MaxIcon

I've got a pair of Areconts - an AV5105DN and an AV5100. The software and user interface are primitive, the frame rate struggles to meet half the spec, and the AV5100 has died recently. I'm not an Arecont fan either.

What you'll find is that ACTi doesn't spec resolution (or much of anything beyond the basics) on their lenses (that I could find). Most vendors don't. The ACTi lenses aren't that expensive, which leads me to believe they aren't high MP rated, but I could be wrong. As an example, Axis publishes specs on their MP lenses, but each spec is from the OEM and has different parameters, and the ones that do spec resolution mostly say "> 500 TVL", which is not a number that gives me warm fuzzy feelings for an MP lens. This is another area the vendors need to get under control. With C/CS mount lenses, you can choose the tradeoffs between price and features, but M12 lenses are mostly a roll of the dice.

Looks confusing. D1 is standard analog resolution, and CIF is 1/4 resolution, which is very low res. This used to be popular for some reason. 4CIF is another term for D1, so it's not clear if they mean 4 x CIF inputs, or 4CIF = same as D1. It may mean it will record 4 each D1 and 4 each CIF in realtime, and would have to reduce the frame rate if you wanted higher resolution on the 2nd set of channels (or they may not be capable of full D1 at all).

Never heard of them, but here's what I'd do: - Figure out which of the pins are grounds. There should be one ground for each input. - Figure out which pins are video in. Typically, you'd have 8 pins going to 8 identical resistors on the card, which then go to the video encoder. You can test them by using clip leads to connect a camera's output to the connector side of each resistor and seeing if the video appears. If you can see the traces where they connect to the input connector, you should be able to spot which pins are connected directly together on the board, which would usually be the grounds. You'll want to be careful not to short anything, of course!

Buellwinkle has some experience with LPR systems. You might ask over in the IP cam forum.

This is a problem with these no-name capture systems. Blue Iris works with a variety of analog capture boards, but you won't know about this one until you try it on the demo version. If it doesn't the developer is very responsive to requests, and could tell you if there's any hope. I don't know what your budget is, but the software costs $50 if it works for you.

CS lenses can work with up to 1.3" sensors, but these are uncommon. It's rare to see sensors larger than 1/2", and most vendors are moving to 1/3" unless you go for the high-end gear. As the sensor element gets smaller, lens quality becomes much more critical. It doesn't help to have a 1/3" 10MP sensor if the lens resolution doesn't support that pixel density, and high-resolution MP lenses are expensive. Note that the actual sensor size is about 2/3 the spec size, due to the quirks of vidicon camera tubes. A 1/2" sensor is really only about 1/3" diagonally, depending. Also, going from 2MP to 5MP gives 2.5x the total resolution, but you don't get that much improvement in pixels per foot (or whatever unit you prefer), so it's important to check where you need the higher resolution. 1080p - 1920×1080 ACTi 5MP - 2592x1944 Total resolution improvement - 2.4x Horizontal resolution improvement - 1.35x Vertical resolution improvement - 1.8x Multiply those together and you get the total improvement: 1.35 x 1.8 = 2.4x ETA: Just realized you meant 3MP. Same thing, less improvement - just a 25% increase in linear resolution! ACTi 3MP - 2048x1536 ACTi 5MP - 2592x1944 Total resolution improvement - 1.6x Horizontal resolution improvement - 1.26x Vertical resolution improvement - 1.26x Multiply those together and you get the total improvement: 1.26 x 1.26 = 1.6x

Your exposure time is too long. You'll want to limit it to 1/30 second, which will still give motion blur on cars going by, but not as much with people. 1/60 is better, but will reduce your low light quality even more. Anything longer than 1/30 second will give more and more motion blur.

If your NVR software doesn't display network usage, you can measure it at the PC with process explorer, which will show a 5 minute running plot of this, CPU, HD and I/O usage. http://technet.microsoft.com/en-us/sysinternals/bb896653.aspx

I also ran an Altronix supply for 10 years with no problems. Finally retired it when I went all POE, but it was a great piece of gear, with multiple fused outputs.

Most decent cameras let you set the bit rate, which lets you calculate how many hours of recording you can do on a given card. Resolution and frame rate don't matter if you're using fixed bit rate. If you're using variable bit rate, you'd have to test how that works on the camera you chose, because different programmers do it differently. Motion detect is where it gets tricky. You would know the pre-trigger recording time or frames and post-trigger time, and would need to estimate how many motion triggers and how long the triggers will last. So, at 4096 kb/s: (4096 kb/sec - that's bits) / 8 = 512 KB/sec (that's bytes) 512k * 60 sec = 31 MB/minute 31MB * 60 min = 1.9GB/hour For full time recording, you'd get about 17 hours on a 32GB card. For motion detect, if you had 5 seconds pre-trigger recording and 30 seconds post-trigger recording, each event would take 35 seconds minimum. If an event lasted 2 minutes, that would be 2:35 recording time for that event. Best bet, if you can, is to install it and start gathering data. Otherwise, you'd have to guess at the number of events and how long they last.

I've got a couple of fixed 1080p 4mm cams that give me a 140 degree view, and they're an OK substitute for a PTZ at a low price, but they definitely don't give the resolution a PTZ would in an area of interest. Once you get past 30' or so out, they just don't have the pixels per foot to give "id-the-perp" quality images. This is the standard trade-off of coverage vs resolution, of course. I also don't like how the images split my front coverage right in the middle, as that's where things are often going on, so I not only get lower ppf at the edges, but overall lower image quality compared to the center. I'm thinking about adding a 3rd cam and switching them to 2.8 mm lenses (or whatever the numbers work out to), which would be approaching the cost of an inexpensive PTZ, though it would give me that full-time panoramic coverage at better resolution, and would give me a single view of my center coverage.

What about an independent recording system in a small box? Depending on how many cams and what resolution/fps you're looking at, you could set up a small Blue Iris or similar system in a tiny mini-ITX box that would record independently all the time. It would run a bit more than $250, but not much, if you weren't recording too many cams. Getting it to upload the content after an interruption would take a little script coding, but it's a PC, so the sky's the limit there.

This is a common problem for people starting out with IP cams. The specs look good and the prices are low, but you don't realize how important software and support are until you get the cams and spend weeks struggling with them. Definitely worth sticking with the common name brands with lots of community support!

It's hard to go wrong with the Hik bullets or domes, if you don't need audio or a 3 axis dome. They're the sweet spot in the under $250 cams right now. The 3MP setting actually covers less horizontal FOV than the 1080p setting, though, so I keep mine on 1080p for the wider coverage. PTZs are only worthwhile if you set them to patrol or are ready to man them when required, and good ones are pricey ($600 and up). Many people are better off with 2 dedicated cams instead of 1 PTZ, but it depends on your situation. Most of the cams with audio are a good bit bigger (or lower res) than the mini-bullets and domes. My Dahua HFW3300C has audio in/out (no built-in mic) and is huge compared to the 3200S or Hik bullet, as well as being a fair bit more expensive.

I bought a cheap ebay mic the other week to play around with, but haven't gotten around to hooking it up yet. Maybe next weekend...

Looks like the tech support people need to help out. You're almost there. For Android, try ip cam viewer free, which supports a lot of cameras and dvrs. The free version only supports 6 channels, but if it works, the pro version is only $4.

Here's a link to 0200d for the IP8332: https://docs.google.com/file/d/0B5RXJz1sMJ9VWk85WlBVU0N2S00/edit?usp=sharing Filename: ip8332_0200d_firmware.zip

4096 is for the bit rate. That's usually a good setting for 1080p h.264, as BW says. 15000 could be contributing to your problems, as that's quite high. GOP is more commonly called i-frame or keyframe, and the setting depends. This sets how often the full frame image is sent out, and everything between that and the next i-frame is an update of the changed pixels. Lower i-frame gives better quality, but uses more of your bit rate, or bandwidth, on these frames. Higher i-frame leaves more bandwidth for the in-between frames (called p-frames). A good starting place is to have the i-frame or GOP at 1x or 2x the frame rate. If your frame rate is 10 fps, try 10 or 20 for the GOP setting and see how they do.

Those are the actual filters. Better quality filters are pieces of glass, though some are thinner than that, like the one I removed from a 5MP Arecont. If you go to ebay and search on the following, you'll find more choices. I sort by lowest price first, to get rid of the full lens filters and such. ir blocking filter ir cut filter ir filter If you can't find a filter that fits, you'll need to look for a lens with a built-in filter, like the GoPro and similar.

Many newer DVRs have image streams for each channel that can be fed to an NVR. Several people at the BI forum use BI to record the channels from their analog DVRs like Zmodo. If you go to cam-it.org and search on zmodo or dvr, you'll find some threads that discuss this.

They used to have several older versions of firmware on the support site, but seem to have gone away from that. I have a copy of 0200d and can post a link to it later. I haven't tried version 3, but when I went from 1.xx to 2.00d, it wouldn't let me go back to the older version again. Support said they could do it for me, but I never got around to it, as it was only one camera. I left all the others on 1.xx, as 0200d caused worse night images and more IR halo for me.

There are a couple of common ways to do this: - Connect a mini-PC to your TV and display the signal from that, like discussed here: http://www.networkcameracritic.com/?p=1049 - Connect a dedicated laptop to your PC and use it to view the NVR. I like using a 10" tablet as a dedicated viewer, but it's quite a small screen compared to a TV. I also have an old dedicated laptop with a monitor attached in the area where I need a bigger display.

Are you viewing live locally, right at the NVR, or over the internet? You could try enabling one camera at a time to see where the problem picks up. The NVR specs can be misleading sometimes, and based on their recommendation that you reduce the frame rate and/or resolution, you may have one that can't handle the full load. More info on the model and specs would be useful.

It's probably an M12 lens. Changing these can be tricky; best bet is if you can get one from the same vendor that they know will work with this cam. Here's a post about the issues you have to consider for your cam: viewtopic.php?p=233184#p233184