MaxIcon

Getting good images through glass is tricky. It's not a camera design problem, but an optics problem. IR reflects from the glass and overwhelms the outside light, so that's no good unless you disable it, and if there's light on the inside, it'll reflect off the glass as well unless the room is kept dark. I used to view through glass on vacuum equipment by making a shroud that went from the camera up to the glass, and this kept the reflections out. It's easy enough to test with poster board or cardboard and tape. You still can't use IR, so you'd need to have decent lighting outside, or a good low-light cam. The original analog cams probably won't have the resolution to see much detail unless you can zoom way in. IP cams, like the HD models mentioned, have much better resolution, but tend to suffer in low light without IR unless you go for the expensive models. They run on network cables, and are good for a few hundred feet, no problem. One option would be an Axis P1214e covert IP camera, which would allow you to put the sensor outside and run the wire through the window. Very inconspicuous, but a little expensive, at about $500-ish per camera. Another option would be to install a box that could go in a partially open window and extend into the room, so the camera would be inside and in shadow. Not exactly inconspicuous, though, and if someone looked closely, they'd see the camera. Likewise, a flower box in the window could be a good hiding place, but might also draw attention if someone notices the difference.

Your external IP address may have changed. Mine's been stable for years, but they could change it any time. Try connecting from a browser outside your internal network to see if that's still working. If not that, you might want to check that your port forwarding hasn't changed. I just had to delete a forwarding rule and re-create it on my uverse router.

If you google marine electrical tape or all weather electrical tape, you'll see many options. If there are reviews at Amazon, that would be a good place to start.

Yeah, that Axis looks pretty awesome, though a bit pricey. I ran an indoor dome under the eaves on my patio for years with no problems, and I'd bet that it would work fine in that protected area. They make an outdoor version, too - the P1214e - that looks like it's the same price range. If you could put that right above the door, and if they support effective backlighting or WDR settings, that could be a perfect solution. Watch out for that slippery slope of IP cams...

Yeah, it's relatively easy if you swing that way, but time-consuming, and there's a tradeoff between inexpensive and compact, plus you'd have to build the thing, so there's some time value there as well. The problem is providing 48Vdc at 15W in a field portable package. You'd either need a 48Vdc power supply that ran off of a small 6V or 12V battery, like a UPS or motorcycle battery, or a compact 48Vdc rechargeable battery pack and charger, which tend to be expensive. If you didn't mind big and heavy, 4 x 12V sealed lead-acid batteries would be cheap and easy. Unless you're doing it for the fun of it (which is a perfectly good reason), you might as well buy the $200 Pointsource.

The HFW3300C doesn't play well with Blue Iris, even set at 1080P, which makes it essentially the same as an HFW3200C (same software and hardware, just a slightly larger sensor). Mine is OK if the scene is low contrast or not brightly lit, but when the sun is bright and there's a lot of detail and contrast (like a yard with bright sun on it), I get constant camera dropouts, pauses, and resets. Dropping the bit rate to the minimum helps, but doesn't completely fix it, and causes lots of pixelation when there's a lot of movement. Pointing the camera at the sky is no problem, and leaving it in my garage is no problem. Many people report ghost images (dropped frames leaving visual blocks from previous images behind) with this cam and BI. I tried it again this weekend, and during the day, it dropped out and reset constantly. At about 5:30pm, when the sun has my yard mostly in shade, the dropouts quit, and it ran fine the rest of the night. If someone's had good luck with the HFW-3X00C cams and BI in a bright outdoor environment, I'd love to hear about it.

Yes, any IP cam with a BNC connector for setup can feed that signal to an analog DVR. It has to be always active - some can be disabled, and some are only enabled at certain times, so read the manual. It will be analog resolution, of course, and not the high resolution of the digital signal. The multiple stream feature will only send out IP streams, and can't feed an analog DVR without a decoder.

I took my HFW3300C apart this weekend, and got some pics of the sensor up close. It's still hard to tell, but the black rectangle on top could be an IR filter over the actual sensor. It should be pretty obvious if this is different from yours. It appears the IR is directly controlled by the electronics, not by the CPU and software, as it will turn on and off as soon as the camera's powered up, long before it's booted. My front glass also had some haze on the inside surface, which took a little effort to clean off. If there's residue on the glass that's IR reflective, it could cause flickering. The good news is that it will work fine with the IR board disconnected, which is different from cameras that use the IR board sensor to control the day/night transition (like the Vivotek 1MPs). It's the only way I know of to disable it, for those who want to use external IR.

Could be you got a bad lens. The QC for these inexpensive lenses is not the best (part of how they keep the cost down), and often the customer provides the basic QC when they install it. You can roughly check the back focal length by pointing the lens at something bright a little ways off, watching the light coming out the back, and moving a piece of paper back and forth until the image focuses on it. A light fixture works well, or the sun shining through trees can work better for distance. Since it focuses up close and won't focus at a distance, you could also try holding the lens slightly away from the threads to see if you get better focus. This takes a steady hand and a dark area, with something lit that you can point at. If you can focus further out by holding the lens a little further away from the camera, it means your focal length is too long, and once you start screwing it in, your distance images are focusing behind the sensor. Ideally, you want the focal point in front of the sensor when the threads first engage, so that by turning it further in, you move the focal point back onto the sensor.

OK, there's always confusion on this topic. Actual TVL is a combination of 3 or 4 things: - The sensor - The lens - The camera electronics and processing/compression - For IP cams, the decompression and viewing software In addition, the display device can reduce the apparent resolution of the camera - a fuzzy monitor will not show a high TVL device output well - and the recorded resolution will usually be different from the live view resolution. Live view is what you care about, since that's the camera's output. The weakest link defines the best TVL you can get. A junk lens will give a $1000 camera a poor TVL result. Now, to measure actual TVL, you need to be able to focus on an ISO12233 chart that fills the screen vertically, then find where your resolution becomes unreadable. This is easier said than done, but this is the only real test. So, measuring the actual TVL is tricky, but measuring relative TVL is easy. Go to this site: http://www.graphics.cornell.edu/~westin/misc/res-chart.html and download this PDF: http://www.graphics.cornell.edu/~westin/misc/ISO_12233-reschart.pdf Print it on the largest sheet of paper you can, on the best printer you can find (laser printers are better than inkjets, in general), and mount it where you can view it with both cams. View it through both cams with the paper taking up the same amount of the screen for both (easy if they have the same lens size, takes moving it if they don't). Focus the lens for the best resolution of the lines, then figure out where the lines become blurred, and that's your relative TVL for that cam. You can then test things like switching lenses between the cams to see how much effect the lens has. Here's a good overview of TVL measurement: http://www.acti.com/getfile/KnowledgeBase_UploadFile/TVL-The_True_Measurement_of_Video_Quality_20101027_002.pdf

Yeah, you're really just paying for the equivalent of electronic pan-tilt across the sensor. It was probably an innovative idea at the time, but unless you needed these specific capabilities, it's definitely too much for what you get.

The wider angle will give you more of the path, but you'll lose resolution overall, and since you'll be looking at faces further away, recognition/identification will be more difficult. More MP in an IP cam would fix this, but this is a basic trade-off. This is a tricky spot, due to the long, narrow entrance to the door, and the window above the door limits how far down you can put a camera. If you could mount one between the door and window, or just to the right of the top of the door frame, that would be ideal from a face capture perspective, but it's not so aesthetic or wife-friendly. A compact dome would be the most friendly form factor. If your camera doesn't have IR, you'll also have the problem that someone standing close to the door will have their face in partial shadow at night, since the light is a few feet back. You can get pinhole cams that look like viewports or doorbell buttons, but these typically don't have very good low light performance due to the tiny lens, and I'm not sure there are any IP cam versions.

I run both internal IR and motion detect floodlights in some areas, and I get a lot of false alarms, mostly due to the cats (and the occasional possum or racoon) triggering the lights. I'll get a detect when the lights go on, and another when the lights go off. How much trouble this is depends on the software you use for recording. I've gotten good at spotting which ones are routine and which are something different. Another option is to increase the motion time it takes to trigger motion detect. If you go long enough for it to recover from the lights turning on and off, you'll get fewer alarms, but may miss shorter motion events. Whether you miss anything before the motion event will depend on how many pre-trigger frames you capture.

I took a look at my HFW3300C and HFW2100, and the sensor is a tiny chip on the bottom of the IR board. It's hard to tell if it's got an IR filter over it, as I didn't have time to take anything apart, but it would need one if it were IR sensitve, since you'd get quite a lot of reflection from the back of the glass. If you can compare the boards easily, you might be able to tell if there's a difference between the sensors.

I've been interested in this NVR, but it's a bit limited in how many MP it will handle (not surprising, given the price). Here's what the manual says about MP limits: 1.3 MP with H.264/MPEG4/ MJPEG on the first 4 channels and 2.3MP with MJPEG on the 1st channel for IP cam connection. This implies that it won't run H.264 on 2MP cams at all, but it's not super clear about that.

Yes, sounds like the IR is reflecting back into the sensor, causing an oscillation. The Dahua doesn't use the lens signal to turn the IR on and off, just the sensor on the IR board, so loss of lens isolation shouldn't cause flickering IR. The sensor for the IR responds very quickly, while the through-lens response for light levels is quite slow. It sounds like either the sensitivity of the sensor has changed, causing it to turn on and off with the IR, or there may be an IR filter that goes over the sensor that has fallen off. This would make sense, if the sensor sees IR at all, as you only want it to respond to white light, not IR, to prevent this type of problem. CdS sensors aren't very sensitive to IR, while Si sensors are, so it'll depend on what kind they use. I haven't looked closely at mine to see if there's a filter or not. Sounds like another weekend project.

I've been buying outdoor bullets for some time now, and have a pretty good grip on that market. I now need to find an indoor box cam for general purpose troubleshooting in an industrial environment, and am out of touch with this market. This cam would be placed wherever needed for short periods, mounted to a tripod, clamp mount, whatever. It'll generally be set up and recorded with a laptop, though edge recording would be handy if the laptop can't be left in place. It won't always have network access. Need: $300-ish would be ideal 720p or better, 30 fps Varifocal lens or CS mount No IR, or able to shut it off Standard screw mount for tripod Good low-light sensitivity User-friendly recording software Nice to have: Dual power - POE and PS SD card recording Compact size I like the looks and features of the Acti KCM-5211 and the Axis P1344, but they're a bit out of our price range. Any suggestions?

The Dahuas, in particular, are problematic with external lighting. Both my 1.3MP and my 3MP bullets are very slow in responding to big light changes, often taking 5-10 seconds to get the exposure back under control. It's worse when the lights go out, as the IR filter is in. The IR LEDS go on immediately, but it takes 3-5 seconds for it to decide to switch the filter out, then another 4-6 seconds for it to adjust the exposure. The transitional images are badly washed out for those times. When lights go on, it's not so bad, and the transitional images are much clearer, but still not optimal. In contrast, my Vivotek 1MPs have the exposure tweaked in less than a second, going either way, and lose very little of the video.

Thanks - The E12 looks like a pretty good bet, and gets 30 fps, which we need, when dropped to 1080p. No info on the sensor size in any of their docs, so I sent them an email. I'm guessing they don't have great low-light, but that's what you get in this price range. Does Acti supply recording software? If not, we'll probably use Blue Iris, since I'd have to support it, and it's what I'm most familiar with.

There's a drop-down where you specify which lens you want. This doesn't specify IR corrected, which can matter. Sometimes they just don't bother putting the details in (or don't know). If the lens isn't IR corrected, your focal point will shift between IR and white light, so one or the other will be in focus, but not both. If it is IR corrected, both will be in focus. I'd keep looking for one that specifies IR (or ask them and see what they say), but it may be worth a gamble.

This is a fascinating camera. I've never heard of it, but their goal is to have onboard video analytics with digital zoom and PIP tracking, so they use the 3MP sensor in place of a motorized PTZ solution, and put out 2 4cif streams - one full field of view, downsampled from the 3MP sensor, and one zoomed PIP view that the analytics processes. http://www.dvtel.com/UserFiles/File/ioimage/ioicam/en-ioicam_mmp100dn-datasheet.pdf I'm guessing it's an older design (I see articles about it dated 2008), since it doesn't support h.264, which virtually all newer cam designs do. It sounds like this was a bridge solution for analog customers looking for digital PTZ solutions but not needing full MP output. Looks like the price was in the $2-3k range back in the day. It sounds very innovative and intelligent (and expensive), but it's a bit of a one-trick pony, and I wouldn't be surprised if it's not very well supported by 3rd party software. The vendor would have to tell you if it can have the full 3MP output enabled.

One thing to be aware of is that many no-name cams aren't supported by the popular software packages, and the software that comes with them is usually part of the problem, not part of the solution. If the manual includes the stream settings, you'll be able to make it work with some packages, but many no-name cams don't include this important data. Here's what I'd check before buying something like this: - Do any popular software packages support this brand and model? - Are software updates available online? - Is the manual available online (and is it in a language you understand)? - Does the manual include the stream settings? I'm a big fan of experimenting, but I hate to throw money away.

Last time I had to set up a camera in a tricky position, I used a tripod (I have a very tall one) and a ladder to roughly position the camera so I could see how the view would look. It saved me quite a bit of time and hole patches. An extension ladder leaned up against your corners would allow you to set up the camera and roughly position it. I used a towel and bungee cords to hold the cam down while shifting the position and having it stay. You want to make sure everything's stable, of course, so the camera doesn't come crashing down while you're testing.

Do you have the model info on the camera? Many no-name cams listed on ebay claim to have high MP sensors but only record at limited resolutions. I have no idea how they come up with these numbers, but it's always a warning sign.

Don't know the Easycam, but if it's the same as another brand and you can find that, you can look here to see if it's included: http://www.soleratec.com/rtsp/rtsp_listing.php