MaxIcon

A recent update to Blue Iris enables Geofencing, which uses your phone's location to enable or disable profiles. I don't believe it's working yet, but should be in the near future. How effective it will be is still unknown, but it's a neat concept. You can follow it at the BI community forum at cam-it.org - search on geofencing.

Maybe not much help, but the Aver software for the NV series of hybrid cards does this. I have an NV-6480 as a backup system that records 24x7 with motion detect, and the timeline shows different colors when there's motion. The cards are expensive new but come up cheap used now and again, and there's no need for them for a full IP system, so they essentially act as a dongle for the software. I believe Aver used to sell a stand-alone version of the software, but I've never seen it on their web site.

I set up something like this for my wife in her home office, and after trying a few things like old laptops, Blue Iris client viewer, and such, I ended up with a tablet. I got a Galaxy Tab 2 10" tablet (which wasn't too expensive since it's last gen, but has decent resolution), and a cradle. I loaded IP Cam Viewer Free (which supports 6 cams), and she leaves it on all the time while she's working so it's easy to keep an eye on. Even when it's off, a button push and swipe brings it back up within a few seconds. Power use is pretty low. The only downside is that it sometimes shuts down and goes back to the ip cam config screen. I haven't spent any time on that yet.

Yes, assuming Dahua's spec is correct (which is always a big assumption with them), they have it classed incorrectly. Since the Zyxel's unmanaged, it's not easy to determine actual power draw without using an inline device to read it. For OP, since it worked for a while, it appears something changed recently: - Cable resistance increase, which is possible if there's a mediocre crimp with a lot of current flowing, or if the cable is copper clad aluminum. - Higher current draw; I've had one Dahua overload, then kill the IR LEDs. This would be easy to check by looking at the LEDs in low light to see if some are brighter than the others. Other internal problems could cause this as well. - Switch not putting out as much as it used to; Zyxel's fine for inexpensive gear, but inexpensive gear tends to not be as reliable as the big boys, and it would be running right at the edge. If it were mine, I'd probably re-crimp the cable ends, because that's fast and easy. Still, it's close to or over the edge, so a higher power supply would be the long term fix.

Or take 10 seconds to learn about PoE and don't waste money on cat6. You can't power a class 4 device with a class 3 PSE. This is basic engineering, not POE. The 12.95W spec is worst-case, and is due to wire resistance at max power load. The extra power is dissipated along the cable due to current and cable resistance. Lower resistance wire and shorter runs will decrease the total cable resistance and increase the power available at the end of the cable.

This limitation is due to power drop along the cable. With a short cable or a larger gauge wire, you'll get more power at the device for a given current. This is a good reason to use cat6, even though few cams run over 100Mbps, since the wire gauge is slightly larger. Presumably, this minimum power delivery rating is based on maximum spec cable length and minimum spec wire gauge.

That sounds like your minimum exposure is set too low, not an IR LED problem. I've run a number of cams in BW with the IR LEDs on during the day, and they expose just fine. Daylight is far stronger than the IR LED output. I believe these are Dahua cams, right? When I upgraded one of my Dahuas, this happened to it. I had to go in to the image settings page and reset the exposure times to get rid of the white-out.

Not sure about the Dahuas, but several users of the Hik 2032 cams reported that they worked fine through the polar vortex cold snap, where it often got below -10F in some areas.

Have you tried switching IE to compatability view? I've had several cams that don't work right unless this is done. My latest Messoa NCR870 firmware upgrade shows the video if compatability view is on, but doesn't show all the menu options. Switching it off loses the video but gains all the menus... I haven't upgraded my Dahuas recently, though.

Yeah, this is the conclusion I'd expect. Not much difference, but with slightly better 1080p performance, though not enough to matter in the real world. Maybe the previous tests were with a lower res image like 720p. This takes it back to the preferred FOV and aspect ratio.

I tried it both ways with alert snapshots where a bag of topsoil was in the pic - with lettering on the bag. Zooming both pics to the same apparent bag size showed significantly better resolution in the higher-rez image. More than just a bit of the fuzzies: lettering was readable on the higher-rez image that was illegible on the other. I'd be interested in seeing some samples of that. I did a variety of resolution tests on the Hik 2032 bullet, and saw very little difference between 1080p and 3MP. It was close enough that there was no obvious winner. I'm assuming both alert snapshots use the same jpg quality setting. Maybe something's changed since then, but on cams that do direct sampling, like Dahua, the resolution is identical between the 3MP and 1080p, just the FOV is wider. That is, you still get the same number of pixels per foot in either mode, and get more pixels horizontally and vertically in the higher res. The Hik doesn't do this, but scales one of them like so: Hik 3MP going from 1080p (1920 x 1080) to 3MP (2048 x 1536): 7% increase in horizontal pixels, 20% decrease in field of view 42% increase in vertical pixels, 18% increase in vertical field of view. Compare this to the Dahua: Dahua 3MP going from 1080p (1920 x 1080) to 3MP (2048 x 1536): 7% increase in horizontal pixels, 7% increase in field of view 42% increase in vertical pixels, 42% increase in vertical field of view.

I run my Hik 2032s at 1080p for the wider coverage, and recently did some night testing at 3MP while comparing to a Dahua 4300S. I found that the 3MP Hik image had some artifacts that make me think it's the scaled image and the 1080p is the direct sampled image. They weren't major, but they were noticeable when zoomed in. I haven't looked to see if the same artifacts show up in day mode. Aside from that, it's all about the aspect ratio, as BW says.

The actual FOV will depend on both the sensor and the lens, and one 4mm 1/3" lens will give a different FOV than a 4mm 1/3" lens from a different manufacturer, depending on lens design. In addition, some cams read each pixel directly at the higher resolutions (like Dahua), while the Hik reads them directly at 1080p and up-samples the 3MP image, so the Hik gives a narrower FOV at 3MP than at 1080p. I've measured the actual FOV of both the Dahua 2100N and the Hik 2032, and they're virtually identical at maximum resolution, while the Hik gives a wider FOV at 1080p. Measurements I haven't made yet have a ? in the table. Hikvision DS-2CD2032-I Resolutions: 3MP = 2048 x 1536 1080P = 1920 x 1080 720P = 1280 x 720 OEM 4.0mm lens - Hik's original lens 3MP = 69 degrees H, ? degrees V 1080P = 77 degrees H, ? degrees V 720P = ? degrees H, ? degrees V Dahua 2100 Resolutions: 1.3MP = 1280 x 960 720P = 1280 x 720 D1 = 704 x 480 OEM 3.6mm lens - Dahua's original lens 1.3MP = 68 degrees H, 50 degrees V 720P = 68 degrees H, 38 degrees V D1 = 56 degrees H, 38 degrees V I also measured a Deal Extreme 3.6mm lens on the Dahua (I don't recommend DX lenses on MP cams): DX 3.6mm lens, from this set: http://dx.com/p/2-8mm-16mm-fixed-iris-lens-set-for-webcams-and-security-cctv-cameras-6-lens-pack-15774#.UvgOOGJLXTo 1.3MP = 75 degrees H, 55 degrees V 720P = 75 degrees H, 40 degrees V D1 = 61 degrees H, 40 degrees V

If it's got an optical drive, I'd disconnect that and re-use the port. I use an external DVD recorder for all my NVRs on the rare occasion I need one, which has the benefit of reducing power consumption. Otherwise, you can get add-on SATA cards: http://www.newegg.com/Product/ProductList.aspx?Submit=ENE&DEPA=0&Order=BESTMATCH&Description=sata+controller&N=-1&isNodeId=1

If the eave or any other item is too close, and the end of the camera is back from the eave, that can reflect IR onto the glass. If you can move the camera so that the end is just beyond the edge of the eave, that may get rid of it. You can tell if it's internal or external reflection by pointing it straight up with nothing around it. If there's still IR halo, it's internal; if it's clean, it's external.

Here's their press release, with some branding support from Hikvision. They claim the firmware is optimized to support multi-MP cam streaming and such.

To use this injector, you'd need to put in 12Vdc at one end of the cable via the DC plug, then pull it back out of the same plug at the other end of the cable and connect it to the camera's DC input with a matching cable. This requires 2 assemblies and the appropriate cables, of course. Passive POE won't power the camera directly from the network cable without modifying the camera end box and injecting 48Vdc at the other end.

POE supplies 48Vdc, and the cam converts it to the voltages it needs internally, so you can't run your mic from that. The easy solution would be to run a POE power splitter and power the mic from that. It's a bit of overkill and bulky to boot, and the splitter should be in a junction box, but that would give you the voltage you need. http://www.amazon.com/TP-LINK-TL-POE10R-Gigabit-Splitter-compliant/dp/B003CFATQK/ref=sr_1_2?s=electronics&ie=UTF8&qid=1393447831&sr=1-2&keywords=poe+splitter If you're an electronics geek, you could whip up a 48V to 12V converter, but I'm assuming you're not or you'd already know that!

Smart IR doesn't work on mine, but I'm still on 5.0.0, and they may have fixed it in the newer version. I use WDR, set at 21, though some people like it lower. Too much WDR increases noise at night a good bit. Did Hik add profile switching in 5.1? That would also help; you could optimize the two profiles and have it switch automatically.

I've found dedicated NVRs to be more stable in the long run than PC based NVRs. That said, I only run PC based systems personally, but prefer dedicated NVRs for the few commercial systems I support.

Dahua does something similar, and someone posted the algorithm a while back. I don't know if they still use the same setup or not, but it's a handy thing to know, since Dahua's support is not so good.

The Dahua 4300S I'm testing now has a live view jerk when it decodes the i-frame. It's very repeatable and changes frequency when you change the i-frame setting. I haven't checked to see if it does it in the recordings.

Sorry, brain fart. I said ISO 12233, and I meant Siemens star - http://en.wikipedia.org/wiki/Siemens_star. Print one out one from the bottom of this page - http://en.wikipedia.org/wiki/File:Siemens_star.svg - and use it to focus. The Siemens star is much easier to use than the ISO chart! There are lots of websites with lots of info, as you've found, but not many comprehensive ones, and much of the info you'll find is outdated and inaccurate. Most of us have learned through the School of Hard Knocks. That said, these guys have some free general training that's good to start with: http://ipvm.com/report/ip_video_training_course_101 Some of their training is only available to subscribers, but their basic course is free. As for threading and BI, nobody's really answered that with data. For instance, whether an i7 performs better than the same speed i5 is still unclear, though most people will recommend an i7. Whether the i3 will work for you will totally depend on frame rate, as 24MP is a pretty big load for an i3. I'd be surprised if you could run those cams at 10 fps on that CPU, but it's often hard to predict. You might read up, then ask this, over at the BI community forum, cam-it.org, where there's a lot of discussion about CPU performance.

I'll add that there's a relatively small group here interested in the low-level technical details like this, and the write-up definitely was not a waste for those who want to understand this better. Most people may not want or need it, but having it documented is a good thing.

Many cams have terrible information on field of view. It's either inaccurate, or it only shows one resolution, and not the others. FOV calculators are not useful for M12 lens systems, though it looks like this is what the manufacturers use. There's a fairly simple way to measure actual FOV with a ruler, a few parts, and a little math: http://www.cam-it.org/index.php?topic=5424.0 All you need is to have the ruler a known distance from the cam, and be able to read what the ruler says at the edges of the image.