I'm A Firm Believer In What I Call "Just Enough" Surveillance + Casino Example

John,

We've actually taken the "Just enough" concept to its logical (and perhaps most controversial) conclusion. In 2013, based on our then-existing count of ~1,000 cameras, we specified that our new NVR/VMS system should have 960 analog encoder channels and 240 IP-camera capability. We purchased 100 720p cameras at the time and figured we would expand our IP megapixel count slowly, as the need arose. In fact, we are now at around 350 IP cameras and we still have nearly 650 analog cameras.

Why so slow? Budget is one reason. Suitable IP cameras aren't cheap; in part because we have specific requirements for number of usable streams, ability to handle casino-specific lighting challenges and type of camera/mounting requirements. 

We also decided to replace our PTZ cameras first, another very expensive goal. Our experiments with SD and 720p PTZs led us to choose mostly 1080p units. We found that the best modern PTZs have overcome most of the challenges analog and older IP PTZs had trouble with. Handling our lighting and providing a fast response to control inputs were among our primary goals. 

Another reason was our infrastructure. It is nearly 50/50 with an equal number of RG59 home runs and CAT5e 25-pair backbones terminated at each end in "66" punchdown blocks. Despite the addition of fiber from our Server Room to a number of IDF closets on our gaming floor, there remain challenges to edge networking our cameras. IDF closets were laid out to accommodate floor wiring with conduits going down to our IT Department in the basement and walker duct to the gaming floor. Pathways to ceiling-mounted cameras are often obstructed by things like HVAC ducting and fire blocks. The result is that we often have to resort to IP-over-coax or IP-over-single pair devices, which also add to the cost and complexity of deploying IP.

The other, and still the primary reason for our slow adoption of IP is the "Just enough" concept. We wonder why we would waste many hundreds of dollars covering a 10' by 10' room or a jackpot sign. Likewise, if an older analog camera dies, it's often vastly easier to just replace it with another analog camera and often, especially with older cameras, the new camera provides better image quality despite not being megapixel.

So despite IP camera purchases consuming a much larger share of our budget than analog, we are still buying analog in appreciable quantities and have no plans to stop until, and unless, we are given no other choice.

Cheers

Undisclosed,

We have no plans to eliminate PTZs. In our opinion, they're just too valuable. I've had similar discussions here in the past and at one time JH and I even discussed what it would take to eliminate PTZs entirely.  The conclusion was, I believe, that it would require multi-gigapixel cameras to totally eliminate PTZs, when they are used in the ways we do.

While it is an admirable goal to replace mechanical devices with purely electronic; Arthur C. Clarke himself said "The ideal machine would have no moving parts." (paraphrased), in practice, PTZs just offer too many advantages over fixed cameras. At least for us. 

We don't use PTZs as a primary camera except in very limited, temporary, cases. As a primary camera, we do use them to temporarily substitute for a failed fixed camera. We also occasionally use them as the primary for temporary promotions set up around the casino. 

But the primary application has always been as an adjunct to our fixed cameras. That is, at least in part, what allows us to deploy fixed cameras in the "Just enough" method. Basically, we cover every square inch of our floor with fixed camera overviews that always see what's going on with reasonable resolution. Then we use PTZs to zoom in on scenes like an incident or a jackpot or a person of interest or...  If we deploy 180/360-degree cameras in the future, they would basically replace many of our fixed camera overviews and PTZs would still be used to zoom in on the action. 

I would understand an argument that "Just enough" surveillance used as we do would only apply to applications that have both excellent situational awareness and where the cameras system is manned, as is required in Casino Surveillance. In our case, "excellent situational awareness" means we not only monitor the space ourselves, but we also have an excellent rapport with other casino departments, including Security, that don't hesitate to call us if they see something they feel we should be aware of.

I can't speak for Carl, but the Casino's I work with covering more with less is nice in some areas, but in others we kind of need to focus on specific areas. Like table games. We don't really want to see two tables at once. We want one, well actually two cameras per table. One general overview of the patrons of the table, and then the overhead view of the cards and the chips. Putting a 4k camera above a single blackjack table, really isn't going to gain you a whole lot. Just going to increase costs on the camera. 1080p is more than enough.

I believe with some of our tables it wouldn't even be possible to cover two tables via the overhead shot. The angles would be poor.

I do get what you are saying, but things are just a bit different in casino's, or so I think. It's been an adjustment for me, as I just did regular surveillance for like office buildings and such before, where you were trying to cover as much of an area as possible, while maintaining good clarity and PPF. Heck most customer didn't even really care about PPF, just wanted to cover as much area with one camera as possible. But at the casino's I work at now it's different.