How do you think of HDcctv? It can be a sustitute for IP?

[Soundy 1]

Fine, I'm sold. So, recommend me some HDcctv cameras I can go buy right now. And a capture device.

 

No?

 

Hmm.

 

 

Comparing vaporware is pointless.

[survtech 0]

I said I am keeping an open mind, not jumping in with both feet. I see some real possibilities for HDcctv, especially in casino deployments where the infrastructure for IP doesn't exist and would be prohibitively complex and expensive to install; where frame rate and latency are real issues and where we don't want lossy compression at the source.

[hardwired 0]

Umm, problem? From the HDCCTV page "New and existing installations can use CCTV industry standard coaxial cable (RG/59, RG/6 and RG/11)", not any words about passing through twisted pair, probably the most likely infrastructure in their biggest potential target market, casinos?

 

For my uses, the ability to use multiple transmission medias, often changing midpoint in a single run, is one of the biggest advantages of IP protocol cameras.

 

Also, I doubt that SDI type transceivers and capture cards are going to reach the price of a Ethernet transceiver anytime soon.

 

I have no doubts of the technical superiority of HDCCTV, just that it's benefits are only applicable to such a small niche that it is unlikely to make a large market impact, or be cost effective in many cases.

[Soundy 1]

For my uses, the ability to use multiple transmission medias, often changing midpoint in a single run, is one of the biggest advantages of IP protocol cameras.

 

One of the drawbacks to standard analog CCTV is that it's a point-to-point topology, and there's really no way around that (at least not without extreme expense).

 

HDcctv does little to change that - although "hubs" to consolidate/collect multiple signals and send them over a single line should be POSSIBLE, will existing cable be able to handle the combined bandwidth of three or four cameras? For most intents and purposes, it's still a point-to-point system. And I'd like to know how one would integrate, say, a camera at a remote site into a system using this method? With IP, it's easy - run it through the Internet. No, you won't get the massive bandwidth necessary for 3MP at 30fps... but you CAN GET THE SIGNAL THERE.

 

Also, I doubt that SDI type transceivers and capture cards are going to reach the price of a Ethernet transceiver anytime soon.

 

I have no doubts of the technical superiority of HDCCTV, just that it's benefits are only applicable to such a small niche that it is unlikely to make a large market impact, or be cost effective in many cases.

 

Thank you, that's exactly what I've been saying all along. Any new technology will be initially expensive, to recoup development and tooling costs. Economy of scale is the main thing that drives prices down - as more people buy them, the required profit margin for the manufacturers gets smaller. The more of something that a given assembly line churns out, the less it needs to recover per unit to pay for itself.

 

I don't know that HDcctv would be "technically superior"... from the descriptions, it has some advantages, but still has other limitations compared to IP cameras. Neither would usurp the other outright for every situation. The problem is, while HD continues to be developed, IP prices continue to fall, and the limitations continue to be addressed and improved upon. A 2MP camera that does 30fps will eliminate one of the main stated benefits of HD.

[survtech 0]

I agree. I have been pestering the Alliance since day one to include twisted-pair as a transmission medium. But then again, IP will not pass on a single twisted-pair cable either without a $600 Nitek VR124UTP.

 

Point-to-point is not necessarily a drawback, especially from a signal transport reliability standpoint. You only lose one camera with a bad connection or whatever; not tens or hundreds!

[hardwired 0]

I've been using twisted pair/ Ethernet converters for a while now that are significantly less than that price... and the same company just came out with a converter for coax using HPNA 3.1 that can go up to 100 Mbps bidirectional, point to point or multipoint, up to 4000 feet at lower rates... for $220 for the pair! It was designed for foil RG59, but it's operating frequency band is 12-44 MHz, and attenuation on copper core RG59 is lower than foil in that band.

 

Just ordered some to try.

 

Going to try DC power injectors on the same line to see if I can make a power link on the same cable, but most of my long run coax installs have remote power supplies, but if I could localize power, too, that would be a bonus.

 

Survtech, think through that one, you probably could run an IP camera that also has analog output up the same coax, keep your matrix (no latency for ptz, no forklift upgrade, etc.), and peel off megapixel video for recording. HPNA specs have QoS, too, so you should be able to control the stream pretty well.

 

I'll try feeding baseband video through my test pair and see how it works, what filtering might be needed, etc, and let everyone know.

[megapixel man 0]

I've been using twisted pair/ Ethernet converters for a while now that are significantly less than that price... and the same company just came out with a converter for coax using HPNA 3.1 that can go up to 100 Mbps bidirectional, point to point or multipoint, up to 4000 feet at lower rates... for $220 for the pair! It was designed for foil RG59, but it's operating frequency band is 12-44 MHz, and attenuation on copper core RG59 is lower than foil in that band.

 

Just ordered some to try.

 

Going to try DC power injectors on the same line to see if I can make a power link on the same cable, but most of my long run coax installs have remote power supplies, but if I could localize power, too, that would be a bonus.

 

Survtech, think through that one, you probably could run an IP camera that also has analog output up the same coax, keep your matrix (no latency for ptz, no forklift upgrade, etc.), and peel off megapixel video for recording. HPNA specs have QoS, too, so you should be able to control the stream pretty well.

 

I'll try feeding baseband video through my test pair and see how it works, what filtering might be needed, etc, and let everyone know.

 

Hardwired, we are both thinking along the same lines and I did mention this to him previously, however he has reservations of the Analog to IP converted images viewed on LCD screens, apect ratios etc... I am currently in discussions with a large casino and this is what are planning to do. Keep the Matrix for PTZ controls and live viewing on analog monitors, convert the Analog feeds to IP via encoders after the matrix for recording to NVR's, add megapixel cameras from 2MPx up to 16MPx resolutions.

 

Back to the thread topic. HDcctv in concept has potential for some applications, the proof of concept will be in the detail, and in this regard I have many many questions in my head.

 

"Bring it on" , less talk and more show.

[survtech 0]

I've been using twisted pair/ Ethernet converters for a while now that are significantly less than that price...

 

Please give a link.

[hardwired 0]

Survtech: Here's the link for the twisted pair products / new coax adapter http://netsys-direct.com/index.php

 

Megapixel man: I was referring to Megapixel cameras with additional analog output directly at the camera (Sanyo VCC-HD4000, Panasonic WV-NP502, etc.) with coax adapters at the camera combining the feeds and then splitting at the matrix for the analog and IP feeds.

 

BTW, google G.hn for what I think is probably going to eat HDcctv's lunch.... Up to 300 Mbps on coax, twisted pair, or powerline..

[survtech 0]

BTW, google G.hn for what I think is probably going to eat HDcctv's lunch.... Up to 300 Mbps on coax, twisted pair, or powerline..

Very interesting technology. Now, if they could only get IP latency down to <50ms, they might have my vote.

[megapixel man 0]

As per the promoters of hdcctv, products were due to be launched at ASIS 2009....

 

Did anyone see it, or is it still a myth

[Soundy 1]

Vaporware. They're probably discovering that all the "obstacles" they were hoping to overcome from IP are obstacles to them as well.

[survtech 0]

I met with Todd Rockoff, Jean-Michelle Florent and Gloria Reiss of the hdCCTV Alliance at ASIS 2009. Although I can't attest to the extent of progress toward the group's goals, I will say that they expressed a lot of interest in our needs for High Definition cameras and equipment.

 

They do seem to be taking my reservations toward the technology to heart: Their latest FAQ claims that they are working on extending the maximum distance on RG-59 to 300 meters without requiring repeaters.

 

Now if only they accomodated signal transport via single twisted-pair; had some major camera, DVR/NVR and signal switching manufacturers producing products and were demonstrably cost-effective, I would totally jump on board. Since neither appears to be forthcoming, I remain skeptically optimistic .

[megapixel man 0]

I have heard comment from a respected Surveillance editor that the technology proposed by the hdcctv alliance is akin to "putting wheels on a horse".

[survtech 0]

What is a "Surveillance editor"?

 

Edited September 27, 2009 by Guest

[rory 0]

either way. .. coax isnt going anywhere anytime soon.

[Soundy 1]

I have heard comment from a respected Surveillance editor that the technology proposed by the hdcctv alliance is akin to "putting wheels on a horse".

 

Yeah, I could go along with that.

 

They're building a system that's going to require completely new cameras, and completely new capture hardware, and it's still going to be limited to ~2MP (assuming they adhere to HDTV spec). The main purpose of megapixel in the first place is to get past current resolution limitations of analog video, so bumping up against another pre-defined limit before you even get started does seem pretty backwards to me. It still won't address issues of poor low-light performance, because that's a limitation of physics and sensor size.

[megapixel man 0]

What is a "Surveillance editor"?

 

 

Someone who writes articles on Surveillance systems. "A magazine editor".

 

Good pic.

[survtech 0]

What is a "Surveillance editor"?

 

 

Someone who writes articles on Surveillance systems. "A magazine editor".

 

Good pic.

Whewwww! For a minute there, I thought you were talking about someone who edits Surveillance video footage.

[tesc_cctvpro 0]

It's great that we are discussing this.

 

There is no doubt that Megapixel surveillance is the future of VSS. Will it be IP or will it be HDcctv?

 

One thing that we should not forget is that Television is the forebear of CCTV. Closed Circuit Television developed as an offshoot of commercial television. While we where all engaging the transition to IP Surveillance - we overlooked the evolution of the standard television - which has rapidly emerged to HDtv. HDcctv is a natural evolution which would have occurred regardless of IP surveillance.

 

One point worth noting is the recent agreement between the HDcctv alliance and SMTPE. In terms of this agreement - HDcctv has been kickstarted with a highly developed standard set - adopted from the motion picture industry. The kind of resolution we are talking of is nothing new to the motion picture industry.

 

Regarding the challenges facing IP - many are mute points. Megapixel surveillance requires almost 1/3rd the number of cameras to cover a given area than does analog. So we should actually be talking total pixels per system. For example a 4000 pixel requirement at the forensic level would require about 13 analog cameras. 4 Megapixel cameras would get the job done -- but we are still talking about 4000 pixels either way.

 

Indeed the larger image size does increase CPU load, and if we are using H.264 then even more so. The IP surveillance industry has addressed many of these issues though with technologies such as cropping, ePTZ, adaptive mode recording, H.264SVC, Multiple streams and onboard recording.

 

For me there are three core issues which make HDcctv more appealing than IP Surveillance.

 

(1) The ability to re-use the existing cable and power infrastructure when upgrading from analog to Megapixel.

 

(2) The fact that HDcctv will mostly be based on equipment and technologies that we are already familiar with. Many of the network management technologies that have emerged - as cctv is force fitted to IP networks , make the system more complex, more IT in nature. The complexity means numerous more skill sets to be learned by technicians. I don't doubt that some of these technologies may be necessary in the HDcctv field - but generally, i believe - the average technician is going to be more comfortable with HDcctv than IPcctv.

 

(3) Comparing apples with apples, a real time HDcctv camera is already priced up at 1/3rd to 1/5th the price of a comparable Megapixel, realtime IP camera.

 

Of course we also have the benefit of motion picture standards governing more than just pixel counts - but guaranteeing color fidelity, etc.

 

I am not saying that talk of increased bandwidth, storage and cpu processing is undue, just that it should be considered, in relation to the reduction in number of cameras that results, when switching to the wider field of view attained by HD megapixel cameras.

 

The difference between IP megapixel and HDcctv is that IP is expected to handle both encoding and transmission for storage as well as remote access/transmission on the same media. HDcctv will handle storage encoding at a local site and bandwith issues are mostly limited to transmission from the local site when required.

 

IPcctv has tried to address this issue by building onboard storage into the camera itself. This of itself creates additional network and bandwidth issues when we start to look at redundancy. In addition SD card storage doesn't present the same level of usability convenience for live viewing as we would see in a non-network live view of HDcctv.

 

It must also be said that analog will still be a best fit solution for many installs, HDcctv will be a best fit in many cases, and IPcctv is going to be a best fit in others. We need not adopt an either or approach.

 

Brandon Joubert

TESC (SA)

[survtech 0]

Megapixel surveillance requires almost 1/3rd the number of cameras to cover a given area than does analog. So we should actually be talking total pixels per system. For example a 4000 pixel requirement at the forensic level would require about 13 analog cameras. 4 Megapixel cameras would get the job done -- but we are still talking about 4000 pixels either way.

 

Brandon Joubert

TESC (SA)

Brandon,

 

That assessment would only apply to limited situations. I can envision some places within our casino where megapixel could replace multiple analog cameras. On the other hand, there are many other applications where that would not be true: I don't foresee using one camera to cover multiple gaming tables, for instance.

[cglaeser 0]

I'd find a 5MP (2560x1920) resolution at 10fps (IQEye 700-series) more useful than 1920x1080 at 30fps...

 

Agreed. Assuming, of course, the same low light performance, which is a challenge for the higher res smaller pixel sensors, but I agree, I'd rather have 3x pixels @ 1/3 fps. 10 fps is just fine for many security applications.

 

Best,

Christopher

[tesc_cctvpro 0]

Thanks for the views survtec.

 

I think the situation applies in ALL Cases - as it's purely mathematical.

 

A proper response to your question would demand a brief mention of the classification of surveillance goals. viz -

 

- General Observation

- Forensic

- High Detail.

 

Regardless of what camera you are using - for general observation you will need about 60pixels per meter of horizontal area, for forensic detail 120pixels, and 240pixels for high detail - as would be required at a gaming table.

 

The pixel requirement doesn't change.

 

So from a factual point of view a camera with 320 horizontal measure is going to cover less horizontal field of view than a camera with 1080 horizontal measure. Thats pure maths.

 

one 1080H camera will cover the same area as approximately 4 cameras if we are looking at general surveillance. Make the necessary adjustment for Forensic or High Detail - but it's all purely maths. A 1080H sensor will have a far wider field of view than a 320H camera in ALL CASES.

[tesc_cctvpro 0]

Of course after rereading your position survtech, I would agree 100%

 

My stance is consistently that of "no one size fit's all".

 

 

While it is not viable to have one megapixel camera covering multiple gaming tables, there are obvious benefits to having one megapixel camera replace one analog camera.

 

But indeed - Megapixel, wether HDcctv or IPcctv will not be a best fit in ALL Cases. I do not envisage the total demise of analog for some time to come.

[Erron S. 0]

A couple of things came to mind as I read through this thread.

 

There are companies that are sending IP over coax and doing it very well.

http://www.veracityusa.com/products/products.php

 

There's one thing we really need to consider when we look at these new and 'easier to install' technologies and that's the installers worth in the industry. What I mean by that is over the past 3/4 decade or so the DIY market had increased and began to make it harder for a dealer/installer to make a buck in the industry. What's the dealers 'worth' if you can have your brother-in-law install the system? (meaning no extensive knowledge is needed for a cctv install) This is what can and does put many dealers out of business. How many times has someone lost a job to Sam's Club or the like? Something to consider as we promote these 'plug and play' technologies. It does make it easier on the installers until the end user market gets a feel for it, then it's just the end user doing the install.

 

Having said that, what I'm seeing in the market place is a few things. Either the dealers are getting more education on IP infrastructure and growing their 'value' to the end user or they are hiring guys that are aready trained in networks. I'm also seeing a number of network admin types getting into cctv because it's a fairly easy adjustment when they have IP knowledge already. Not to mention that most new buildings are installing networks for not only the internet but also for phones and cctv gear. They are planning ahead from that standpoint and that looks to be the forseeable way of wiring a building. Have any of you guys seen new construction that pulls coax during construction as a 'just in case' measure? I sure haven't. Not only that, but any medium to large sized building typically has an IT guy on staff or on call just to handle their IT infrastructure after the install is completed.

 

Either way, it will be interesting to see if any of these new technologies take hold or if the IP standards will prevail. Time will tell that's for sure.