Soundy

I fully respect the quality and reliability level of a system like this. thewireguys touched on the constant light levels and since the thermal cameras don't use light you're both saying the same thing. I guess if an installer put in artificial light and wrote more than one alogarithm for the software detection of the camera it might work in a simliar way as a thermal system. But nowhere near as well. Thanks a lot for all your help. CK Actually, the main factor in our system's reliability is the video analytics. Standard motion detection just looks for short-term (within a couple seconds) changes in the image - shadows moving with the sun and lighting changes from day to night won't trigger it, it's just sampling and comparing pixels for variations within a span of maybe 10-100ms. What the analytics does, is look for SPECIFIC movements. For example, "tripwire" is a common, very basic detection type for most kinds of analytics: watch for an object crossing a line that the user defines. This is often refined by watching for movement in a specific direction, or crossing one line and then another. The system can also be programmed to recognize types of objects based on their size - say, to differentiate between a person, a car, or an animal. You can then combine these to, for example, trigger an alarm if a person or car crosses a fenceline, but not if an animal does it. Or you could have to trigger if a car enters a fenced area, but not if it's leaving the area, and not if a person enters or exits. A lot of that ability has to be defined by the user: for example, you'll draw boxes on an image to show the system what size a person is, what size a car is, etc. Doing this at the "front" and "back" of the scene will let it know the perspective it's dealing with, so it can interpolate the sizes of those objects at points in between. It can take a good deal of time to fully set up the system for a detailed level of "intelligence".

Good in some instances... depends on how many cameras you have and how much stuff you want to save. Drives are cheap on their own, but once you get into needing five or six 1.5TB drives, you have to start factoring in the cost of enclosures and/or controllers for them as well. Some can also record at a slow speed on a constant basis, then kick to higher framerate, resolution and quality settings when motion is detected. We do have software-triggered monitored alarms on one site... the alarms are triggered by video analytics software (much more advanced that just basic motion detection), and the cameras are FLIR thermal cameras, which are unaffected by varying light levels. Of course, that's not a CHEAP option - I think the FLIR/VA system alone ran around $80k for the hardware... installation to a 50' tower in the middle of a two-square-block storage yard not included.

Just as a side note, this is why megapixel cameras aren't generally as good at low-light as analog, too... and the same reason pocket snapshot cameras have lousy low-light performance compared to digital SLRs and other large-sensor cameras. Think of the sensor as a field of buckets, and the light photons as raindrops. The bigger the buckets, the more rain you'll collect in a given period of time (say, 1/60 of a second), and every raindrop gives you more data about the water you're collecting. If you want to increase the number of buckets (pixels) in a given size of field (sensor), you have to make the buckets smaller and smaller... so they can't collect as many raindrops (photons) and thus won't provide data as accurate as you get with the bigger buckets. The only way to improve the ability to collect while maintaining the resolution is to increase the size of the field to allow for larger buckets... or in our case, using larger sensors. Beyond that, you either need to devise a better way to process the minimal data you ARE getting, or you need to come with a more efficient bucket.

Ahhh, got it. Okay, consider me enlightened. Thanks Back to the original problem then, he's not likely to be able to get around using conduit to run the wiring - not because of Pelco's specs, but because in most of the civilized world, electrical code would require it

Baluns don't REQUIRE twisted pair - I've run video baluns over station wire (four un-twisted 22ga conductors), I've run them over Cat3, even over 18/2 speaker wire. That's just FYI... I can't think of any possible causes or solutions right off the top of my head. Sounds like you've done pretty much all the necessary troubleshooting steps. The only question I might have is whether you tested the voltage AT the camera with it powered up, to make sure it's not dropping too much, but you've tried powering the cameras locally as well, so... Only other thing that comes to mind is that it IS a bad batch of cable with overly high resistance - not enough that the cable tester fails, but enough that the voltage and signal are dropping below usable levels. What sort of Cat5 tester do you have? Will it give you things like capacitance and resistance/impedance readings, or is it just a continuity tester? I would assume either way, it's showing all the pairs are wired correctly? That's the only other thing I can think of, if the pairs are mis-matched...

Uh... is there not a version of the Arecont software that will run on Win2K? I went to add an AV5100 to a Win2K-based Vigil today and the software install says it only supports XP or W2K3 Server. Now how stupid is that??

Thanks... so the bead isn't required for "proper" operation, just to fit the rules. Good to know ...Assuming the cam does not generate enough interference to lock up a switch, etc..... Not likely, but a possibility, nonetheless. My point was that the beads were just another sign of a design that wasn't fully thought through. Yeah, I'm noticing a lot of that with these cameras.

Why would you not just run the control cable separate from the AC??

Hmm, without looking at the manual, just a couple thoughts: I doubt that the control wiring is 115V; though it may be powered with standard line AC, control is usually via low-voltage RS-232/422/485 (+5 or +12VDC). Either way, use of conduit for the AC is likely required by electrical code, and if so, there's no way around it, unless your local code also allows BX. Even if the platform does use 115VAC for its control signals, it would still have to conform to your local code.

Thanks... so the bead isn't required for "proper" operation, just to fit the rules. Good to know

Silly thing is, it probably does... at least, the finder utility DOES work with Win2K... but the INSTALLER won't let you install it without XP. How dumb is that? do you need the installer though? maybe just extract the required files, and put together your own installer? if i can help let me know. Well, I've done that NOW... but the full package including NVR has to be installed on something first to extract those files; you can't just unzip the included installer and have the files there. At least, that's the case with the software that Arecont makes readily available, both on the bundled CD and on their website. That's what 99% of the world is stuck with. They don't even tell you that there's a smaller "stand alone installer" finder/configuration utility available that will run without needing installation, which ak357 sent me. Why wouldn't an intelligent company include that on the CD or on their website?

The big advantage of the Super Dynamic series has traditionally, of course, been their dynamic range and handling of widely dynamic scenes - that's a big part of why a CP484 runs $600+ compared to about $250 for a CP414. From what I'm seeing, this thing carries on the SD tradition quite well. Granted, the windows are obviously tinted (look outside when they open the door), looks like a 20%-30% tint, but even then, the highlights outside should be blown out on most other cameras given the inside lighting level. And please, barrel distortion? Really? This is a major issue? It sounds like some people have never seen it before - you'll get it on 99% of all cameras, including $8000 DSLRs, when using wide-angle lenses. Correct it? Who cares? cglaeser is right, no cops are going to reject an ID because of it, and no court will toss out evidence because of it. Keep in mind what the PURPOSE of this type of video is, people - this is for security and identification purposes, not high art or making movies. We have a client who wanted an MP camera for an ID shot on their flagship store a couple years ago, and we talked them into a CP484 instead because no MP cameras at the time would handle the severe backlighting. They'd still love to use MP for that purpose - if this lives up to the Super Dynamic name, they'll want to use it, for no other reason than it gives them the high resolution in extreme lighting conditions.

Maybe you can measure the difference with VLC and some random USB adapter... (I'll go sit quietly in the corner now...)

Wow, a lot of people having trouble with reading comprehension. Did anyone actually pay attention to the original question, or is it just the usual wave of spammers with their "MY PRODUCT IS PERFECT FOR ANYTHING YOU NEED!"? Hint: "If I wanted to mount cameras on a race car..." PoE is not going to work well. You'll either need a PoE switch or injectors that run off 12VDC, or you'll need an inverter or some other source of 120VAC in the car. Most race cars do not have such luxuries. "Professional" level cars like stock cars and open-wheel cars are pretty stripped-down. And yes, if he wants something suitable for viewing sponsor logos on underbody parts, he WILL need to eliminate "cheap" from the equation - anything that will stand up to that kind of beating will be pricey, and anything cheap won't last through a dozen laps, let alone an entire race.

Silly thing is, it probably does... at least, the finder utility DOES work with Win2K... but the INSTALLER won't let you install it without XP. How dumb is that?

Well so far with these Areconts, I've found a lot to complain ABOUT. Or maybe I've just been spoiled, because the IQs we've been installing have been solid and easy to use and don't clutter things up with additional software. Seriously, Arecont has NOT made a good first impression around here, and I have a feeling we won't be using any more of them anytime soon.

I don't use the bead, or the foam... just makes the dome more cramped. I also use a right angle connector for the network port to avoid sharp bends in the cable within the dome. Right-angle crimp-on RJ45s?? That would be handy! Never seen that! Hey, here's another one for the rant list: the position of the network jack on these 3155s makes it unnecessarily difficult for anyone with even marginally thick fingers to unplug the cable, the gap between the locking tab and the housing is so narrow. How hard would it have been to rotate the thing 90 degrees?? I'd hate to try it with a patch cable with the rubber shroud over the tab. This whole product line looks like a classic case of engineers slapping things together with no clue as to actually USING the product.

When I was at training Arecont admitted to these problems and said they have since been rectified. Which ones? The bad domes? The "missing" yaw adjustments? The mis-installed screws? The more I read, the more concerned I get about their QA/QC - what other problems can I expect to run across with these things? They're already installing software I don't want, without even asking permission... wonder what's next?? My coworker has a few more rants about these things after today... not the least of which is the use of stainless-steel bolts to mount them to their pendants - when you're supporting a camera 40' off the ground and trying to balance the mounting screws on your driver, it would be really nice if they would actually hold to a magnet. All manufacturers who use stainless or other non-magnetic screws for this sort of INTERNAL location should be lined up and shot. BTW, I'm interested in others' experiences with those ferrite beads as well - we've been putting them on, JUST IN CASE, but man, what a PITA to cram them inside the housings! Never needed them on any other IP cams, so I'm wondering what the deal is with these?

It's not hard to miss - with this arrangement, the camera itself flops around pretty loosely in its mount. You can't adjust the "tilt" movement at all without it being painfully obvious something is amiss.

Well, back (somewhat) to the original bent of the thread... I've just finished bench-testing the last of the seven AV3155DNs, and found a little QC issue... This is how the mounting ring for the "yaw" control is supposed to be attached, using four screws around the upper ring, looking something like this: And this is how three of the four screws look on this camera, out of the box: It appears the three screws were inserted BEFORE the top ring was in place... an easy fix - take out the one screw holding it all together, remove the ring, remove the other screws, put everything back in the correct order - but still troubling. The holes were NOT stripped, mind you, or broken; the screws were NOT pulled through them... they screws were simply installed before the ring was in place. Normally this would be cause to send the camera back to the factory, but it's supposed to be installed tomorrow on a job that's already been delayed three weeks because the cameras took so long to get shipped... and another delay would probably give the contractor a heart attack! (On a side note, these snaps were taken with my phone's camera under a 20W halogen desk lamp... pretty damn impressed with the picture!)

Wow, the spam level is ramping up around here...

^Cut'n'paste fail.

Not "usually". Sometimes, only. A larger sensor can also mean the same number of larger pixels, which equates to better low-light performance.

I think in that instance, you might be better to just use cameras with their own built-in recording - small camcorders, for example... depending on where you want to mount them, of course. If you're looking for things like the underbody cams you see on NASCAR and F1 racers, showing sponsor labels on the struts and whatnot, you need to remove the word "cheap" from your requirements before you go any farther.

Seriously Beyond Hope in Spuzzum. LOL! heheheh, we used to say, "If you ain't been to Spuzzum, you ain't been nowhere!"