atcsam

Not sure which cameras you have, but you need to access and set the camera parameters separately. WDR and BLC generally cut the fps in half. Noise reduction also slows down the auto shutter speed/ fps functions. Things to try. Manually set the fps, 15 if possible, 7fps at the low end. Manual exposure, generally related to shutter speed. GOV/ I frame 10 or less (increases bandwidth) H264 settings Once you get the perfect night time settings, your daytime pics will most likely suffer unless you can have separate day/night settings. Lowering the resolution may help, less pixels to process. Best of luck

My 2 cents.. The idea of separate NVRs/ servers is a good one, the draw back/ tricky point I can see is if you need to work with all cameras/ departments simultaneously. Single system is easier to control data, physical security, etc. Could get pricey depending on your surveillance requirements. Things to consider: Cheap usually gets you cheap cameras, OK for observation and control under ideal conditions. If your looking at forensic grade footage, difficult lighting situations, then some higher end cameras should be considered. Recording 24/7, motion, scheduled, how long do you need data for.. Terabytes can be used up quickly. 30 cams at 8mb/s needs 240mb/s recording, need another 240mbs going out for network viewing, enterprise level hard drives, gigabit network, recommend dual gigabit ports on servers/ nvr. If your looking at loss prevention, security guard needs, then I'd invest in a higher end system.

Let me throw in Axxon next for a try, 16 cam free, 1TB storage limit. If your using motion detection you can generally get a couple of weeks worth of video.

Primarily advantage of frame by frame requires less cpu resources, especially if using keyframes only for analysis. Using a frame by frame comparison with high sensitivity can detect smaller objects in general, including shadows, cloud movement reflecting off glass, frog hopping across the driveway etc. You might want to contact support to get more information Most mfgs keep the details of their analytic algorithms close at hand and just refer to general terms of sensitivity, percentage of area, zones, and the like. Look up video analytics, or video content analysis (VCA), plenty of google hits. Personally I prefer edge detection and find Bosch's IVA to be one of the better implementations compared to the others.

Accumulated background helps reduce false motion events, light rain etc..

Shutter speed is usually the killer for motion blur in low light setting. As others mentioned, turn off the WDR, noise reduction etc. You could try to set a manual shutter speed that works for both day and night (unless you can set separate settings). Another setting is the GOV length for the h264 stream. High GOV and low fps can get messy. I use IPCENTCOM on my lumina to peek at my cameras. I'm not sure if you can get direct access to the cameras through the NVR

I like to use axxons calculator for looking at system requirements, along with my 2cents as a program to look at. They offer a free 16 cam version, 1TB limit, plenty for motion or analytic detection. Here's a link https://sale.axxonsoft.com/calc/calculator.jsf Once you enter your cams and resolution, you can turn on/off the record, client, view streams and look at the cpu loads. Nicely estimates storage and throughput. They support on camera analytics in addition to they'er software analytics. Handles Bosch's IVA well and most of the Samsungs. have an issue with my7082's though. Work in progress. If you add an Nvida CUDA card, it supports hardware decoding which should help cpu usage on the client/ viewing side of the house

Get the specs for your devices first. A relay is not a power supply, two different animals.. You will need a power supply for the siren and/or relay. IF your cameras contacts can handle the voltage and current rating of the siren then I would hook them up as follows.. Power supply + to Siren + Siren - to Alarm Out + Alarm Out - to Power Supply - To add a relay, the contact side has to meet voltage and current of your siren, coil/control side needs to operate at sirens voltage and not draw or exceed your cameras current limits. Power Supply + to Siren and + coil of the relay Siren - to relay + NO (normally open) contact Relay contact - to Power supply - Relay coil - to Alarm out + Alarm Out - to PS - The power supply + will have two wires, one going to the siren and the relay NO contact. The power supply - will have two wires one going to the common contact, another to Alarm out - You really need to get the specifications for both the siren and the camera before burning things up; and perhaps a local expert to help set things up.

You may have to contact support for the alarm out specifications (voltage/ current) I'm fairly certain the camera will not power the siren but the contacts could act as a switch and connect a ground/return. You would need a relay if the sirens current exceeds the alarm out contacts specification for the camera. (some are limited to 5ma sink current) Most likely you will need a separate power supply for your siren, the address below shows a simple connection direct to the camera IF it can handle the sirens current. Otherwise, as mentioned previously, you will need to add a relay between the camera and the siren. A relay set up would also provide some protection to camera's alarm out contacts http://www.hkvstar.com/technology-news/how-to-connect-sensor-to-ip-camera-s-alarm-i-o.html

Thanks, I'll revisit the PC/ software route again. Budget cuts have me back down to $1K or less for the NVR route. I have about $700 tied up in the current collection, mostly NOS. Going for a upper end, low budget system. I was hoping to exploit the ONVIF capabilities of the NVRs, I'm fairly certain I can at least record the rtsp streams directly, just not sure of the motion/alarm. Cameras are 2.0 or above and I can record the analytics on the edge as required. Sammys latest firmware supports ONVIF M.D/V.A events, I leaning towards it for a standalone solution... I'd prefer to get the NVR5216 to play for future growth. Anyone with experience with using ONVIF with the Dahua NVR5216?

Newbe myself but, for lens bandwidth and camera placement take a look at IP Video System design Tools jvsg dot com. They have a 30 day trial and it really opened my eyes regarding Camera placement, resolution, lens, and bandwidth. Its a bit pricey for the home guy but you get 30 days to setup your plan. I captured an overhead pic from Google of the house, inserted as a background and could view the effectiveness of various cameras and setups. Really nice tool. I cannot comment on the cameras though I've heard good things on Hikvision. Stay with the 3mp if you want a chance at forensic identification, nighttime is the rough part unless you use external lighting. My 2cents

1st time post, long time viewer, some great info and experts on this forum. Debating on Samsung SRN-470D or Dahua NVR 5216. System: Samsung 7082 & 1080R Bosch NDN 832 IVA enabled (I like this one) Grandstream 3672_FHD Cisco CE 500 24PC Budget cuts have reduced my 8-10 camera system to 4 for the next 6m-1yr.. 2 3MP at 20fps, 15fps respectively 1 2M 1080P FHD 30 FPS 1 VGA I figure 3TB for 7 days 24/7 more if motion/ alarm recording. I'm still awaiting a response from Dahua for the ONVIF capabilities for the specific models. I can set up a refurbed 470D for the price of the 5216, just limited to 4 cameras. The 5216 would allow expansion whereas the 470D is a fixed solution. Any thoughts or alternatives appreciated