PeteCress

Could it be salt water spray?

Now that you have mentioned it, that sounds like a good candidate - especially since the problem cams are facing into the prevailing SW wind and we have had some 20+ MPH days recently - while the "good" cam is facing more NW.

HikVision DS-2CD2032-I's screwed to a 2x4.

I'm thinking a hinge and a padlocked hasp on the 2x4 mounting pole plus some way to control the descent or, at least, keep it from hitting the deck if/when an inexperienced person takes it upon themself to clean the cams and loses control.

3 HikVision cams - all came out of the box and were installed at the same time.

The two problem cams are on a pole and the third is under the eve of the building next to the pole.

The two on the pole have developed some sort of haze: http://tinyurl.com/pn6lxgn That screen snap was taken late in the day - as in 2030.

Cam site is ExtremeSurfCam.DynDNS.org.

First I thought "Condensation", but why those two and not the third one? And why no droplets?

I guess one lesson is to make the pole installation amenable to easy takedown/cleaning/whatever.

But I can't get down there for at least a week and meanwhile would like to hear some possibilities.

You think that's expensive..

We have 2 of these.

http://www.idealindustries.com/prodDetail.do?prodId=LanTEK2&div=2&l1=certifiers&l2=lantek2

Holy Schimoley!!!!.... that boggles the mind.

OTOH, reading further on the subject I came away with the impression that the $1,200 Fluke device wasn't all that comprehensive and that the $5,000+ devices were more like it...

Oh well.... -)

I installed a EZ-RJ45 Cat 5/5e connector on one end

Part number 202003j http://www.markertek.com/Connectors-Adapters/Data-Connectors-Adapters/RJ45-CAT5-and-CAT6-Connectors/Platinum-Tools/202003J.xhtml

They certainly look like the answer convenience-wise.

They have another one that sounds like it works for both Cat5e and Cat6: http://tinyurl.com/l4ymxx2.

OTOH, reading http://www.audioholics.com/audio-video-cables/bjc-cat-network-cable-quality-interview I come away thinking that the pass-through thing may not be all that wonderful efficiency-wise.

In that context I'm thinking that, instead of the basic "Verification" tester that I have maybe I should have what Fluke calls a "Qualification" tester as in http://tinyurl.com/l78txuj.... but without the $1,000+ price tag -) .... but according the the interview in the link above, that's not going to happen anytime soon.

But I have to wonder about PC/Software-based solutions for cable testing. Seems like I should be able to hook a laptop up to a cable and at least determine raw speed given the right application.

I think the term-of-art is "Layer 1 (Physical) Layer". I see utilities that run on two PC's - one PC at each end of the wire, but nothing where some sort of hardware gadget is connected to one end of the wire and a single PC is connected to the other end.

Edit 2012 06-23 14:08:

Just downloaded the freebie version of TotuSoft's "LAN Speed Test (Lite)" and have concluded that I've been obsessing about nothing.

While the product is not doing "Level 1", it is blowing a file out to the other end of the wire, reading it back, and computing Mbps each way. Seems to me like that is more conservative than Level 1 because it takes into account all the bottlenecks along the way like switches, patch cords, and so-forth. Maybe not as useful in nailing a specific problem, but it it says I've got enough bandwidth for a given task, I probably really do have enough.

It says that my amateurish Cat5e wire to the garden shed is supporting 500-700Mbps. Yeah, test-after-test the numbers move around... but with repeated tests, I'm coming up with around 500 Mbps. How far off can it be? 4 IP cams on that wire.... 2Mbps per cam.... I need to get on with my life.... -)

What problems are you having with the connectors with load bars?

Maybe this will help:

 

The vid straightened me out.

I never thought of clipping the wires on an angle to facilitate threading into the load bar.

Takes a little longer than Cat5 but, OTOH, it would seem less error-prone since once the wires are properly ordered in the load bar, there won't be any crossovers at inserting time.

Thanks!

What's all the hype with these hikvision cameras?

Are they better than dahua?

I have a couple of Dahua's (rebranded as "Q-SEE") and four HikVision DS-2CD2032-I's.

Can't cite a model for Dahuas bc there's none to be found, but the "Device Name" defaults to "TZC2FU00802037" and the BlueIris driver used is "Dahua H.264 RTSP" if that's any clue.

There's no comparison in performance/$ and I find the Dahua's quite troublesome to log in to.

I mistakenly bought a 1,000' spool of Cat6 solid - was supposed to be Cat5a.

But now I've got the stuff....

Been trying to terminate with male connectors with removable plastic inserts/load bars (as in http://tinyurl.com/ply9jqd) and it's been pretty grim - so bad that I've wimped out and teminated with female connectors, doing the final join with a short male-male Cat5e.

Seems like there are two alternatives:

• Connectors with built-in, non-removable guides
  .
  
• Something called "Tooless" connectors as per http://www.firefold.com/rj45-connector-cat6-toolless

Does anybody find one significantly easier than the other?

Cost isn't a big deal because I don't need a lot of connectors - just enough to go through this spool.

Something like this?

 

What would be a target bandwidth?

That looks like what I had in mind presentation-wise if the rectangle can be moved by the viewer.

Target Bandwidth? As small as possible while retaining functionality.

Right now one 1280x720 cam seems to be taking up about 500kB/s.

"Seems to be" because I'm kind of clueless on this stuff and 500kB/s is what my cam server is showing for one camera.

What I was thinking was three cameras stitched and taking up something less than that - the more less the better.

Do you have a link to a live cam? Looks like it's one of these guys: http://tinyurl.com/p7cegy7, but I don't see the rectangle on the demo vids that I watched - so maybe there's some server software between the user and the cam?

The idea of stitching images together/dynamically zooming on them seems interesting, but the multiple-user low-bandwidth issue would not be solved, nor much bandwidth will be saved - due to compression algorithms.

I wasn't thinking so much compression as just shipping fewer bits. Something fairly fuzzy... but with that little rectangle of high def.

Hard...maybe brutal on the CPU... But CPU cycles keep getting cheaper - likewise cams - while bandwidh/streaming site charges seem to be holding steady.

After fooling around with multiple IP cams for a couple of years (http://ExtremeSurfCam.DynDNS.org) on limited bandwidth, it occurs to me that something along the lines of a Gorgon Stare-type system with a few extras is the logical solution to wide area coverage for multiple concurrent users over limited bandwidth.

Streams from multiple el-cheapo cameras merged into a single large image: lots of processing power required by the server, but minimal bandwidth required per concurrent user..

Has anybody heard of such a thing - either in production or under development?

Essential Features:

• Ability to stitch the streams of multiple IP cameras into a single large view. e.g. three 1920x1080 streams => one 5750x1080 stream.
  .
  
• Ability to record that combined stream to disk
  .
  
• Ability to allow a user to log into the DVR and watch the stream over a WAN connection with 3 special features

The 3 Special Features:

1. The stream is presented as a mostly low-rez picture to conserve bandwidth.
   .
   
2. There would be a rectangle overlaying the low rez image and the rectange could be slid around on the image by the user and digitally zoomed by the user.
   .
   
3. Within the rectangle, resolution would be at some much higher level - but the rectangle would be small enough that it would not use too much bandwidth.

This makes so much sense to me that I'm thinking somebody somewhere is either marketing or developing it.

??

Well you are not going to get the functionality of having recorded video clips and the weather overlay without BlueIris. But here's the 411, re-streaming software is not cheap....

Which simplifies my decision greatly...... -).... The client has a low budget for this. He wouldn't even go for a commercial Comcast account; so I had to stream over a couple of radios to a house a mile-and-a-half away to get behind a Comcast cable modem instead of the shop's pitiful 44k DSL connection.

If you want to allow more and more people to visit your site and keep up with the bandwidth, your best bet is to go with a streaming service and the cheapest is YouTube Live, meaning free. The reason is you only have to send one live stream to YouTube, everyone else connects to it, unlimited, 1 user or 100, same bandwidth from your point of view. Then embed the YouTube videos on a web page so it looks like what you have now. Instead of the embedded weather in the video, put the weather somewhere on the page like most sites do.

That was calling out to me a year or so ago. Seems like the logical solution. Tried YouTube and Vimeo at the time, but came up against the problem of automating the upload/delete process (i.e. keep uploading the latest clips and delete the oldest clip at the end of each upload).

Sounds like I should re-visit this in hopes that something has changed - or my IQ has risen - since the last time.

Likely streaming to a Wowsa server or Unreal Media Server. What are you using for live streaming? I'm use BlueIris also for live streaming cameras for a community, works well too. I use the jpegpull.htm that I modified to make it look pretty, easy to do. Get rid of the background wallpaper that's lame, the unnecessary controls and works with any browse.

I use BI too, but I think the limited bandwidth of the Comcast connection that the BI server is behind is most of the difference.

ExtremeSurfCam.DynDNS.org

It's "Good Enough" with just one or 2 people logged in. One cam only puts out 14 FPS - but I think that's just part and parcel of having a cheap cam. The other cam stays in the high twenties.

But with six people logged in, motion gets unacceptably choppy and I am not looking forward to the day when we have a dozen concurrent users.

Seems like, with Flash streaming, there are two advantages:

• The inherently more-efficient Flash file format: more data over a smaller pipeline
  .
  
• The fact that the streaming service's bandwidth is much less limited than a Comcast cable modem.

But I'm still trying to get my head around the whole Flash streaming concept: what the components are, what the terms of art are, pricing, and so-forth.

Also, all the Flash streaming implementations I have seen so far present a single "Live" view (even so it might be a couple minutes old...). The users of my application rely a lot on the stack of five-minute clips that my instance of Blue Iris keeps and makes available in it's web presentation.

http://www.schooners.com/multimedia/hillbarcam.htm

What's got me all spun up is the quality of the motion - fan blades and all.

Makes my dinky little BlueIris setup look pathetic.

He has a couple other cams with the same outstanding quality.

Seems like FlashPlayer is involved because of the progress bar at the bottom of the screen.

But I have no clue otherwise.

Don't feel like The Lone Ranger.

 

Been trying to get my IP Cam Viewer to recognize my just-unboxed DS-2CD2032-I for a couple of hours now.

 

I do not see my model number in the "Model" dropdown on the setup page.

 

Tried all three that are there (DS-2D815F-E, DS-7216HFI-SH/A, and DS-81xxDVR), but to no avail.

 

I wonder if it's just a missing driver issue?

Bump!...

FWIW, the settings that currently work for me:

Make: Hikvision

Model: Hikvision DS-2CD8153F-E

I tried this in the BI forum last week, but no nibbles.

One of my BI cam servers is headed for bandwidth issues as

more than about 4 concurrent users seem to slow the FPS

down quite a bit. Can't imagine a dozen....

So I am thinking about a single "User" which is actually

a stream for each camera to a streaming site somewhere that

is not so bandwidth-bound.

Problem is that I don't have a clue.

Here is what I *think* I understand.

Can somebody validate or correct this:

• IP cam connects to something called "Flash Media Server Software" running on a local PC.
  .
  
• The "Flash Media Server Software" on the local PC is connected over the WAN to somebody's web site.
  .
• That web site is running something called "An Embedded Player".
  .
• On the web site, the web page that presents the camera's video contains HTML that points to the site's "Embedded Player" and tells the user's browser that we are going to feed a Flash Player stream to it.
  .
• The user's browser either throws a "Flash Player required to view this" or seamlessly renders the video.

Am I even close?

Resource Monitor

I'm trying both Resource Monitor and DuMeter, but neither one seems to offer up actual bandwidth use.

RM because it just gives total Network I/O, and DuMeter because it's upload use (1.5 Mbps as I write this) seems at odds with what SpeedTest.net says is my capacity (only .7 Mbps).

OTOH, maybe I don't know how to read the screens.... or SpeedTest.net's numbers are flawed.

I actually never seen the Hikvision weatherproof connector in action...got any pictures after you installed it?

I have not installed them yet - maybe in a week or two when I get to the actual site for the final setup.

Right now I'm semi-obsessing about what to do with the power pigtail that is not used if the cam is running on POE.

Thinking that for the sake of neatness maybe just clip it off, make sure one wire isn't touching the other, seal it with some silicon goop and maybe wrap with mastic for good measure.

Got a couple of cams connected to a Blue Iris server (ExtremeSurfCam.DynDNS.org), toying with the idea of adding 3 more.

No streaming service: users just connect direct to the server.

Of course I'm running into FPS limitations beyond a certain number of concurrent users/cameras.

What I would like to find is a tool that tells me how much bandwidth my browser window is using.

With that, I could fire up IE or Chrome or FireFox or Safari, aim it at ExtremeSurfCam.DynDNS.org, and see how much bandwidth is used by a single user connection.

Then I could divide that number into the bandwidth available to the Blue Iris server over it's IP connection and get an idea of how many concurrent users can be supported.

From there, I could tune Blue Iris and add/remove/resize cameras to optimize optimze the bandwith for number of users vs functionality.

Does anybody know of such a tool?

viz: http://tinyurl.com/mzbh36l

Has anybody figured out what this and how it is supposed to work?

The Hik has a female Cat5 connector, a "Y", and a dongle containing power receptacle - all of which would benefit from being protected from the weather.... and I'm guessing Hik's thingie has a role... But I can't figure out what it is.

??

Edit 2014 04-24 15:53:

OK... Dumb post.

It's intended to be used with a Cat5 wire whose connector is attached after being fed through the thingie... and then the thingie screws on to the cam's female Cat5 receptacle....

Leaving the power receptacle (not needed with POE) to be dealt with by the user.

Getting a seamless composite is not really possible. .... but this isn't about surveillance, it's a cool webcam.

Here's what I am getting now: http://tinyurl.com/mm2ydqc in a sort of "Bread Board" setup looking out my driveway.

Yeah, perspective is plenty goofy... but the panorama is at least 180 degrees - maybe a little more - and it looks to me like it is going to be good enough.

When I take the cam array down to the bay on a windy day with plenty of people on the water, we'll find out if the images are actually good for anything.

These are 4mm cams running at 1920x1080 sitting side-by-side as in http://tinyurl.com/l4zbzz8

Hopefully this helps you visualize it. The angles I drew on here are not super precise, but they will be awfully close.

 

You would be able to cover roughly 100 feet across, 30 feet away from the cameras. That is of course with the cameras spaced 20 feet apart. At 10 foot spacing you would have a total coverage of roughly 80 feet

 

[attachment=0]three-cams.jpg[/attachment]

What happens when the camera on each side of the center camera is rotated so that the cones just touch or overlap just a bit?

I guess I'm going to find out myself in a couple weeks.

Something with perspective is my guess.

I would also guess that I need to determine what distance I want the three frames to touch each other at. With the twenty-foot spacing shown, it looks like those frames touch each other at about 12' from the cams.

Sounds like some backyard experimentation is in order before I go down to the actual site.

I was just fishing for some magic bullet with this post and it sounds like there is no such thing.

Thanks for all the effort - and I'll report back with a URL and description of placements/angles once I get something working.

What type (brand and model) of camera did you purchase?

HikVision DS-2CD2032-I as in http://www.aliexpress.com/snapshot/6063556910.html

4mm lens.

2048×1536 max rez although experience suggests I may wind up running them at 1920x1080.

I tried this with another thread ("2 or 3-Camera Array: Anybody Tried It?") but no nibbles.

Just pulled the trigger on 3 2048×1536 cams with 4mm lenses.

It is a bay-side application for people to watch what is happening out on the water. The existing cam's field of view is insufficient and allowing people to use it's PTZ capability has resulted in "PTZ Wars" between users as different people contend for their own idea of an ideal view.

The intent is to mount the three cameras in such a way that the three pictures align into one very wide image with minimal overlap on the cam server's "All Cameras" web page. Then each user can decide which view suites them and select that view for maximum definition/size.

I guess I'll find out for sure once the cameras arrive and I get down there - but knowing in advance might save me a few man-hours of futility.

The Question:

Is there any hope of getting the desired composite view by mounting these things right next to each other - as in all three within a foot or two?

Or will I have to spread them as far apart as possible? I've got about 25' of building to work with in that respect, but scrunching them together in a faux bird house about 10' above the deck has it's attractions in terms of theft avoidance.

??

Right now, I've got http://ExtremeSurfCam.DynDNS.org where the "Extreme Beach" cam has a rather limited view of what's going on vis-a-vis the bay. FWIW "Pleasantville Yacht Club" is not one of my cams - I'm just linking to somebody else's cam by popular demand.

I'm thinking maybe supplement "Extreme Beach" with an array of 2 or 3 more el-cheapo indoor cams mounted in a wooden box with a glass front such as to give a 180-degree panorama of the beach/bay.

Then they could be presented side-by-side in the cam server's web page - like "Bandwidth Test1" through "Bandwidth Test3 are now.

I already had one of the type of cam I would use there for two seasons - during one of which it survived Sandy just in a plastic bag and a certain amount of electrical tape.

With the proposed array, it seems like the box/glass would be pure gravy weather-wise - more of a theft deterrent than anything else. The box would be high up enough so that somebody on the boardwalk would not see what is in it.

I can't see why this would be any different from any 3 other cameras... but figured maybe somebody's been here and could offer up some tips.

Got two EdiMax 3030 cams where it looks as if the same capacitor (?) has failed - as evidenced by leakage. viz: http://tinyurl.com/ohvyjxx About a hundred bucks each.

The Question: Is it worth spending a half day finding/procuring these capacitors and trying to solder them into the board without messing anything else up?

Or does such a failure usually indicate something else is wrong too?