MaxIcon

Here's what the rest of the connector looks like. As BW says, these are for bulk cable, and have to be installed before crimping the connector. Unassembled - the white rings are soft silicone rubber or similar. The large skinny ring seals between the two halves, and the small tall ring slips over the unterminated cable, and compresses against the loose half and the cable. It looks like it should be effective. I didn't use mine because I'm using pre-made cables. Assembled, ready to be screwed down and connected:

Another test is to connect it to 12V instead of POE, in case the POE power board is causing problems. I've never had this make a difference when troubleshooting a camera, but I always try it on persistent problems.

One's probably the IR filter, as Zikronix said. Sometimes, the IR filter is between the lens and the sensor, other times (like the Dahua 3300C), it's built into the lens and looks like it should be a zoom or focus control. If the lens is auto-iris, this would also require a set of wires and a motor. I don't see motorized zoom or focus in the description, so it's unlikely you have them.

This is pretty much the standard for the inexpensive mid-range Chinese cameras like this. Accessories are often unavailable, even if you can find them on their website, and nobody sells spare lenses, even when they sell the same model with the different lens installed. To get better service, you have to bump up to the better name brands, which usually means spending more for the same performance. The support costs them more, of course.

I just took a look at my HFW2100 and noticed that 12 of the 30 LEDs are out, and one more is dim. This is pretty common on bottom-end cameras, but not what I'd expect from the midrange, and is a bad sign for Dahua quality. The usual cause of this is overdriving the LEDs to get more output, and once they start overheating, the output drops until they die. We've just had the first few pretty warm days of the season. I'll take a look inside to see what's up, but this is not a good indicator for Dahua quality. Anybody else seeing their LEDs dropping out?

After installing a couple of Hik mini-bullets, I pulled my 2100 to check the LEDs. By this time, there were only 6 left on. Turns out they're connected in 5 strings of 6 LEDs, so when one LED opens up, the other 5 in that string go as well. At first I thought I'd replace the board, but finding a match turns out to be non-trivial, so I found the bad LEDs and swapped them out, choosing some with matching parameters from digikey. It's back working again, except one string (I apparently missed one bad LED in the testing), and seems to be holding up well. I've got a detailed post on the process here: http://www.cam-it.org/index.php?topic=4946.0 Nobody else seems to be having trouble with theirs, so it looks like I got a lemon IR board. We'll see how long the rest of the LEDs hold up.

Yes, the hardware is similar, but the software can make all the difference. I'm not sure if all these cameras used the canned TI h.264 compression library, but Dahua's image quality has those fuzzy shadow problems that the others don't, so they must be doing something different. I haven't tested the ESC cams myself, but based on the samples posted over at cam-it, there are 2 difference between the ESC and the Dahua or Hik. - ESC vs Dahua: No soft fuzzy shadows when there's bright light on the ESC. This is a big problem on both of my Dahuas. - ESC vs Hik: The ESC appears to have better low light image with less noise. Not all the posts have all the image setting details, though, so this one's not as sure a thing. Firmware support can be an issue as well with the small vendors. My Hiks are ESC branded, but I can download the firmware directly from Hik, so no problems. One user of an ESC branded Dahua reported he was unable to get a copy of the firmware (normal tech support on Dahua gear), and the vendor wanted the camera sent back or upgraded by them online.

Post the codes they gave you and the info you gave them, and we might be able to figure it out. The Dahua date-based reset system was posted here a while back, and was pretty straightforward.

Interesting. My new Hiks, dated 06/2013 on the box, are labeled V5.0.0_130412 on the box and in the firmware. The firmware lists Encoding Version V4.0 build 130411. So, is 5.0.0 130607 the same firmware, with just updates for the vendor info, or is the firmware functionally different? Usually, functional differences would be reflected in a version change, like 5.0.1. Dahua did something similar - changing the date, but not the revision number, for different vendor versions.

As others have said, it really depends on the VMS you plan on using. Some, like Blue Iris, can require a lot of resources. Others that are only recorders/servers, and not clients, can use very low CPU for a lot of cams. My main experience is with Blue Iris, and here's what you need to take into account: - Camera resolution - higher resolution increases CPU load dramatically - Frame rate - higher frame rate increases CPU load dramatically - Number of cameras doesn't matter as much as total MP and fps loads - Motion detect and pre-trigger frames increase CPU load a medium amount - Connecting remote clients increases CPU load a medium amount - Bit rate doesn't have a huge effect - Recording can increase CPU load a good bit, if not using direct-to-disk - Use of direct-to-disk recording reduces CPU load, sometimes dramatically, but removes some options like overlays This will be different for each system, but in general, it's important to specify number of cams, frame rate, and resolution as a starter.

I have a couple of US 3MP bullet cameras, sold by a vendor we can't name here and with their specific model name, and they show English days of the week on the OSD. They came with 5.0.0 installed.

You can figure out a bit more by watching your POE switch LEDs when the resets happen. On startup, the Dahuas should light up the POE indicator right away, and the Link indicator will blink a few times, pause, blink a few times, and keep doing that for about a minute. Then it starts blinking constantly and the camera's online. If your switch shows it doing this when you lose your signal, it may mean your camera's resetting.

One more data point - I updated my HFW2100N to the 2013-05-17 firmware last night, and it went fine. Many of the improvements are there for the 2100, but no easy-focus, no 90 degree rotation, and no SD card options (of course). ETA: I played around a bit with EasyFocus on the 3300C, and found that it's the number's peak value, not how close they are, that indicates focus. When I de-focused on purpose, the numbers were both low, but very close to each other. Focusing increased the numbers. In my case, great focus gave numbers in the 6000 range, decent focus in the 5800 range, and poor focus in the 3000 range. Looks like it may be using the fuzzier portion in the shot above for these numbers, rather than the better-focused background, but I haven't tested that yet. I moved a chair with webbing on it in front of the camera, so it saw the yard through a mesh, and got it as high as 10,000, so it looks like high contrast may trick it.

You're right, the Costco version looks like a mini version of the Hik linked. I don't see the smaller version on the Hik pages, but they're pretty clearly related. The Hik's got a junction box and is a bit longer. The Costco spec shows 2.8 – 12 mm, but no power consumption that I can see. Here's the Costco vari-focal: and the Hik vari-focal:

If someone's actually done the swap, that's your best bet. Let's see if anyone responds with details. The problem with swapping M12 lenses is that there are few standards having to do with lens fit and focus, unlike CS mount lenses, so whether you get one that works with your camera is a shot in the dark. A 3mm lens from vendor A may focus in a completely different position than a lens with the same specs from vendor B, and will likely give you a different field of view in the process. The most common problem on IR cams is that the lens needs to be screwed in too far to focus, and hits the IR filter housing. I tested a half-dozen lenses of various sizes last weekend, and 3 of them hit the filter housing on the camera I was checking. Unless they specify the back focus length on the lens (few vendors do this), you won't know for sure, and even then you'd need to know the clearance of your IR filter. If you don't have IR, there's a lot better chance that any random lens will fit. Anyway, here's what you want to look for, in general: - IR corrected if you've got a true day/night camera (not IR filtered - that's a permanent IR filter mounted on the lens). - MP rated (most vendors don't specify how many MP, but it's a start). - Right size lens for your sensor (1/3" lens for a 1/3" sensor). This changes the FOV a lot if you don't match them up. If they specify glass lens elements, that's usually better than plastic lenses, and an aluminum body is better than a plastic body. All these things can make the difference between a $3 lens and a $30 lens.

The Swann appears to be this camera: http://www.hikvision.com/en/us/Products_show.asp?id=7673 2.7 ~ 9mm @F1.6, angle of view 101° ~ 30.4°. They don't match the spec you mention, but this isn't uncommon. This also shows up as a 12W max camera, which is a good bit higher than many recent models. It's possible this is the max load when the motorized focus is moving, but would require testing to know for sure.

My 2100 came in at 63 degrees, and calculates to 3.9mm. It's got a 3.6mm lens, and the Dahua website only gives an FOV spec for the 6mm lens.

IIS is a common user of port 80. If you put in that PC's IP address in another PC's browser, you'd get the generic IIS page if it's turned on. If that's the case, here's how you disable it: http://forums.iis.net/p/1150139/1872824.aspx

The usual reason for wanting this is when there's motion after the trigger that isn't enough to continue triggering recording. I see this at night, when cars park on the street. The motion detect catches them, but if they're in a poorly lit area, the low contrast movement of people around the car won't trigger more motion sense. By setting the break time for 30 seconds or more, I get more of the action. The other option is to increase the sensitivity, but that can cause too many false alarms. Likewise, people or animals moving away from the camera into the dark; you can still see them, but the motion trigger doesn't see them. Sounds like there might be a bug in the software, if it doesn't continue recording after the motion detect is over, or possibly this isn't what the setting does at all. You could ask Dahua tech support, but I think you're getting more response here than you would from them!

The real problem is that the relationship between lens size, sensor size, and FOV is not very predictable with M12 lenses, due to a lack of standards. CS and C mount lenses don't have this problem as much. I measured the actual FOV on the Hik in degrees, and plugged the number into the Theia FOV calculator: 1080p: 68 degrees, 3.6mm equivalent 3MP: 61 degrees, 4mm equivalent Other FOV calculators give different numbers, as low as 3.2mm, but what really counts is the actual FOV, of course. The manufacturers need to be doing a better job of specifying the actual numbers, not calculated ones, and for every resolution as well.

I finally got around to modifying my lens and mounting it up on my Arecont AV5105DN 5MP camera. Removing the auto-iris was simple - 3 screws to remove the shroud, 2 screws to remove the iris assembly, then put the shroud back on. It's fully open, but that's how I run most of my lenses. Anyway, the quality is decent, but not great. The colors are a bit dull, and it definitely doesn't seem to be 5MP resolution. My Arecont MPL4-10 has a charper, clearer image and tends to cost around $100 or so. I'd guess that they got a batch of these that the customers were unhappy with, possibly when they switched from Fujinon to the Korean manufacturer, and they dumped them on the surplus market. Well, it was worth a try!

The newer version of the firmware has online help, which is not always useful, but this is what it says for this screen: It sounds like this is the break time after the motion trigger stops; that is, this is how long after a motion detect event ends the camera will record. The shortest clip would then be motion detect time plus 10 seconds.

Well, it depends. In cam-world, you don't always get what you pay for, but you rarely get what you don't pay for. Cheap Chinese gear tends to have poor quality control, and the user is often the first QC inspector. Dust inside the lenses, elements mounted crooked, poor quality optical components are all what you get with cheap lenses. With CS mount lenses, you can get $10 lenses, and $300 lenses with the exact same specs. Good optics cost more, and reliable manufacturing costs more. With M12 lenses, there are no moving parts (if they're not varifocal), and that's part of the problem (focus is achieved by screwing the lens in and out), as well as part of the reason for cheapness. So, I wouldn't say a $10 lens and a $50 lens are the same without having tested them. The fact that these guys publish actual specs is worth something, since it means it'll work the first time if you do your homework. Whether their lens quality is better than DealExtreme would take testing, but whether to choose a $10 DX lens vs a $40 Marshall lens for an important application is a personal choice.

Do you or somebody uses FTP recording? If so, it works with the 2013-05-17 firmware? I haven't used it before, but just set up an FTP server and enabled motion detect on the camera, and it seems to be talking. I'm not sure what the original bug is (or what's normal), but it's chattering away, pinging the server constantly, and when it sees motion it saves a bunch of 0 byte .dav_ and small .idx_ files, but no video or snapshots. If that's the bug, yes, it's still there.

The Swann will be identical to the Hik if the flash takes correctly, unless they've changed something recently. When I first flashed a Swann to 5.0.0, it was missing items in the white balance area, as well as other stuff. At one point, after both the camera and PC had been turned off, it prompted me to reload the webcomponents file, and everything was OK after that. Try powering down the camera and restarting the PC. If that doesn't work, delete the webcomponents folder and it should reload. For the bricked issue, Hik has a procedure for this. Someone else tried it without much success, but if you're bricked, it's worth checking out. Here are the files and instructions: http://www.hikvisioneurope.com/portal/index.php?dir=Technical%20Materials/Special%20Tools/TFTP-Auto-Update/