phred

Good idea regards

Good man ---- You're hired Attaching the camera housing in that way will ensure it becomes part of the same Faraday Cage so all voltages will rise and fall at the same time, just as the experts at www.lightning.org recommend. Also see this post by Rooney http://www.cctvforum.com/post-49997.html&highlight=#49997 who lives in an area where lightning is an issue. .

Excess light alone should not cause overheating by the mechanism you describe ie 'overvolting' CCDs have a very sharp cutoff in response beyond which additional light will not increase the output. The excess light will however cause additional heating as visible light is absorbed by many materials and becomes IR (quantum mechanics). It is actually pretty difficult to get rid of all IR without cooling the CCD. There are many different designs of CCD and some can tolerate more heat than others. Those with microlenses and filters made of plastics, for example, are more easily damaged by overexposure. The problem with using your video cameras is that they have not been designed for 24/7 use and the CCDs will not be designed with longterm overexposure in mind. I also suspect they may suffer from image burn as a result of viewing the same scene for long periods. (good questions btw )

CSG your post sums you up perfectly.

No

Ok jisaacmagee, my misunderstanding. Lightning is not something I would bother about here either. Doing a proper protection job is quite involved and at the end of the day you can still end up with toast. My objection was to the pseudo science nonsense and false claims about lightning protection and 'building zap' by CSG – I think the guy has been reading too many Marvel comics. In an area where lightning strikes are a real threat his suggestion would actually make matters worse. Regards

Very possible you have a damaged CCD. Sometimes the tiny wire connections give up. Effects of this can vary from complete failure to false colour rendering.

If you have ever used a magnifying glass to focus the sun on a piece of paper you will understand the problem. The damage to the CCD is usually caused by excessive heat build up as the lens focuses the sun on the CCD. I have tested some CCDs to destruction in a lab setup, usually they just stop working but just a few have exploded You really should do all you can to avoid pointing your camera at the sun. Reflected sunlight and car headlights are less of a worry - they just tend to ruin the image. Smearing is... as Kensplace says.. Google 'CCD smearing' you should find some examples.

Thank you for the reassurance - I know from past experience that what you say is true. I was however, referring to licenses for security installers. Perhaps a more likely explanation is that CSG is not who or what he claims to be.

Only a good earth bond will give you protection from lightning. Plastic insulators are useless at very high voltage – the only thing that comes close are those huge glazed ceramic isolators you see on overhead power cables or the Soapstone blocks used at the bottom of high power transmitter antenna - they still need special grounding in case of lightning strikes.

CCD technology has been pushed so far into low light territory for things like thermal imaging, I can't see CMOS catching up for several years, if at all. Cmos does have advantages in almost every other respect now - like 250 times the contrast ratio and twice the bit depth. What I would like to see is more cameras that have both types of imager cmos for day and CCD for night.

OK, you got me, that was a wind up just to get me on my soapbox

no ground - apart from the cable http://www.lightning.org/?page=faq

Two of the safest places to be in an electrical storm Inside a metal clad building or a motor vehicle. The metalwork acts as an efficient conductor allowing the energy to discharge to ground. Attempting to isolate a camera from a metal building is a waste of time. There is no insulator that will hold back 100,000,000 volts. Even if you could do it you would then create a situation where, if lightning struck the camera, the path of least resistance would now be the cable from the camera. Not a good idea to put 100,000,000 volts into a DVR. There is nothing you can do that will 100% protect sensitive outdoor equipment from the effects of lightning – the best you can do is provide an efficient earth conductor. Suggesting that a plywood sheet is protection from lightning is simply nonsense. By all means use it to stabilise the camera but please, drop the pseudo science crap. CSG calls himself a professional licensed installer, I can only assume that they hand licenses out in New York in exchange for a couple of cereal packet tops.

I was thinking more of the 'corporate culture'. HP used to be similar years ago while IBM were a bunch or arrogant ***** For such a large company Kodak are pretty far down the lawsuit table compared with Microsoft and Intel. Surprising when they hold over 1000 patents on camera technology alone. As for the Sony case they obviously agreed that Kodak had a valid point as they eventually paid up and signed a licensing agreement without going to court- same agreement was already in place with Sanyo and Olympus. Kodak also accepted some of Sony's technology as part payment. Courts also found in favor of Kodak in the Sun Java case. The same technology has been licensed to many, including Microsoft, for many years so why should Sun get it for free - Sun were effectively charging their own clients for technology belonging to Kodak. Many of Intel's lawsuits against Via failed despite their huge legal budget. Just thinking about it, I have lost count of the number of lawsuits Intel filed against Via and I don't remember too many succeeding.

Sure, put the two together and you have a potent brew. They are not all bad though – Kodak are really nice people.

I did, now YOU Goggle "Intel chipset problem" and give me a count on how many 'hits' you get. Most/many of the problems with the early Via chipset drivers were the 4in1 driver package from 5 or so years ago. Oh, as far as the "rocket science" part, I was using it in the extream. ---------------------------------------------------------------------------------- I was not offering an opinion, just pointing you to possible explanations. There is no doubt that Intel wanted to maintain a monopoly on chipsets for the P4 and patented several technologies related to the PCI bus and the way data was handled. This meant that anyone attempting to make a chipset for the P4 had to pay license fees to Intel. Via considered this unfair and produced chipsets for the P4 without paying the license to Intel. Intel sued Via in several countries – Via counter-sued Intel. It is remarkable that Via managed to make a chipset at all for the P4 considering their acrimonious relationship with Intel. My opinion, for what it is worth - Intel may have a legal right to protect their patents but they have an unhealthy appetite for bullying their rivals.

Just as I suspected. Thanks for the info

Google 'Via chipset problem' Creating something as complex as a chipset and bios engineering is probably more difficult than rocket science.

Don't know every card so can't say for sure what is best. Best bang/buck I can find in the UK is the VGuard RT4. Google VG4C-RT4 and you should find the same card re-badged under different names. All use the same software. Prices vary considerably.

All you need is a camera with 1million:1 contrast ratio. A lens with 20 f stops. Effective resolution approx 600 megapixels with a 130 degree field of view. That should match the human eye but you still will not be able to look at the sun. sorry for being a smartass but hope you see the point.

No light = no image, unless you use a thermal imaging camera (you don't want one -trust me). You will need to provide some light but a low watt soft light from a small lamp is all you need. You can move it around for best results. Unfortunately the lux figures they give in camera specs are often exaggerated. Brand name cameras tend to be more honest so don't be confused by what looks like a lower spec for a higher price. There are some inexpensive b&w bullet cameras – ones without LEDs but with grossly exaggerated spec - they should give you a decent image. For better quality nothing will beat a proper box camera and lens. Just as a guide, bright moonlight is 0.1 Lux Btw my own kids would freak if they woke up in pitch darkness. good luck

Yep, There is no surer way. Just to add to those comments - One of the benefits of the H.264 cards is that they have to have good chips. There are few to chose from with enough power to encode H.264 in realtime. The TI DaVinci (DM64x) chips seem to be what most people have turned to and they offer very high performance (probably still the fastest DSPs on the planet). Mpeg4 cards are more of a problem as everyone and his wife got on the bandwagon, buying in cheap mpeg4 encoder chips. There are some good ones amongst a mountain of dross. Like the man says - you know for sure when you have full hardware compression running, I managed to run a 4 channel H.264 card, recording 100frames/s 720X576 on a Semperon 1.8Ghz with very low %cpu. The one thing you have to be careful of is the H.264 codec used. The TI chip is not an H.264 ASIC. It is a complex 'system on a chip' that can run any codec you can write the code for. Writing code to run H.264 efficiently on the TI chip is a job for DSP engineers and is well beyond the reach of your average programmer. Those who buy in cards, write a GUI front end and use an off the shelf H.264 solution will offer poorer performance than those prepared to do the hard stuff.

Price difference can be more than 100%. Look at the price of intel Conroe CPUs – same core processor, some can run fast and some not so fast. There is a wide range of price/quality for CCDs Other factor in camera price is the performance of the processing chips inside the camera. If this was not the case then all cameras would be equal. Also survtech is right about Pixim, they are giving Sony a bit of a bashing at the high res. end of the market

Gracef, Do the two streams have different bitrate? Which H.264 codec are you using? Is it TI SDK codec or someone elses?