Winning the Race to the Very, Very Bottom
THERE’S been a lot of grumbling from well-known quarters lately about the race to the bottom when it comes to price in the video surveillance industry – the way innovation has fallen by the wayside as manufacturers chip away at the performance of their biggest selling cameras by reducing quality in multifarious ways.
It’s not just about trying to hang onto margin in a maturing market. The way systems are purchased on the basis of price, forcing manufacturers to play low-cost ball or lose jobs is a diametric that feeds into the process. What’s interesting as an observer, is that price and performance are materially linked when it comes to CCTV cameras and there is a point past which the loss of performance does not make up for the dollars saved.
The challenge for installers and end users is trying to work out where this point actually is. Depending on your application, there may be no need to spend $A2500 trade price on the latest bullet camera, when another camera and housing combination will do the task just as well for $500.
Before we go on, I want to hammer home a point. You can buy cameras for less than the best cameras that may be lacking desirable features but are close in on-screen performance. However, something that’s noticeable when testing cameras is that very low cost units really do offer much reduced performance, even if manufactured by reputable makers.
In my opinion, the race to the bottom really is a race to the bottom – but it’s not price I’m talking about. Maybe one day in the future people will goggle at the amazing contrast and resolution of vintage Vidicons and say to each other: ‘they don’t make them like that anymore’. In fairness, they do make them like that anymore but cheap prices have a way of anchoring the human brain to a miserable diet of bad performance and that's disastrous for serious security investigations.
There’s no doubt in my mind that we’re well past the point at which low cost means performance so wretched the savings are no longer worth it. Speaking plainly, there really was some crap in the way of low cost cameras at ISC West last month, particularly cameras with cloud back-ends. It’s curious to me that users of cloud surveillance systems are categorised as a group with no need to clearly identify any moving person or vehicle numberplate recorded by their CCTV solution, ever. Sure, low upfront hardware costs are important to a recurring revenue business model but not at the expense of a security system's core functionality!
But it’s not just cloud solutions providers which cut underlying performance to the bone. Some of the corners being cut in the wider CCTV market are plain nuts. If you can buy very affordable cameras of excellent quality, why buy total junk for half the price? Lenses wrought by distortion, complex aberration and spooky green ghosts; processors that can’t keep up with the gentle wafting of summer trees, bitrates that are eye-watering at 1080p and simply unmanageable at UHD.
Then there’s the re-introduction of low light noise, strange colour renditions in strong or low light with (or without) WDR, a combination of tone mapping halos and slow shutter blur extending from one edge of the image to the other that disappear, by colour, in weird, temporal pulses. And finally, that wondrous combination of blooming and veiling flare in strong backlight that so much brings to mind the last opening of the gates of heaven.
How can new cameras be worse than those that came before them, security people? It’s not that there aren’t affordable and capable CCTV cameras available on the market today, because we play with them all the time. Why is the first generation 1.3MP Pelco Sarix (built in the digital dark ages of 2009) that I randomly plugged into my workstation last week seeking contextual inspiration, so much better in very low light (and so much easier to network!) than many of the latest compact domes?
The sad truth is that CCTV solutions are reaching that point of ubiquity long ago inhabited by glum-faced alarm sensors, prox readers and network routers. But users are mistaken if they believe the exquisite balance of sensing, processing and optics that goes to make a competent video surveillance camera can be stamped out at no cost, will function in every application, with no further consideration.
There is a way to win the race to the very, very bottom when it comes to camera performance. Set concrete operational parameters and insist the cameras you buy exceed them. And make sure you come along to SecTech’s Camera Shootout in starting in Adelaide this coming Monday, May 9 (register here for free parking!). You’ll see 24 cameras tested in 3 key groups – Best Low Light/All Rounder, Best Compact Entry Dome and Best 4K/5MP. If you care about camera performance, it’ll be well worth your while. ♦
By John Adams