of the sort: it can double as a signal source for troubleshooting. Bad picture on a line, is it the line or the camera? Unplug the camera, plug in the DVD's output, and see what kind of picture you get at the other end. I made a DVD with a variety of standard TV/NTSC test patters that's I've also used for setting up projectors.

(server glitch, double post)

Actually, at that size, you're more likely to be getting into "TV" range, rather than "monitor"... and with that, the design parameters generally change, along with a reduced resolution. TVs that size are usually designed for 720p HDTV display, and so won't have a resolution over 1366x768. With a TV, higher resolution isn't necessary, and a LOT of cost is saved by keeping it low: around $600 at my regular retailer for a 32" 720p Viewsonic, vs. $3800(!!) for a Samsung 30" that does 2560x1600(!!!), or $4100-$4600(!!!!!) for a 30" LaCie with the same resolution. Also, something that size, at TV resolutions, is not intended to be watched up close like a computer monitor. 1366x768, watched from 2-3 feet away, is going to look pixelated no matter what you do. Try this with a 20" LCD for about 1/10th the price and the customer will probably be a lot happier... But of course, this is just guessing until we know the actual model of the display in question...

The thing is, the CARD is only a small part of the equation. As Erron indicates, how the SOFTWARE deals with the video and the streaming of it has a much bigger effect on the overall quality, especially when viewing over the internet.

On ANY computer with an LCD monitor, you want the desktop display resolution to match the native resolution of that LCD. In Windows, that usually means installing the driver that's included with the monitor, or making sure the monitor type is set as "Plug and Play", and then maxing out the resolution setting. That will give the sharpest, clearest display that the computer and monitor are capable of... without that, it doesn't matter what you do with the remote view, it's not going to be at its best possible quality. Beyond that, it really depends on the system and the remote client. Some will increase the compression of video streamed to the remote, which makes more efficient use of bandwidth while reducing the quality somewhat. Some will want to scale the video, others will default to showing it at its "real" resolution. Anytime you blow up the video resolution beyond its actual recorded resolution, it will start to look fuzzier/blockier/pixelated. If you're recorded at CIF (352x240), for example, viewing at anything larger than that will show off just how low the resolution really is.

As the old saying goes, "Good, fast, and cheap... pick two."

I was looking at devices to do this and all the ones I found ran in the $75+ range... one actually showed a schematic, and I couldn't believe how ridiculously simple the design was, nevermind how they could get away with charging $75 for $5 worth of parts.

No... you have to make sure the computer's display resolution matches the monitor's native (maximum) resolution. If your computer display is set to a 4:3 resolution (1024x768, 1280x1024, etc.) and you're running a widescreen monitor that supports a higher resolution (1680x1050, etc.), you'll end up with a fuzzy, distorted picture of EVERYTHING displayed on the computer.

Realistically, the only way you're going to get wireless covering that sort of range is with the aid of external antennae on both the camera(s) and the DVR end... which is going to mean disturbing the walls to pull in cables. All you'll really be able to avoid is needing to dig up the street/parking lot to bury cables. Any wireless that will cover that sort of range with any degree of reliability is going to be spendy as well. Frankly, there's no way I can think of get away with this cleanly AND cheaply - they're just going to have to accept that as a consequence of their own shortsightedness.

"Better" is relative in this case as well... "better" for what? Better for fine dtail recognition. [snip] First is the Canon 400F5.6 telephoto: Now the 20f2.8 wide angle: [snip] Huge difference, no? The corners of the wide angle (the curves on the right) have nearly no resolving power, compared to the telephoto which does not suffer hardly at all as you go from center (left of the curve) to the corner (right of the curve) of the lens. So if you have the bad guy in the corner of your camera, you better pray you are not using a wide angle like above! Okay, I have a background in photography as well, so I actually understand all that... however, there's a bigger concern: field of view. If the bad guy is in the corner of your 20mm lens, he's gonna be about 50 degrees outside your 400mm lens's FOV, which makes comparisons of resolving power rather moot By the same token, if the bad guy is within the FOV of the 400mm lens, he'll be so tiny in the center of the 20mm lens that CA and resolving ability also become moot, because he'll be little more than a 2x5 pixel speck on most cameras' sensors, and even a 5MP camera wouldn't have enough detail to make him recognizable And again, given the relatively low resolution (we're not talking a 1Ds mkIII here), the camera sensor's resolving power becomes a significant limiting factor, moreso than the lens for the most part. True although one has to be aware of the design period. Once more, Canon's newer zooms tend to outperform their older prime (fixed focal) lenses in most criteria. For this reason, most pros use zooms without hesitation. Such was not the case 30 years ago when I got into photography. The key word, though, is 'comparable'. I wouldn't use my EF-50 f/1.8 as an example of prime vs. zoom while attempting to stack it up against some of the latest L-series glass. And while it's not a factor most of the time with CCTV lenses, when it comes to pro photographers, primes have another distinct advantage over zooms: less complexity and less glass means less weight (important if you have to be mobile, such as shooting sports), and the ability to put a much faster (ie. larger maximum aperture, for those not familiar with the term) lens in the same size package. And of course, less glass also generally equates to less cost. The trade-off, naturally, is versatility... why carry three lenses when you can carry just one? Depends on the needs of the job at hand. Either way, the question remains, "better for what"? An 80mm CCTV lens might give me better clarity than a 3.5mm lens, but that doesn't do me much good if I need to view an entire parking lot

Ah, the same old FUD being trotted out. High cost? Cost is relative. An analog Panasonic SDIII (like a WV-CP484) has a high cost compared to your average all-plastic, generic-1/4"-board-based off-shore dome. As with anything else, you get what you pay for. Do you try to dissuade customers from an SDIII in favor of a cheap dome because of the "high cost"? Special housings? Special how? I've regularly put IQ megapixel cams into the same Pelco EH-3508/3512 housings I've been using for years for analog cameras... likewise into DF-5/DF-8 ceiling dome enclosures. Network problems? Sure, and analog CCTV is susceptible to all manner of wiring problems too - bad cable, bad terminations, bad connectors, bad runs, damaged wiring, electrical interference... nothing is perfect. The argument is a non-starter. Internet problems? Irrelevant unless you actually need an internet connection to your system, and those problems would affect analog-camera-based systems that you're using over an internet connection too. Lens problems? Like what? Cheap-ass lenses with poor optics? See above regarding cost. Despite the proliferation of "digital/megapixel" lenses, the technology is really no different. Lens optical technology is hundreds of years old. The transmission media and image/video format has no relation to it whatsoever. "Better" is relative in this case as well... "better" for what? A more relevant tidbit from the photography world is that a fixed-focal-length lens tends to give better images than a comparable varifocal lens, because the optics are less complex and thus have les negative effect on the image.

Part of the problem is that the camera is switching to B&W for night mode, and your DVR's video codec doesn't handle B&W well, so you're ending up with color artifacts. IF the DVR allows it, try dialing down the "saturation" control for that camera and see if it clears up (the catch is, daytime shots will then show up as B&W).

What settings will "work best for you" will depend on exactly what you need the camera to do. I'd think you'd want to start with the quality vs. framerate setting (switch to the former). The higher the bitrate, the less (lossy) compression is being applied, so increasing that will help quality as well. You can reduce the bandwidth needed by lowering the frame rate - even cutting it to 15fps you probably won't notice the difference unless you're capturing race cars or something fast-moving. Disable the digital zoom, as well - all it's doing is cropping an area at the center of the shot and blowing it up to fill the frame. As with digital pocket cameras, digital zoom is a gimmick at best.

Pelco sells a version of that camera with a 2.9mm lens - on a 1/3" CCD, that's pretty wide-angle. If you bought the camera new, you might want to check with your supplier about trading it for the version with the wider lens... if used, check with Pelco or your dealer, you may be able to buy the lens separately.

Just a random thought, is there some particular reason for running a *FORTY SECOND* buffer? At that point you may as well just set for constant record...

Every manufacturer has traditionally had the occasional bad batch or series of drives, be it WD's particular model of 500s or Seagate's line of 1TB units... while other drives of theirs may work flawlessly. I wouldn't dismiss either brand based solely on that. Where the problem really arises is how each them deals with their issues and stands behind their products, and more importantly, their customers. I haven't had significantly more or worse problems with either brand failing (yet.... touch wood), but while I've not yet had a problem returning failed Seagate drives for RMA, Western Digital has been a whole other story. Their warranty status check is misleading at best - even six-month-old drives have shown up as "in limited warranty" and offered me a "customer loyalty" option of "upgrading" to new, larger drives... at about the same as I'd pay for those drives retail. Gee, thanks, WD. Yet on contacting them to ask about it, I've been told that the drives were still fully covered and eligible for direct warranty replacement... hmmmmm.