Squiffy

Which other video players have you tried? The usual recommendation is to try VLC. You should obviously be looking for MPEG-4 H264 CODECs rather than just MPEG-4 but I would assume you know that

Try an antilog potentiometer. You could also try two diodes on the wiper in inverse parallel, a technique used in music keyboards for pitch-bend wheel potentiometers.

Sure enough that was the problem, removing the filter requires removing a couple of boards inside but it's simple enough to do, results are now as expected.

The manual supplied doesn't mention any removable filter, or that it's optional. I'll take one apart tomorrow and investigate further.

I've been playing around with some Lilin PIH-7030 cameras with ExView HAD sensors (old stock but previously unused). Visible light sensitivity seems good, but the IR sensitivity is really poor. A/B comparisons with cheap CMOS cameras and much older CCD cameras is quite striking.

It's possible I suppose that they're from a bad batch or perhaps they're fitted with a IR cut filter on the sensor (presumably in error). Looking at the sensor there's a cyan hue to the glass cover/filter very similar to this:

www.cleaningdigitalcameras.com/resources/n70-filter.jpg

I'll appreciate any comments regarding this.

If you're displaying the video on the computer, you can try pressing "print screen", which, depending on your OS, may copy the contents of the screen to a clipboard buffer. You can then paste this into an image program and print (after cropping and reprocessing if desired) from there.

However, if the video is displayed in 'overlay mode', this may not show the video on your copied desktop image. The best option would be to use a video editor and step to the frame in question and do the aforementioned, as it's less likely to be in overlay mode.

It may not be the same company, but you might enquire at: http://www.nightvisionsystems.com/ (they were NVEC) to see if they can help.

Clearly it's a 12v camera but to ascertain the connections in the absence of more info, remove the cover and see if you can trace the wiring from the 9-way D connector inside. The power connections should be quite obvious.

It would be easier (and probably cheaper) to source a 12V PSU rather than a 24V AC PSU. An AC supply can be regulated (if using switched mode chopping and filtering techniques) but usually it won't be, therefore it will simply be regulated as much as the local mains voltage is.

A regulated supply is best when a fixed DC voltage is stipulated, because it will not (or should not) change with different loads (i.e. the current drawn, if within the limits, will not affect the voltage).

Note that a cheap way to provide a high current 12V DC source for general domestic use is by using an ATX computer PSU and a bit of DIY, but if you're not technically minded it's best to buy a commercial PSU.

http://www.instructables.com/id/Convert-an-ATX-Power-Supply-Into-a-Regular-DC-Powe/

> Wireless does not work well where either end (transmitter or receiver) moves

In my experience the biggest problem is when you get multipath reception, with AM you'd get ghosting but with FM video it results in a nasty unsynchronised mess as you move (through nodes/antinodes).

If diversity reception is used then it's far less problematic.

Presumably this model:

http://store.keytechsecurity.co.uk/Digital_Recorders/Digital_Recorder_KT8004

If all else fails try running the file through pvastrumento:

http://www.doom9.org/index.html?/DigiTV/pvastrumento.htm

Note the program is not just for PVA files. There are a number of options within the demux (demultiplexing) section which may cure the problem.

I've also found that when VLC has failed to render files correctly, another way to fix the problem that works for me is to use the Womble MPEG Video Wizard to write a new file, which I guess is just tidying up errors in the original MPEG file header.

Try http://204.50.56.163

If that fails then check your hosts file (if you have one) isn't redirecting it.

Thanks for the link but I'm aware of the standard pinouts, what would be useful is the voltage levels normally encountered and if pin 4 (with DC output) is normally ground (as opposed to a floating voltage differential if you see what I mean).

The model is a Sanyo VC-2152, probably around 10 years old. I know that this model was often used in laboratories for observing infra red objects and its sensitivity in that respect is excellent, it shows a bright glow for a soldering iron in the dark for instance. Obviously normal daylight results in total saturation as there's no electronic iris option, so it'll be useful to get an auto iris lens working with this.

In case anyone's come across this connector on other Sanyo camera auto iris connections, I thougt I'd upload a photo of it. Obviously it's an old camera that's seen better days, but the IR sensitivity is superb, so I'd like to get it working. Obviously it only supports DC lenses.

Has anyone any advice on determining the connections by measuring the voltage outputs? I've done this but results aren't clear enough to me to be sure of which connection is which.

> most wireless transmitters were not manufactured to handle video-type of wireless transmissions

If what is meant is just one-way transmission of analogue video and audio, then such devices have been around for decades and are perfectly capable of doing so.

The average wireless cameras/transmitters are not really designed for a reliable 500m line-of-sight path but it's easy enough to change the receiver antenna on these to give adequate gain to make it plausible. You'd still need to mount the receive antenna outside if possible with the receiver module mounted directly to the aerial, to reduce coaxial cable losses.

> Regarding the fps discussion, would 30 fps make a difference in trying to get a screen grab of a license plate or face that is moving?

It would help in instances where the target object is moving fast, the higher the frame rate the better the chances of capturing a useable frame in cases of borderline recognition.

> Are the frames in 15 fps the same as the frames in 30 fps, just spaced farther apart?

Yes.

> go straight to 500G PATA drives

Just to clarify PATA=ATA, but SATA is a different type (serial rather than parallel)

I suggested 250GB as that was about the maximum capacity I could find when I bought one recently, but obviously more is better, assuming it's supported by the DVR.

133 refers to 133MHz, which is the PCI bus speed.

IDE drives are still possible to buy but you'll need an ATA type unless stated otherwise in the manual. It should also show the maximum capacity possible but I would think up to 250GB should be fine.

Well I wouldn't expect many to have access to a Snell and Wilcox test card generator unless they're involved with broadcast TV...

But manufacturers and reviewers can put DVRs through such tests. As for not having access to test cards and electronic test card generators, the latter are easy to find online for printing out or hooking up to equipment:

http://home.tiscali.nl/~peterdb/testcard/

http://www.barney-wol.net/testpatterns/testpatterns.html

Luminance frequency sweeps and multiburst would be ideal for checking resolution.

Worth adding to your tool kits, even if you don't know how to use them it makes you look like you know what you're doing

> You cant really measure image quality properly that way with many DVR's as a still image such as a test card is different from a moving real life image.

Well you can measure it but I agree that a static image will not give realistic (real world) results because of the compression employed on a moving image.

But Snell and Wilcox have moving test card images (for example rapidly moving grating and shapes) which could be assessed for quality and compression artifacts.

> There is no way to measure record quality

Of course there is, with bandwidth expressed in MHz. This can be measured with test equipment and objectively discerned by using a frequency grating pattern on a testcard.

Unfortunately it's not a parameter which seems to be mentioned in most DVR spec. sheets, probably because it would look so disappointing.

I'd only suggest using wireless cameras in locations where a video cable simply cannot be run under any circumstances.

There are wireless cameras that use 1.2GHz but that is illegal in most countries (the allocation is usually for licenced radio amateurs). It's also prone to radar interference in some areas. Beware that many transmitters are not PLL so will drift with temperature changes, in other words they're useless.

In the UK there is an allocation for CCTV wireless at around 1.394GHz but most cameras will use 2.4GHz which as stated are prone to serious interference, the worst being from microwave ovens. While only intermittently used they can create havoc.

You might want to check they offer a full no-quibble refund.

You could also check actual images (I'm sure there are plenty on this forum) giving A/B comparisons between live display/recorded video to see if you think it'll be adequate.

No problem with the FPS, only the resolution (i.e. it will not look as sharp as the live displayed video).

If undecided, why not buy it and try it out, then take it back for a refund if not adequate for your purposes?

Or ask for a demonstration in store? Maplin (their stores in my area anyway) usually have a working Swann model on display connected to a 7" LCD but you'll need a bigger monitor size to assess quality properly.

> Maplin Stock Code N80GC

Ah, well that shows it as 50FPS per channel, which is more than necessary.

Given that you don't actually need all of the features it provides, I would say it's a lot of money to pay for something that will probably disappoint you regarding the recording video resolution.

Rather than wading through lots of manuals trying to find specs for DVRs, it would be great to see a table of recent and new models to make it easy to locate models which provide D1 resolution for example.