MaxIcon

The problem here is you may be comparing apples and oranges, as you suggested. Very few vendors spec the actual IR output of their LEDs for either wavelength, so comparing one to the other is nearly impossible, though this is starting to change. Also, measuring actual IR output requires expensive gear, typically thousands of dollars. IR LEDs output specs are for total Watts, and/or Watts per steradian (basically the area of a portion of a sphere). An LED with 1/2W output could have a 10 degree angle or a 60 degree angle, each of which would have different W/sr. Here's an example of the rare illuminator giving actual specs: http://www.ebay.com/itm/Professional-IR-Illuminator-IR-25m-Array-940nm-Invisible-Light-Waterproof-2500mW-/200907291308#ht_3155wt_1226 They tell us the IR output power is 2500 mW (2.5W), with a 30 degree output angle (but they specify half/full angle, so it's not clear if it's 30 degrees total or 60 degrees total), and the input power is 6W. Whether we can trust their numbers is another matter... Here's another with actual specs: http://www.ebay.com/itm/940nm-IR-Invisible-illuminator-Indoor-MIX-LED-Array-IR-angle-120-degree-/200816839509?pt=US_Surveillance_Accessories&hash=item2ec19dcb55 This is what you need to know, and what both camera and illuminator vendors should spec - total IR output power and output angle. This would allow us to make logical choices and compare apples to apples. When I replaced some IR LEDs on my Dahua 2100 recently, I compared the original and the new LEDs by powering them up at the same current the camera uses, pointing them at grid paper, and taking a snapshot from the 2100 with the camera's IR unplugged. Since I knew the power rating of the LEDs I bought to repair it, I could make some guesses about relative output power on the originals. Pics of the comparison are here: http://www.cam-it.org/index.php?topic=4946.0 However, even if you buy an illuminator with known output power, you'd have to guess about the power of the original LEDs, but if you use the originals as a reference, you could point both the cam and the illuminator at something a few feet away, and get a pretty good comparison. So, best starting place is with an illuminator that specifies actual output power. Best bet with what you have now is to set up a sheet or something in a dark place, point the Hik at it, and point the illuminator at it to see how they compare.

So the one I bought from . isn't the official US version? It's an official US version, but not Hikvision branded. They have their own model number, both on the box and in the firmware, but it's identical to the Hik version in every other way I can check.

BI has a phone app that users seem pretty happy with. I haven't used it, but you could ask over at the BI community forum, cam-it.org. BI has a demo version, but I don't think the app does.

In general, recorded video on IP cams should be very similar, if not identical, to the live view. This is different from analog cams, where the live view is uncompressed and the recordings are compressed. On IP cams, the compression is done in the chipset before it's sent out over the network, and the received live view will be the same as the recorded view. Some NVR software re-encodes the video, which can cause a change in quality as well as use a lot of CPU, but most software that comes with the camera just records the incoming stream directly. Blue Iris, for instance, lets you choose between direct to disk recording or re-encoding, while the Aver software records directly and doesn't re-encode. I'm not familiar with other software. Exported video is another matter - some exporters re-encode the original recordings, some don't. Re-encoding is generally much slower and results in really different file sizes, either larger or smaller, depending on your settings. Exporting the same video without re-encoding into a different wrapper is faster and results in closer file sizes. You have to test your system to find out for sure.

They've been popping up on ebay lately, including ones with the standard E27 plug like a normal lightbulb uses. Search on 940 illuminator. Many of them don't look weather proof, and being ebay, there's no telling about specs and quality, but they're showing up at least.

There was a post a while back where someone had this problem with their Dahua bullet, and since then, I've seen it a few times on my Dahua HFW3300C full-size bullet. It definitely seems to be heat related, and if I refocus it while it's hot, it gets blurry again when it cools down. No idea why yet (no time spent chasing it), but you might be right - these are so sensitive to the lens focus that a pretty small mechanical shift could cause it.

Cameras that can see 850 nm IR can generally see 940 nm as well, but with lower sensitivity. How much lower depends on the sensor specs, and those can be hard to get. You can assume half the sensitivity as a rule of thumb, but you won't know for sure until you try it. Edit: Here's a typical response chart; each sensor will be a bit different.

An inexpensive and flexible option is Blue Iris, which supports the Hiks nicely and is a good value. It needs a fairly powerful PC if you want to run a lot of MPs. You can download a demo version to see how you like it.

Someone reported loading their Dahua with QSee firmware to check it out, then trying to go back and it wouldn't let them, so it sounds like the QSee firmware prevents you from doing this.

Yeah, I'd agree with this. One of the main complaints I hear about dash cams is failure due to heat. The recordings get corrupted and unreliable, at which point you've wasted your money. You want them to work without having to think about them, and when you do need them, they need to be good to go. I bought one recently on a whim, and will mount it up as soon as I get time. For the price, it's hard to go wrong if you stick to cams with a good history of reviews. I see a future camera option for cars that'll start as front, maybe front/back (since backup cams are already popular, if a bit low res), and eventually all 4 sides, for 360 degree coverage. A module sitting on the roof with a single network cable would be awesome; you'd just have to remember to keep the lens covers clean (though the luxe versions would have wipers and washers!). This would go to the car's internal network, complete with POE to run the cams, that you could connect to with your wifi from your house or laptop. I keep waiting for the future to show up. No jetpacks, no food pill, no teleporters - somebody's slacking off!

Well, here's the test. Set one of these up to record a week's worth of video of your front yard, and see how much it costs. IP cams have POE power adapters and network boards, more complicated firmware, IR LEDs, moveable IR filters, and weatherproof enclosures. You can add all that to your car cam (might have to remove a fixed IR filter if it's got one), and by the time you're done, see what it all costs. Then, you have to come up with a way to get the video off of it automatically, or stream it via USB and record to an external device. If you need it very far away from the recording device, you'll need USB extenders, or to locate the device close to the camera. It doesn't say what the f-stop is for the lens, or what the sensor size is, but this will affect the night images and sensitivity. Likewise, if you need to replace the lens because you need a different field of view, it doesn't specify anything about it. It appears to be a nice piece of gear for the price, but by the time you turned it into a surveillance camera, you'd have spent many hours and dollars on it, and you'd still have a cobbled together solution. Another question to ask - if it were straightforward, wouldn't someone be tapping into that market already? Maybe they are and we'll see low-cost surveillance cams based on these units any day now.

Another decent 1/2" zoom lens for a reasonable price is the Arecont MPL4-10, though it's a manual iris lens. They typically run just over $100. Your current lens shows a fair bit of chromatic aberration (all those purple fringes, pretty common on inexpensive lenses), but looks fairly sharp, for looking through a window. Getting sharp focus on a 5MP cam with a CS mount lens can be tricky, as very small adjustments can change focus a good bit. You should be able to set your camera for a maximum exposure time with variable exposure below that. That is, it would auto-adjust exposure, but never go longer than your maximum exposure time of 1/125 or whatever. That would give good daytime control but avoid the motion blur at night. I'm not sure how the Grandstream supports that type of setting, though. The license on that image is only about 25 pixels wide, which isn't enough for ID, in general, and wouldn't be helped by a sharper lens. You'd need to zoom in more, or hope for a capture with the car closer to the camera. For a straight on view, you'd need 40 pixels per foot minimum (a license plate is about a foot wide), and more when it's at an angle like this or if the lighting isn't great.

Actually, it makes sense from a pixel perspective. Most of the pixels change all at once when it changes mode, and generally the exposure has to change for the new setting as well, making it look like the mother of all motion events. Blue Iris has a sophisticated motion detect algorithm, allowing you to detect based on things like color changes or brightness changes only, groups of pixels where the motion is detected as an object, and a variety of other settings, but I haven't seen an NVR with that level of adjustment.

Technically, both 720p (1MP) and 1080p (2MP) are megapixel.

Hmm... Good question. I'm not sure what the plastic is over the standard illuminators, but anything that passes most of the IR would work. If you google plastic ir transmission, there are a lot of discussions. It looks like both acrylic and polycarbonate transmit IR, and these guys make black acrylic that passes IR if you want to hide it: http://www.eplastics.com/Plexiglass_Acrylic_Sheet_Infrared_Transmitting This site has data and suggestions, but their charts aren't showing up for me: http://www.plasticgenius.com/2011/05/infrared-and-ultraviolet-transmission.html One test would be to get scraps of materials to try and see if a remote control works behind them. Note that you'll probably still get reflections off of any of these, so the illuminator would work behind them, but the camera might not.

Yeah, it might be worth asking them for this feature. It's not there in the current firmware. You might be able to unplug the IR board, which works on some cameras like the Dahua. I haven't tested it on the Hik. Some cameras use the photosensor for day/night switching, like my Vivotek IP8332. This gives fast, responsive switching, but you can't disconnect the IR board. Vivotek lets you control all 3 aspects independently - IR filter state, IR LED state, day/night mode - so it doesn't matter.

My Hiks have the build date in the serial number. If you go to the SN page, you'll get some variation of this - mine is a vendor SN, not OEM Hikvision, but the Hiks are the same, I believe: KBU320120130607CCRRxxxxxxxxx The first part is the model number, and the bold part is the manufacture date. Look for the 2013!

I always prefer image quality over frame rate. Having smooth, high FPS video of pixellated, blurry images doesn't do you any good, while 10 fps of high quality images is useful. When you do tests, be sure to include fast motion where a lot of the image changes, as that's where the problems show up. You can have a beautiful static image, but things get all pixellated when lots of motion starts if the compression's too high or the bandwidth's too low.

I set up a pair of Hik 3MP bullets in 1080p mode (3MP is a narrower FOV) to get as close as I could to 180 degree, and they're a good bit short. http://www.cam-it.org/index.php?topic=4828.msg25682#msg25682 Their specs claim 76 degrees with the 4mm lens (they don't say if that's 1080p or 3MP mode), but mine measure 68 degrees at 1080p, and I'd definitely like closer to 180 degrees total. I may swap the lenses, but that's a pain in the butt with these cams and M12 lenses in general.

I can't watch the video at work, but a common cause of image problems at night is the IR LEDs coming on and loading down the power supply, especially if it's a long or light gauge power cable. You can test this by hooking up a power supply much closer to the camera with a short run. IR reflections are another big problem, but those usually cause washout, not wavy images.

There are few relevant specs given for most IR illuminators. I'd like to see them spec the total IR output power and the beam angle, since that would let you compare illuminators easily, but nobody does this. There's little difference between 850nm and 940nm except for the sensor sensitivity and visibility to the human eye. Beam angle is determined by the LED's spec - you can usually buy the same LED in multiple beam angles. Likewise, IR output power is not very related to the input power; a 1W LED can have a huge range of actual IR output. Actually testing IR output is tricky. Decent IR power meters can cost thousands of dollars. It's a jungle out there, and you pretty much have to buy something and test it to see how it works for you.

Some lenses have a lock ring instead of the allen screw, and the Hik mini-bullets have a spot of glue on the threads holding them in place that needs to be scraped off first. Most M12 lenses have an image spot large enough to cover slightly larger sensors, but M12 lenses don't have strong specs for this kind of thing, unlike older CS mount lenses. Likewise, if the new lens needs to be screwed in too far to focus, it'll hit the IR filter assembly before it's in focus. This is a common problem on replacement M12 lenses if the back focus length is too short, and few vendors specify either the BFL or the image spot size. You pretty much have to order one and see if it works. If not, well, try, try again!

i found setting my contrast to 0 pick up the coons very well (only need this for night use profile for day/night) Thanks, I may try this. I only have Hiks on the front of the house, and the raccoons mostly play in the back, but I have better lighting in the back. Last time they were up front at night, I found them because one of the Hiks started recording due to a bunch of dust in the air being seen by the IR - the raccoons had passed under, messing with stuff, and it didn't catch them, but the dust they raised gave me a time stamp. I went to my backup 24x7 system and saw them messing around.

They need a field on the web page for a discount code: iheartcapitalism, for instance!

Which lens did they come with originally, and which ones did you replace them with? I have a few of their MPL4-10 lenses, and they're pretty good for the money, but not great compared to the pricey 1/2" lenses.