Thomas in BP

With 32 cameras, quite a bit of bandwidth is required in order to store video off-site in a quality/bit rate that is worth anything. The risk that I see is that your are forced to reduce the recording bitrate to fit the bandwidth rather than choosing the much more expensive option to increase the bandwidt to fit the necessary bitrate. If the bit rate is too low for images to document anything else than that somebody walked in front of the cameras, then there is the risk that the insurance policy has small print saying that coverage is void if no CCTV system is in place or if recordings are of inadequate quality. This is a risk, but I am not saying that the insurance company has evil intent. The thing is just that I have heard far more stories about insurance companies doing everything they can to avoid paying than I have heard stories about insurance companies honoring legitimate claims without resorting to any tricks.

I second (or third) the Avira Antivir. If you only need Internet access because you need to access the DVR and you will do that from fixed IP addresses, then I can recommend a small application called Protowall, which is sort of a firewall on IP packet level. This does however require that you do not mind getting your hands a little dirty with some bits and bytes! Protowall is popular in the P2P community to block out computers belonging to the recording industry and government agencies, but it also works great if you want to limit network access to just a few known (fixed) IP addresses. It sits as a network device driver which will block IP ranges based on a blocklist file. I use it on my home PC to block more than 2 billion individual IP addresses among others edu ranges (a lot of attacks come from schools and universities either via live studens or zombie PCs), known zombie IP ranges, and even a lot of nosy companies and government agencies who have no business tracking my use of the Internet. In my case the blocklist takes up almost 10MB of physical memory, but if you basically exclude all existing IP addresses except for just a few, then you will only have a few lines in the blocklist and then the blocklist will only take up a few kilobytes of memory. Regardless of the blocklist size it uses very little processing power (less than a regular firewall). Most firewalls allow packets to go through the network adapter and into the firewall application in order to analyze whether packets and IP addresses should be granted access through a particular port or not. Even if access is blocked, there will still be communication between your network adapter and the network adapter of the blocked IP. With Protowall, the network adapter will not even respond to communication attempts by returning packets or acknowledging receipt of packets to IPs in the blocklist, so basically your PC appears not to exist at all. A locked door is good, but I think an invisible locked door is even better

Chisel, I am not sure what you define as high resolution, but if we are talking typical PAL/NTSC resolution then you may be fine if you find a 1 channel USB capture device and work with software video compression. An alternative could also be to buy a 4 channel stand-alone DVR and just use one of the channels. If you already have an old PC (for one channel you do not need anything fancy) then a USB device is most likely the most cost-effective solution for you. The added benefit is that with a USB capture device, most softwares (normally included with the device) will include recording of sound through the sound card of the PC and then you have an easy way of turning your VHS and other analog recordings of weddings, vacations, and family parties digital as well We, Comart System, are not the only manufacturer who offer 1 and 4 channel USB capture devices, so if you contact you regular wholesaler of CCTV equipment they should be able to tell you which of the USB capture devices are available through them.

Down Under, Each EU member state has its own national laws regarding privacy and data protection, however they are all primarily based on EU 'Directive 95/46/EC on the protection of individuals with regard to the processing of personal data and on the free movement of such data’, which is a collection of principles that all EU member states have legally committed to. Article 29 of this Directive provides the legal background for an official interest group, called Work Group 29, which provides interpretations and recommendations on 95/46/EC in order to ensure consistent compliance EU-wide. Council of Europe 'Convention No. 108/1981 for the protection of individuals with regard to automatic processing of personal data' has established that “voices and images are considered personal data if they provide information on an individual by making him/her identifiable even if indirectlyâ€

G'day down under, mate! I will go so far as to agree that MPEG-4 is a sophisticated codec family, but we cannot agree that it is leading the development race unless we decide which race we are talking about. If we talk about the preparedness for compression of megapixel video streams that will be a viable alternative to today's CCTV cameras some time in the future, then MPEG-4 needs a serious redesign in order to beat the performance of JPEG2000 on large resolution images when the race goes into the megapixel range. In most European countries you cannot get MPEG-4 certified for use in banks because they have very high requirements for the quality of individual images and lossless is often explicitly specified in the guidelines they have to follow. MPEG-4 has an edge in compression of high fps motion video by using P-frames, but if the race is to up the fps in banks and still meet their strict requirements, then MPEG-4 is incapable of winning. In some applications using P-frames is acceptable, but not in many other, so if the development race if for higher fps with complete images (I-frames in MPEG-4 jargon) then MPEG-4 will clearly loose to JPEG2000 and also to MJPEG though with less of a margin. I have nothing against MPEG-4 (Part 2 & Part 10/H.264) if it is in the hands of professional CCTV installers who know the technology and are aware of its limitations, but all the marketing hype and misrepresentations combined with the ability of any Tom, Dick, and Harry to buy a cheap MPEG-4 chip and set themselves up as a manufacturer of video boards has mislead a lot of IT people with experience in DVD ripping to believe that they are equipped to be professional CCTV installers and therefore a lot of CCTV customers end up with video recordings that are completely useless the day they have to document a crime - a quite important reason for installing a CCTV system to begin with.

Hmm, that sounds really scary, but I can't wait for the first Brit to challenge that on EU level. According to Council of Europe Convention No. 108/1981 it is established quite clearly that "voices and images are considered personal data if they provide information on an individual by making him/her identifiable even if indirectly." In combination with EU Directive 95/46/EC on processing of personal data and, especially, the 'Charter of Fundamental Rights of the European Union' in which Article 7 explicitly provides protection of "private and family life, home, and communication" I don't really see how they would be able to get away with that...Unless our law makers gets so scared by the big bad terrorists that they strip us for some of the central rights and protections we currently have. In the meantime I will continue to sleep well at night. BTW, who cares about what the IOC's code on audio recording is? They still have to follow UK and EU law regardless of what they do under the London Olympics.

Is 'auto tracking' acceptable/legal in the USA? In the European Union using 'auto tracking' would be deemed unlawful in most cases!

C'mon kangtree, gimme a break! I recognize that you want to sell as many of your H.264 video boards as possible and that you may be very proud of your product, but what happened to honesty in marketing and the principle about that advice should be based on the needs of the customer? I am sure your four steps are the best way to choose your product, but I do find your categorization of codecs crude and biased. I know it is very popular to pretend that H.264 is not MPEG-4, but in many cases that is just marketing hype in order to sell H.264 as a better mouse trap to people who already use MPEG-4 Part 2. I am not saying that MPEG-4 (part 2 & part 10/H.264) is bad per se, but it is definitely not God's gift to mankind and claiming that MJPEG is '3rd class' is just outrageous! If a customer's needs is given higher priority than a manufacturer's needs to sell a particular technology, then the choice of codec should be among the final conclusions of the analysis of the customer's needs, because MPEG-4, MJPEG, and JPEG2000 can each claim 'superiority' in different customer scenarios. You and I are not competing for the same customer segments, but I still would like to ask for a bit more sober product promotion from your side as my partners and I have to waste our time 'cleaning up' after a lot of the H.264 hype on a regular basis - even in our segments!

No matter if you use IR or not you should expect images to be distorted to some degree, because you will probably get a little bit of 'fish eye effect' where the sides of the image arches slightly and you also risk getting a 'ghost image' by the room behind the camera being mirrored in the window.

A dear child has many names, which is why MPEG-4 and H.264 are the same but different! MPEG-4 is actually a 'family' of codecs, but often people refer to MPEG-4 as a misnomer for MPEG-4 Part 2 (ISO/IEC 14496-2). H.264 is a different 'generation' of the MPEG-4 codec, the MPEG-4 Part 10 (ISO/IEC 14496-10). To make the confusion complete, H.264, MPEG-4 AVC, and H.264/AVC are all used to describe MPEG-4 Part 10 depending on which partner in the standardization group talks about the codec. H.264 is the name used by ITU (the International Telecommunication Union) while MPEG-4 Part 10 is the name used by MPEG (the Moving Picture Experts Group) which is a working group established in a cooperation between the International Standardization Organization (ISO) and the International Electrotechnical Commission (IEC). Some of the key improvements between the two generations are that Part 10 requires less processing power relative to higher compression rates than Part 2 and that features have been implemented to reduce the macro-blocking problem often seen when Part 2 is used at the limits of its compression capabilities. At the same file size Part 10 has better image quality than Part 2 and at the same image quality Part 10 has smaller file size than Part 2, so the main questions are whether there is justification for replacing an existing MPEG-4 Part 2 hardware compression card with a MPEG-4 Part 10 hardware compression card if the existing card is still working or how much it is worth paying extra to buy a card with MPEG-4 Part 10 rather than a card with MPEG-4 Part 2. These are two questions that can be hotly debated and I doubt there are any correct answers to them. It just depends! The closest you get to images that directly compares to a camera connected to a CCTV monitor is if you use a lossless codec. Some MJPEG codecs as well as codecs based on JPEG2000 are lossless. However, at low to moderate compression rates for MPEG-4 I think most people would not be able to notice the difference in image quality between lossy MPEG-4 and lossless MJPEG/JPEG2000 compression. I know that a lot of people are selling the H.264/MPEG-4 Part 10 as the answer from above and that they say it is much better than MJPEG/JPEG2000 and anything else in motion video compression ever invented, but I think that makes as much sense as saying that a screwdriver is much better than a hammer. If the most important feature is high compression for transmission over the Internet, then MPEG-4 Part 10 may be better, but if it is important to be able to pick out one individual image of high quality from an image stream, then MJPEG/JPEG2000 may be a better solution. Somewhere in between high compression and individual image quality, JPEG2000 my be the better option as it allows you to store the image stream in high resolution while you can pick out a lower resolution for Internet transmission without needing to recode the video stream. But, it just depends!

I think it sounds relatively expectable that the drives start failing at some point after around 3 years when you consider that it is the single component with the most mechanical motion in any PC or DVR. Especially if your clients are running their systems 24x7. When the HDD writes information on the platters it will write a certain amount of info before it moves the read/write head to update the file allocation table around the center of the platter. With continuous recording you have the read/write head moving back and forth an incredible amount of times every hour, so it is only natural that the HDD will be worn out at some point before most of the other components of the PC/DVR. If you read previously stored data at the same time as you write new data then the read/write head will be even more busy. Of course there may be some variations in the life-time of HDDs from different manufacturers, but unless your DVR uses some obscure brand of HDD, I don't think you should judge the quality of the DVR based on the survival of the HDD. I cannot give you a good article, but the following link may be interesting for you: http://ssdirect.com/ca/news/2005/05.htm Thomas

Hi Guys, I think it is fair to warn you that I am very excited about the products from Comart System, but I will try to be objective when I post For the second time I am new to the CCTV business. I worked with CCTV technology for a very short while back in early 2004, where I went under the name of CCTV_Source here on CCTV Forum. To cut a long story short, I left a company developing and selling fingerprint recognition technology in order to work for a small developer/manufacturer of DVR technology. It turned out that the guy owning the company had promised quite a bit more than he could live up to, so I only stayed there for about 2 months. After having searched for another company to work for in the security industry without luck, I started peddling international relocation services. As I did not really find the job of selling relocation services very stimulating I went on to work for a software company doing ERP/CRM systems. Selling software development services definitely did not do it for me either, so I continued to search high and low for the right job in order to get back into security. On October 1st this year I finally manged to get back into the security industry. Needless to say, I am quite excited about my new job as the products are good, the colleagues are nice, and the company seems to be a good place to work. My background is in export sales and marketing and through my career I have worked with export sales of pharmaceuticals, fingerprint recognition technology, software applications, and now I am (back) in CCTV. I have worked several years with sales and marketing of both physical and logical access control technologies with focus on system integrity and reliability of authentication, so I admit that I may have gotten a bit environmentally damaged, which makes me very conservative and sceptic so new technology needs to be much more than just new in order to impress me! I am a Danish immigrant in Hungary and I have been living in Budapest for eight years. My job is to serve new and existing partners all over Europe from the regional office of Comart System, a Korean developer and manufacturer of CCTV technology.

If you are running on Windows and you are worried about the temp of your HDDs, then I recommend that you try to access the S.M.A.R.T. info from the HDDs. I am messing around with making my home PC quiet and I use SpeedFan (http://www.almico.com/sfdownload.php), which gives me the S.M.A.R.T. info as well as info from a lot of other sensors in the system. For a HDD you should not let temps go much higher than 45˚C, for a powerful VGA card you are in the clear up until at least 90˚C, and the CPU, well depends very much on the processor, but if you are lower than 55-60˚C then you should be relatively safe (check the specs of the CPU). For all the components it goes that the higher the temp, the lower the performance (however marginally) and, more importantly, the lower the expected life-time of your components. If your HDD temps are less than 45˚C then the reason why the system crash may have to be found elsewher, probably the PSU. I have read many reports on PCs acting strange and irradically if the power supply is under-dimensioned. If you are using RAID with several HDDs then you put quite a bit more strain on the system than if you only had one or two drives. In general a HDD consumes around 9-10W, so if you have, say, 6 HDDs then you pull at least 40W more than on a 'normal' PC. If you also use a powerful frame grabber card, then you can add 8-10W more. In general, a 'normal' PC with typical specs pull somewhere between 250-350W, while gaming PCs pull up to 450W or more (with SLI). If your PSU is rated for 350W or lower, then I bet that is where the problem is to be found. The ambient temperature does play a role, but not a whole lot. If you have 35˚C ambient that just means that you will not be able to air cool anything lower than...surprise...35˚C However, the higher the ambient temp the higher the air flow you need to keep things cool. The ATX specs are quite old and were published back when a 200W system was enormous, therefore most PC cases (I have no experience with rack cases) still only have an 80mm fan in front of the HDDs and the air inlet is dimensioned accordingly. On my home PC I have dremmeled the inlet for better airflow and I actually use a tower case with room for two 120mm fans. As my goal is low noise I undervolt the fans to about 6V, but by just switching from an 80mm fan to a 120mm fan at standard 12V you may get up to 15-20CFM more airflow! Yate Loon fans are quite good I hear (they are standard in most name brand PCs), but personally I use S-Flex fans from Scythe which are a bit more expensive, but they have a better noise signature as they use 'SONY Fluid Dynamic Bearings' which promise double the life-time, but that may just be marketing blah blah. If you don't know ˚C, then (˚C x 2) + 32 gives you the ˚F with a low margin of error!

Kevin, The short answer is that if you get ~15fps per channel on 8 channels with your 120/100fps device, then, yes, it sounds absolutely normal. If you see a manufacturer mention the speed of their device in only fps (120/100fps in your case), then you can assume that it is the fps at CIF (NTSC: 352×240, PAL: 352×288) or half-D1 (360x240) resolution. As far as I know, CIF used to be the standard analog storage resolution while half-D1 is the digital equivalent. The standard output of an analog NTSC/PAL CCTV camera is actually 2CIF (NTSC: 704x240, PAL: 704x288) which corresponds to D1 (720x240), the full resolution for digital storage. That is why you will get half the fps mentioned by the manufacturers when you want to record full resolution. A typical A/D converter chip can handle 120/100fps (NTSC/PAL) in CIF, so if you see a device advertised with those figures you can be relatively sure that it uses one A/D chip for all the channels - in your case 8 channels. BNC connectors are relatively cheap, so the price of a device/capture card is normally not based on the number of channels, but rather on the number of A/D converter chips and other components, so you always get what you pay for - in terms of performance, image quality, and build/hardware quality. In my opinion, it does not make much sense to judge a system solely on the fps performance, because you can easily have a high-end system designed to deliver 1 fps per channel if that is what the customer wants and you can also have a low-end system designed to deliver 15 fps per channel. Image quality is relatively easy to judge, but you should always judge it on the quality of the playback, because on a system with 'real time monitoring' feature it simply loops the input video directly through to the monitor, so what you see (on your monitor) is not necessarily what you get (on your harddrive)! The build/hardware quality is more difficult to judge, as that is dependent on the electronic components chosen, but again price is a relatively good indicator. As I am not a techie, I judge build quality on whether the soldering and layout looks 'clean' and then I focus on the condensators (maybe capacitor in English?) on the PCB as that is the easiest way for a non-techie like me. If the condensators have a blank surface on the top, then they cost two-three times as much as the condensators with three lines engraved in a star shape on the top. Of course the more expensive condensators will have a longer life with reliable performance and a manufacturer who spends extra on the condensators is not very likely to have picked the cheapest A/D chips etc. That is the logical deduction I use in lack of deep technical knowledge about electronic components!