Thomas

It would destroy the remote viewing unless you know what you're doing with QoS set up and want to drop the money on a real router. The real world answer is no. On a cheap DSL connection you will not have enough upload bandwith or latentcy to make this feasible. Most DSL connections are ADSL (Asyncronis Digital Suscriber Line). Which means your upload and download bandwith will be differant. Most lower tier DSL packages are 256k to 756k upload speeds. If motion is occuring then you will generate video faster then you can move it. So if you get robs when you're a half hour or so behind on your video...well that system is useless. This isn't like a P2P transfer of video, there is a live requirement to make it useful.

Short of looking at a T-1 or better connection you will not have enough upload bandwith for off-site storage to be feasable.

You actully have the two things backwards. It has almost no effect on video processing (Almost all of that is CPU time with extremely little ram use and almost never touches a video card) but it will affect how much CPU usage you see in play back.

Generally a Resolution testing chart is used.

Except that your concrete room has valves in it. To get into that room I have to have kicked at least one camera off the network. Are you telling me that the ability to trvially remove cameras from the network is a secure design? Or the ability to remove all of the cameras from the network? Perhaps it won't allow for data loss, but will it continue to protect physical security?

Except that the encryption isn't anywhere near the protection of any of the things you mention. It's more akin to a gate with a broken lock. It keeps the really honest to and doesn't act like much of a speed bump to those who aren't. And even if I put all of my cameras on the differant network, I still run the risk of some wanna be script kiddie taking the cameras offline at the same time something happens that needs to be recorded. Telling people that "wireless can be secured if you know what you're doing" is a really false statement because the wi-fi standards were rushed. Too many people forgot that processing power increases at rapid rate. What is "unbreakable" today is trival to break tomarrow. Wireless systems have too many issues for current standards to ever be "secure". All you can do is limit the damage to a single camera going off the system. And there may be times when that is an acceptable risk. But acceptable risk is not the same thing as secure.

Have you tried contacting the person you got it from?

It helps a little bit. With live monitoring are we talking about 24/7 Security guards or periodic checks? Are you aiming more for physical security or anti-thieft monitoring? Ideally if you could write up six or seven paragraphs with your goals, needs and wants it really would help alot.

Except that both encryption standards for WiFi are a joke. WEP might as well not exist and WAP can be broken in a fairly short period. If you think wireless can be properly secured then you aren't paying attention. The flaws in WEP aren't new and have been well documented for the last five years. The flaws in WPA are three years old. We're not talking about zero-day exploits here. We aren't talking about bad firmware. We are talking about fundemental flaws in how the standard is writen. WEP: Easily Broken WPA: " " SSID Broadcasting disabled: Easily bypassed via tools like Kismet Mac Address Filtering: Yeah, here is where the standard is great. Yep, lets broadcast MAC Addresses in PLAINTEXT. Yep. God knows the MAC Address spoofing hasn't been known for over ten years. Yep. Great thinking. Subnetting: Okay, so I can't get at the rest of the network but I can turn your cameras into expensive paper weights by taking them off the network. VPN Tunnel: So I can't see the camera but hey it's still off the network. Point to point antenna: Better then 90% of the other suggestions as long as you can somehow keep the reciving end from talking to an onmi-directional source. Did I miss any tricks? Short of looking at microwave based gear then it should be assumed that any wireless install is insecure. Simply turning on encryption isn't going to get you anywhere.

You didn't really answer my question. What is the focus of the system going to be? Will there be live monitoring? Is it retail? Is it a wharehouse? External or internal thief? Large objects or small objects?

Compared to the CPU power to do the encoding the hashing functions are trival in CPU cost. Kensplace, that's why there are chain of evidence issues. And it's not trivial to just "remove" someone or "add" someone to video. You have to go in and look at thier shadows, make sure they are angled right, see how they interact with other objects...it's not trival for a pro.

Oh you can do some nasty things to prevent tampering of the watermark. Hash each frame, hash each file and then I can run it through some one way transformations...yeah.

What kind of application are you doing? Retail, home or small bussiness? 4,8, or 16 cameras?

Cons: Extremely expensive to do correctly. Analog wireless is unencrypted. IP Wireless might as well be unencrypted.

And residental end users should not be generally looking for 30 fps/25 fps recordings. The best benfit of realtime recording is when looking for slight of hand or with a human monitoring the cameras. Any recording over 15/13 fps and you are burning through drive space.

Why the real time requirement?

Odd. They should see a differance unless the application is single threaded...which would be odd.

For some crimes you throw away the book. For some crimes you offer treatment.

Dual cores will always run circles around chips with just HT.

Ah cable techs. Should we ever run out of manure they still able to provide us with ample amounts of fecal matter from their mouths. The camera signal sent out would be a stream of packets. They would be like any other stream of packets. Now it may be possible that the router, or modems are having an issue, but it wouldn't be the camera. The joy of TCP/IP is that it exists above the physical layer. A packet is a packet, is a packet. What generates the packet has no effect on the physical movement of said packet.

Not much in the low profile area. The cards are pretty well cramed with chips. There are some external USB devices but they can be very tempermental about enviromental condisions. The chipsets used for the A to D conversion are hot little chips.

Passing the video through the hinge would be putting it in contact with a large metal object that makes a great anntena. Your interfearance problem would be huge.

The best way would be to come into via the side of the door (Less pulling on the cable) but that would be a *** to do in an apartment.

I have a very small amount of experance with Imron but it has been positive for some of our dealers.

You could try a contact but given it's location you're not talking about a long life span.