Hello from down under Australia

natharas

n3wb
Jun 10, 2022
14
5
Australia
I've only just found this site and can't believe I hadn't found it earlier.

I'm looking to put some cameras out the front of my house, only need 3 cameras, been looking at HIKVision and Dahua, as I'm new to the whole security cameras and what type to get here I am, the plan is as follows:

Cam 1 to cover the front driveway / yard, Cam 2 will be to cover the front door and the third camera will be down the side of the house above the gate.
NVR, I'm looking to spin something up on my server which is Proxmox with TrueNAS on it then Blue Iris on a VM.

Looking to connect all the cameras via POE, with that comes the requirement for a POE Switch, looking at a 16 port POE Switch and open to recommendations, I will also be putting the cameras behind a VLAN with no access to the internet.

With regards to the cameras it has been recommended to go varifocal and turrets, would this one suffice for down the side of the house, I'd then be looking at something similar to cover the front door (5 or 6MP) and from the garage covering the driveway and the front yard - https://www.securitywholesalers.com.au/product/dahua-dh-ipc-hdw2431emp-as-0280b-s2-4mp-starlight-ip-turret/

I'm after any advice as it is all fairly new to me and can be quite overwhelming with so many different cameras on the market.

Thank you in advance!
 
  • Like
Reactions: gwminor48
:welcome:

The first thing is don't chase megapixels. In the chart below note that there are no three or five megapixel cameras listed. That is because there is no sensor that performs well under low light conditions in a three or five megapixel camera.

Start out by looking in the WiKi in the blue bar at the top of the page. There's a ton of very useful information in there and it needs to be viewed on a computer, not a phone or tablet. The Cliff Notes will be of particular interest although the camera models listed there are a generation old at this point. The best way to determine what kind of camera you need in each location and where each location should really be is to buy one varifocal camera first and set up a test stand for it that can be easily moved around. Test using that, viewing using the web interface of the camera, during the day and at night. Have someone walk around behaving like a miscreant and see if you can identify them. There is also information for choosing hardware and securing the system along with a whole bunch of other good stuff.

Don't chase megapixels unless you have a really BIG budget. Chase sensor size and bigger is better. To confuse you more sensor sizes are listed in fractions so do the basic math to be sure, 1/2.7 is bigger than 1/2.8 or 1/3. General rule of thumb is that a 4MP camera will easily outperform an 8MP camera when they both have the same sensor size. Reason being that there are twice as many pixels in the 8MP versus the 4MP. This results in only half the available light getting to each pixel in an 8MP that a pixel in the 4MP "sees".

A dedicated PC doesn't need to be either expensive to purchase or to run. A used business class machine can be had from eBay and various other sources. The advances made in Blue Iris make it easily possible to run a fairly large system on relatively inexpensive hardware which also makes power consumption low, as in under 50 watts in many cases. The biggest expenses turn out to be hard drives for storing video and a PoE switch to power the cameras and, of course, the cameras themselves.

The three basic rules of video surveillance cameras-

Rule #1 - Cameras multiply like rabbits.
Rule #2 - Cameras are more addictive than drugs.
Rule #3 - You never have enough cameras.

Quick guide -

The smaller the lux number the better the low light performance. 0.002 is better than 0.02
The smaller the "F" of the lens the better the low light performance. F1.4 is better than F1.8
The larger the sensor the better the low light performance. 1/1.8" is better (bigger) than 1/2.7"
The higher the megapixels for the same size sensor the worse the low light performance. A 4MP camera with a 1/1.8" sensor will perform better than a 8MP camera with that same 1/1.8" sensor.

Disclaimer - These sizes are what the manufacturers advertise and may, or may not, be the true size of the sensor in the camera.
720P - 1/3" = .333"
2MP - 1/2.8" = .357" (think a .38 caliber bullet)
4MP - 1/1.8" = .555" (bigger than a .50 caliber bullet or ball)
8MP - 1/1.2" = .833" (bigger than a 20mm chain gun round)

Focal length, lens size, is another critical element in camera selection. A 2.8mm lens will produce a nice wide view but will be useless for identification at distances greater than abut 10 or 15 feet.

Compiled by wittaj

Don't believe all the marketing hype no matter who makes the camera. Don't believe those nice night time captures they all use. Look for videos, with motion, to determine low light performance. Any camera can be made to "see" color at night if the exposure time is long enough, as in half a second or longer. Rule of thumb, the shutter speed needs to be at 1/60 or higher to get night video without blurring.

Read the reviews here, most include both still shots and video.

Avoid Reolink, Foscam, SV3C, Nest, and all the other consumer grade cameras. They all struggle mightily at night and never get anything useful on video. Here's a link to a whole thread debunking Reolink in particular.

Compiled by mat200 -

A collection of various consumer grade failures -

Avoid WiFi cameras, even doorbell cameras. WiFi is not designed for the constant, 24/7, load of video that a surveillance camera produces. At best, with two cameras on WiFi, they will still experience dropouts multiple times daily. Murphy's Law says that will happen at the worst possible moment.

Lens size, focal length, is another critical factor. Many people like the wide, sweeping, views of a 2.8mm lens but be aware that identification is problematic with a lens that wide. Keep in mind that it may take two cameras, or more, to provide the coverage you need or desire. Another factor that effects view angles is the sensor size. Typically larger sensors will have a larger field of view in any given lens size.

The 5442 series of cameras by Dahua is the current "king of the hill". They are 4MP and capable of color with some ambient light at night. The 2231 series is a less expensive alternative in 2MP and does not have audio capabilities, no built in microphone, but is easier on the budget. The 3241T-ZAS has similar spcs as the 2231 and has audio. There are also cameras available from the IPCT Store right here on the forum and from Nelly's Security who has a thread in the vendors section.

Again, courtesy of Wittaj -

The 5442 series of cameras by Dahua is the current "king of the hill". They are 4MP and capable of color with some ambient light at night. The 2231 series is a less expensive alternative in 2MP and does not have audio capabilities, no built in microphone, but is easier on the budget. The 3241T-ZAS has similar spcs as the 2231 and has audio. There are also cameras available from the IPCT Store right here on the forum and from Nelly's Security who has a thread in the vendors section.

Review - 8MP 1/1.2" sensor full color camera


Dual Sensor 4K

5442 Reviews

Review - Loryata (Dahua OEM) IPC-T5442T-ZE varifocal Turret

Review - OEM IPC-B5442E-ZE 4MP AI Varifocal Bullet Camera With Starlight+

Review-OEM 4mp AI Cam IPC-T5442TM-AS Starlight+ Turret

Review IPC-T5442TM-AS-LED (Turret, Full Color, Starlight+)

Review: IPC-HDBW5442R-ASE-NI - Dahua Technology Pro AI Bullet Network Camera

2231 Review
Review-OEM IPC-T2231RP-ZS 2mp Varifocal Turret Starlight Camera

3241T-ZAS Review

Less expensive models -

VPN Information Thread

Dual NIC set up

Stop Win10 Updates
 
Welcome to IP cam Talk. :welcome: Stick around and read up on things. Around here you never know what will be discussed.

Sounds like you are ahead of a lot of the folks that show up here. You have read some and know what cameras to look for and such.

Just be warned, we love to spend your money...... That is we will point you to the good cameras so you can
pay once, cry once, rather then buying cheap cameras and then buy the good cameras to replace the cheap stuff.
 
Start out by looking in the WiKi in the blue bar at the top of the page. There's a ton of very useful information in there and it needs to be viewed on a computer, not a phone or tablet. The Cliff Notes will be of particular interest although the camera models listed there are a generation old at this point. The best way to determine what kind of camera you need in each location and where each location should really be is to buy one varifocal camera first and set up a test stand for it that can be easily moved around. Test using that, viewing using the web interface of the camera, during the day and at night. Have someone walk around behaving like a miscreant and see if you can identify them. There is also information for choosing hardware and securing the system along with a whole bunch of other good stuff.

Don't chase megapixels unless you have a really BIG budget. Chase sensor size and bigger is better. To confuse you more sensor sizes are listed in fractions so do the basic math to be sure, 1/2.7 is bigger than 1/2.8 or 1/3. General rule of thumb is that a 4MP camera will easily outperform an 8MP camera when they both have the same sensor size. Reason being that there are twice as many pixels in the 8MP versus the 4MP. This results in only half the available light getting to each pixel in an 8MP that a pixel in the 4MP "sees".

A dedicated PC doesn't need to be either expensive to purchase or to run. A used business class machine can be had from eBay and various other sources. The advances made in Blue Iris make it easily possible to run a fairly large system on relatively inexpensive hardware which also makes power consumption low, as in under 50 watts in many cases. The biggest expenses turn out to be hard drives for storing video and a PoE switch to power the cameras and, of course, the cameras themselves.

Thanks for the detailed response it is most definitely appreciated. I'm very confident that I've worked out I will either be going 4MP or 6MP, as you explained with 8MP I don't think I will need it for my home.

So, with regards to Blue Iris, I will be running that on my home server that I already have and it will be a VM on TrueNAS, it already has 96gb of RAM and soon to be 12TB of storage. I do though need to upgrade my switch and while I'm doing that I will be getting a managed POE switch so that is something that I do need to look into further.

The three basic rules of video surveillance cameras-

Rule #1 - Cameras multiply like rabbits.
Rule #2 - Cameras are more addictive than drugs.
Rule #3 - You never have enough cameras.

Yeah, in my case it will be a hard limit of 3 cameras, I don't want cameras in my backyard and where I live the chance of anyone entering from behind our property is almost 0%. I also don't want a camera over our pool obviously for privacy reasons.

Quick guide -

The smaller the lux number the better the low light performance. 0.002 is better than 0.02
The smaller the "F" of the lens the better the low light performance. F1.4 is better than F1.8
The larger the sensor the better the low light performance. 1/1.8" is better (bigger) than 1/2.7"
The higher the megapixels for the same size sensor the worse the low light performance. A 4MP camera with a 1/1.8" sensor will perform better than a 8MP camera with that same 1/1.8" sensor.

Disclaimer - These sizes are what the manufacturers advertise and may, or may not, be the true size of the sensor in the camera.
720P - 1/3" = .333"
2MP - 1/2.8" = .357" (think a .38 caliber bullet)
4MP - 1/1.8" = .555" (bigger than a .50 caliber bullet or ball)
8MP - 1/1.2" = .833" (bigger than a 20mm chain gun round)

Focal length, lens size, is another critical element in camera selection. A 2.8mm lens will produce a nice wide view but will be useless for identification at distances greater than abut 10 or 15 feet.

Thank you, I will definitely be using this as a reference from now on.

The 5442 series of cameras by Dahua is the current "king of the hill". They are 4MP and capable of color with some ambient light at night.

Is there a lower version of the 5442 series that would suffice, I'm just having a quick look at those and in AUS they still appear to be very expensive at $400+
 
  • Like
Reactions: gwminor48
Contact Andy at EmpireTech, a forum member here. He ships "down under" and has excellent pricing and support. You can check out his prices on Amazon and AliExpress as well but buying directly from him works quite well and he ships very quickly. He's in Hong Kong and works with Dahua to get firmware updated when members here find problems as well. That firmware usually never even makes it to the Dahua site and is always an improvement when it's released.

Andy
IPCT Thread

Andy's Store

King Security/EmpireTech Store

Email
Andy Wang kingsecurity2014@163.com
 
As an Amazon Associate IPCamTalk earns from qualifying purchases.
Cam 1 to cover the front driveway / yard, Cam 2 will be to cover the front door and the third camera will be down the side of the house above the gate.
That is a good start but is not really a cam plan, and you do need a cam plan. Stating something like 'to cover' is meaningless.

How big an area are you talking about?

What is the purpose of having a cam in each of those locations?

Do you just want an overview to see if someone is there? Or do you really want to be able to ID that someone when they do something bad there? It is imperative to have a true plan or you will not get what you want from the cams.

It is suggested that you get one varifocal cam to use as a test cam. Put it on a test rig as described in the WIKI and place that cam where you think you want a cam. Walk it day and night, recording that walk. Is that the view you want? Can you ID the face? Try it out in different spots to get a feeling of what can be expected from cams. Once you have an informed plan, which should change as you gain experience, only then should you buy additional cams to complete your plan. Test BEFORE you run wires and mount cams.

This is my test rig:

Test Rig.JPG

I have two of those 2431 cams in 2.8mm in my garage. See the thread below on how well they perform.

 
Contact Andy at EmpireTech, a forum member here. He ships "down under" and has excellent pricing and support. You can check out his prices on Amazon and AliExpress as well but buying directly from him works quite well and he ships very quickly. He's in Hong Kong and works with Dahua to get firmware updated when members here find problems as well. That firmware usually never even makes it to the Dahua site and is always an improvement when it's released.
As an Aussie member, I can vouch for Andy. I've purchased 11 cameras from him in two batches about 14 months apart. No problems whatsoever. All but two of my 13 cameras are the 5442 models and the night-time quality is far superior than my 8mp cameras. In my case the 5442 models all have a larger sensor than my 8mp cameras.
 
Hello, welcome to the forum.
 
Do you just want an overview to see if someone is there? Or do you really want to be able to ID that someone when they do something bad there? It is imperative to have a true plan or you will not get what you want from the cams.

It is suggested that you get one varifocal cam to use as a test cam. Put it on a test rig as described in the WIKI and place that cam where you think you want a cam. Walk it day and night, recording that walk. Is that the view you want? Can you ID the face? Try it out in different spots to get a feeling of what can be expected from cams. Once you have an informed plan, which should change as you gain experience, only then should you buy additional cams to complete your plan. Test BEFORE you run wires and mount cams.

Sorry for the long delay, life has been extremely busy at the moment. So to clarify the cameras are to monitor but be able to ID a person if needed. I live in a relatively quiet street and there hasn't been many incidents so to speak of.

I'm aiming for a varifocal cam but unfortunately can't get one as a test cam due to time constraints, this also means that I won't be able to test before running the wires and mounting the cams which I'm not overly concerned about if I can adjust them once I have it all setup with Blue Iris.

All but two of my 13 cameras are the 5442 models and the night-time quality is far superior than my 8mp cameras. In my case the 5442 models all have a larger sensor than my 8mp cameras.

Thanks for the feedback, I assume you're referring to the ones that are these 4MP Starlight Ultra Low Light Motorized Varifocal Turret IP Camera IPC-T5442T-ZE - IP Cam Talk Store or did you go via Amazon? How have you found them over time have you had any issues?
 
I'm aiming for a varifocal cam but unfortunately can't get one as a test cam due to time constraints, this also means that I won't be able to test before running the wires and mounting the cams which I'm not overly concerned about if I can adjust them once I have it all setup with Blue Iris.
"adjusting them" after installing them really can't happen as they are already installed in a specific position and height/view. We are not talking about adjusting the settings, but adjusting/testing the FOV.

Good luck and I hope it works out well for you.
 
  • Like
Reactions: sebastiantombs
Well I've ordered the cameras from Andy and they should be here early next week. I'm now trying to work out POE switches, current thinking is I'll go 2 of these the first will be for the cameras and the second for an Aruba AP TP-Link POE 8 port switch, there is a 16 port option but it has two fans and I'm concerned about the noise level as it will be next to my work desk TP-Link 18 port POE
 
Well I've ordered the cameras from Andy and they should be here early next week. I'm now trying to work out POE switches, current thinking is I'll go 2 of these the first will be for the cameras and the second for an Aruba AP TP-Link POE 8 port switch, there is a 16 port option but it has two fans and I'm concerned about the noise level as it will be next to my work desk TP-Link 18 port POE
I have a 16 port TP-Link TL-SG1016PE V3 switch and the only time I can barely hear the fans running is when I literally have my head by the exhaust vents if I am behind my desk working on wiring. The fans on both my desktop PC's and my Dell laptop are much louder than the Switch fans are on the TP-Link. Only my Mac Mini is quieter. The fan noise it does produce is not a deal breaker.
 
I have a 16 port TP-Link TL-SG1016PE V3 switch and the only time I can barely hear the fans running is when I literally have my head by the exhaust vents if I am behind my desk working on wiring. The fans on both my desktop PC's and my Dell laptop are much louder than the Switch fans are on the TP-Link. Only my Mac Mini is quieter. The fan noise it does produce is not a deal breaker.

Interesting, from the stores that I can purchase from there are a lot that mention fans running louder than usual, though they don't say which version it is and those reviews are a few years old at least so maybe it was the older revision ...