Top 5 IP Camera ProblemsBy John Honovich, Published on Jul 25, 2008
The overwhelming majority of people buying cameras today do not choose IP cameras. While most people see the move to be inevitable, serious debate exists on how long it will take to get there. Recently a number of analysts have even slid back projections for when IP camera sales will overtake analog.
To figure out why you should select IP cameras and to figure out when the mass of users will buy IP cameras, we need to honestly and clearly assess what is holding back IP cameras.
[Note: An updated 2009 version of this report is now released.]
The common explanation is a lack of training and a lack of support by traditional integrations and manufacturers. I think these problems are secondary and a consequence of more fundamental problems. Here they are:
- IP Cameras are too Expensive Compared to Analog Cameras
- Storage for Megapixel cameras is too expensive
- Smart Cameras are still in their infancy
- DVRs offer limited support
- Lack of Integrator Training
IP Cameras Too Expensive
It is common knowledge that IP cameras cost more than analog cameras. Compare a Pelco CC3701H-2 to a Axis 211. With the same lenses, the Axis is double the price of the Pelco ($600 to $300). However, this is not the main point I want to make.
To identify the real issues, we need to understand the flaws in Axis well publicized TCO Report on IP vs Analog. The report claims that systems of over 40 IP cameras are cheaper than analog. There is a lot in the report and it is certainly worth reading.
The fundamental problem of the report (and IP camera expense) is that over 90% of customers already have analog systems – DVRs, coaxial cable, analog cameras, etc. The report assumes a greenfield installation with no coax in place and no analog cameras. This is not representative for the overwhelming majority of customers. Once you factor in customers that have analog in place and are looking to upgrade, the costs look far worse for IP camera systems. I would encourage Axis to conduct a follow-up report using this more realistic assumption.
Moreover, being 5% cheaper than an existing technology (which is essentially what the Axis report claims) is rarely sufficient to motivate customers to switch. Indeed, in the applications where IP has been most heavily adopted, it is the elimination of analog video fiber networks that has provided the ROI necessary to migrate to IP cameras.
To make people switch, IP needs to be either significantly cheaper or offer significant economic benefits that analog can not offer. As I reviewed recently, I am encouraged by the ability of ACTi to provide solid, low cost IP cameras that can narrow the cost gap. On the other front, to increase economic benefits, megapixel cameras and video analytics hold the most promise. Let's now examine them.
Megapixel Storage Too Expensive
Compared to analog cameras, megapixel cameras can increase cases solved and reduce camera counts. The biggest economic problem for megapixel cameras is the cost of storage. Almost all megapixel cameras in production today use MJPEG encoding which is 2x to 4x less efficient than the codecs used for analog cameras. Combine this with the massive increase in resolution of megapixel cameras and the cost of storage per camera can be $500 - $1500 per camera up from $50 - $100 for analog cameras.
This is a significant tax that many customers are justifiably concerned. You often hear from vendors that this is not a problem. I believe that for certain high end customers this is true (remembering less than 2% of all cameras are megapixel). However, this will not be the case for mainstream customers. The extra cost for storage will make it very hard to justify mainstream deployments of megapixel cameras.
Yes, H.264 is coming but the questions on how well it will work for megapixel cameras are significant and unanswered.
Smart Camera Infancy
Another way IP cameras can add value is by embedding video analytics into the camera. The challenge here is that smart cameras are not commonly available (even from Axis) and that big questions remain about how well smart cameras will work. Making this even more challenging is that DVR manufacturers are putting analytics directly inside their units. This would extend the life of analog cameras making the case for smart cameras harder.
DVRs Do Not Support
Many DVRs offer limited or no support for IP cameras. This certainly reinforces the problem for IP cameras.
While many IP vendors turn this into a morality play, the lack of support is a reflection of a weak business case. I often hear claims about the DVR companies are stupid or greedy. At the heart of it, I believe the real barriers are fundamentally issues of economics – the problems I listed above. Once those problems are resolved, good DVR companies will be be motivated to support and bad ones that refuse will be quickly crushed.
Lack of Integrator Training
I see the same issue for security integrators as for manufacturers. Security integrators correctly see that IP is not ready for most of their customers. As such the motivation is weak. At the same time, security integrators are still doing well and growing. Again, once IP cameras solve their problems, security integrators will be forced to support or will be displaced.
None of the above means that any specific customer should not use IP cameras. Use of IP cameras depends on specific application and logistic uses.
However, I am contending that until the 3 primary problems are solved I do not believe IP cameras will be selected by a majority of customers. The solutions of these problems are:
- Standard Definition IP Camera only costs $100 more than equivalent Analog Cameras
- Megapixel cameras support H.264 and H.264 has no serious side effects on client or servers
- Smart Cameras are widely available and the analytics work reliably
Once these solutions are delivered, the business case will become strong across the board. Integrators and DVR manufacturers will then be forced to support IP cameras or be ousted by rivals that offer the clearly financially preferable IP solution.