Need some help

[Newtothis 0]

Advance notice: Please don't flame me. I'm really new to this industry

 

I've been tasked by my boss to look at building security and I hate to say it but I really don't know much about it. I came across this forum and most of the people seem to be really helpful and am wondering if any of you can answer some of my questions.

 

Basically I've been tasked to see what options there are to add to this new building that we are going to build. I think (at this time) the idea is to have about 100+ cameras. We will probably be asking an integrator to do the job but we want to have some knowledge about it ourselves so he doesn't try to pull a fast one on us.

 

There are so many brands out there and so many options (ie standalone/ PC etc etc). It seems that with the standalones you get a pretty good quality and it can be scaled up for 100+ cameras if you go with a high end company. However, one company that seems to have a pretty good product for PC Based systems (at least from my limited scan) is GeoVision. But according to the distributor sites it is not stackable. Does this mean that the maximum number of channels it can have is 16? Is it possible to go to 100+? And if so how would one do so?

 

Question 2.) I listed GeoVision as an example but who are the players one would reccomend for a 100+ camera solution?

 

Thanks

[kensplace 0]

Even if you could have 100 cams connected to a single pc, which at present, due to the design of the pc hardware I dont think is possible, due to pci bus limitations, I would not recommend it.

 

Dont think I have ever seen anything that will record 100 cams on one device (certainly not at full frame rate), it may exist but never come across it yet.

 

Whatever you choose, either pc based, or standalone hardware solutions, you will be better of with splitting the project up into groups of cameras, with 100 cameras one person will not be able to monitor them.

 

Having the recording split over several machines will also avoid having all your eggs in one basket, if you have a hardware fault, you dont want to lose the recording capabilities of ALL cameras in one go.

 

If its split over several devices (be it pc, or otherwise) then if one fails, you still have the others, and if any important cameras were on the failed device you can swap them to a working one until the repair is completed.

 

 

With a large system like this, you will need something that is capable of splitting the cameras into groups or zones, so you have say a zone for each floor, or area you want to protect.

 

Standalone hardware solution would probably be a large matrix system, which lets you select which zone you want, then which camera in that zone, then when selected you can also control the camera (if fitted with telemetry and zoom/pan tilt etc). Similar with a pc based system, a pc for each group of cameras, but as you mention it depends on how scaleable geo is for multiple pc systems if you want central control from one station, no idea on that score. Hardware based will allow selection of any cam from one point with a decent matrix system.

 

If the system is to be monitored (rather than just recorded and reviewed after any event) then the system must provide a method of allowing several operators to control different cameras, as if you stuck a 100 camera display in front of one person chances are someone could be murdered in one of the screens and they would never see it, as there is just to much to look at.

 

Four cams is pretty easy to watch, eight still possible, but not as easy, it gets trickier, and much easier to miss things with 16 cams, and anything above that would just be asking for trouble. So several displays would be needed for realtime monitoring, unless you only view screens that have motion detected on them or alarms triggered, which could reduce the amount you need to look at (unless all hell breaks loose and motion occurs on most cams!)

 

As far as recording goes, either multiple pc or dvr solution, with frame rates depending on your needs, its not going to be a cheap system, as thats a lot of cameras to record, and unless its monitored realtime by more than one person, reveiwing the footage would probably be a full time job for more than one person.

[eyespye.security 0]

There are a lot of complexities that go into not only implementing a large installation such as this, but also monitoring it.

 

Security cameras are great (obviously), but looking at them with some meaningful intentions is completely a different beast.

 

One thing your company/boss should keep in mind is who is going to monitor this? Will it be some poor guy stuck in front of a TV, or will you be hiring security professionals to do the monitoring? Or will it be outsourced to some company that does surveillance monitoring?

 

I'd concur with the previous post that it is best to break them up in 4 or 8s. If you go to large buildings, you'll see that each security has typically 4 channels and certainly not more than 8.

 

We have done some large installations, so I'd be glad to talk with you about it - just send me a PM.

 

Also, the other things you need to think about are:

• 
  Where will all of this equipment be sitting?
  Who would have access to the room(s)?
  Do you have the necessary cooling/power for the hardware?
  Is all of this in the budget?
  

The way our company approaches any installation, let alone something of this magnitude, is to to follow this process:

 

-------------------------------------------------

Assess - evaluate the location, identify control points, categorize personnel/guest usage, and note main areas of concern.

 

Interview - engage key personnel to collect various views on issues and needs.

 

Audit - identify current policies and security procedures while highlighting known issues.

 

Objectives - examine the issues that have prompted investing in a security solution.

 

Design - identify the appropriate technologies and services and taylor them to the specific business requirements. Identify risks and limitations of the technologies.

 

Plan - start planning and coordinating the project with all of the necessary parties. Produce a complete specification document along with all of the costs.

 

Implement - test all equipment prior to deployment and install during a phased time line depending on the job complexity.

 

Train - appropriate stakeholders are trained in the use of software systems and the key workings of hardware.

 

Support - a trained professional team handles all activities from product/software use to repair of equipment.

 

-------------------------------------------------

 

Bottom line: Your budget is a HUGE determining factor, because you can bascially accomplish the same thing from most of the brand-name technologies. We can get different manufacturers to compete against each other for better pricing. That is if you can find a manufacturer-agnosting integrator (for example, US!!! )

 

You can go off-brand if you have to keep the money on the lower end. Some of the off-brand products come with more warranty than the branded items.

 

It all comes down to money and what is REALLY required.

 

Again, feel free to PM me, or give me a call and I'll explain it more.

[Panarama Man 0]

The guys before me have already posted some good information.

 

You are certainly not looking at a small scale project. For 100+ cameras, make sure the dealer/integrator you will be dealing with comes with some references, because a project this scope is not for your everyday alarm installation company - things could go south awfully fast.

 

Depending upon how you are going to view the cameras, how you intend to use them overall, will dictate a long as to how you want to set up your system.

 

Good luck in your dealings.

[Newtothis 0]

Wow Thanks folks I really appreciate the advice that you all gave. Over the past day or so, I have done some more research too but would any of you recommend an NVR solution for something so big?

 

You can see how new I am to this! I totally didn't consider the fact that one guy can only see so many camera shots at a time. Do you think software analytics will be helpful in this case? Do most brands come with some kind of analysis tools like indexing or motion detection/ alarm?

[cctv_down_under 0]

Advance notice: Please don't flame me. I'm really new to this industry

 

Do not worry, that is what we are here for

 

We will probably be asking an integrator to do the job but we want to have some knowledge about it ourselves so he doesn't try to pull a fast one on us.

 

Not a doubt... not even a chance that you would do this on your own, besides that you will need not your average installer for this kind of work, so you will need to choose wisely

 

Even if you could have 100 cams connected to a single pc, which at present, due to the design of the pc hardware I dont think is possible, due to pci bus limitations, I would not recommend it.

 

This is possible on IP solutions only, and even then it would still be advised to have several NVR stations becuase of bandwidth and directional flow.

 

I would suggest you break your system up into three segments and stick to IP cameras (remembering though that unless you pay big bucks that ... if your network goes down you will lose recording). Imagine making a big circle around the whole floof plan and grouping the cameras together in sections, IE one wharehouse may have 8 cameras so that would all go to an 8 input switch in that section etc etc.

 

You need to realise that with a PC or a standalone you need to either operate it from the BOX's location which would be scattered around the network or you will need to wire everything back to one location and then use an anologue matrix to control it from remote locations (a very expensive excercise in labour)

 

The best option is to use IP cameras and then you may only need to run a few rings of Fibre, this should be on a seperate network to anything else and should be designed by a true data proffesional.

 

Keep in mind if you do not set up this fibre ring you will be adding massive labour amounts to the job by having to run 100 long cables etc (there are ways around this but none as easy as the networked option).

 

I realise you still need to wire to each switch however 8 or 10 or 20 cameras coming back to one point is a lot easier than trying to run 100 to a matrix.

 

The other option is to use DVR's in each location rather than a switch (considering that you have to cable to the switch poitn anyhow) but 99% of most DVR's either PC or Standalone are not going to have sufficient enough a virtua matrix to support your needs (now before anyone flames me, I am talking about the bandwith issues and abilities to use managed switches that IP represents).

 

You really need to work out if you will be monitoring the data or just using it for retrieval becasue this makes a huge difference to your infrastructure that you require.

 

NVR or Hybrid machines are going to be your best bet and setting them up in groups on a token ring network is going to give you the most redundancy and flexibility.

 

The scale of the job would normally dictate that you would need to contact a very good integrator and someone that has doen this scale of job before, sadly I am In Australia as I would have loved to help you (I will bethere at christmas though).

 

With 100 cameras every man and his dog will be PM'ing you to assist I suggest you speak to a integrator that has VERY strong links with a manufactorer, because if the integrator is worth his salt he will engage the manufactorer to assist in the design, ensuring no mistakes in calculation and design occur.

 

Firstly consider if you can afford the services of a Consultant, this way he can design and specify the equipment and this way you are covered.becuase the consultant takes the responsability of ensuring the end result for you and will specify the job and put it out to tender, he can also decide wh he invites to tender on the job, ensuring you get the best eductaed proffesionals.

 

Otherwise just pay me to fly over and I will consult for free for you LOL

[Thomas 0]

Given the sheer numbers, I would be looking at multiple machines for IP or analog. It's simply too much to depend on a single machine.

 

And for the record, the best layout is a 3x3 view. Casinos spent alot of time and money researching it.