The HookUp ranks the Empiretech T5842 (Dahua IPC-HDW5842T-ZE) as the best affordable camera if night time performance is the main concern

[wittaj]

Hot off the press, The Hookup ranks @EMPIRETECANDY T5842 as the best affordable camera if night time performance is the main concern.

I still think he has flaws in his analysis trying to do to much with a camera and trying to get a further distance IDENTIFY out of it than the focal length is set to. I also think with proper settings he could have improved the daytime.

Reolink at night got a much higher score than it should based on the test of standing still and seeing if you could read a Test sign he was holding. Motion at night is where we need to see these cameras perform and the Reolinks are about as bad as it gets in that performance.

He ranked the Reolink 811A as the higest value camera...

He ranked the Axis P3248 as the best overall performance (but it is also an $1,100 camera).

 

[wittaj]

Crappy cold out, so what better to do than sit back with some popcorn and watch the comments come in on the video comments LOL:

 

[NightLife]

Oh, to be a neophyte once more ..

 

[garycrist]

It is real easy to slow down the bad guys!!

 

[wittaj]

His videos are well produced and pack a lot of information into them, but I think his tests suffer from the same mindset of the general consumer of trying to do too much with a field of view and thinking we can digital zoom like they do in the movies. It would also be interesting to see what the settings were for each camera.

For example, he said that the EmpireTech camera had the daytime image in the bottom tier of the cameras reviewed. One parameter he does mention he changes was WDR. Looking at the video, I think WDR was cranked up too high. I think if he either turned it off or turned it way down, the results would be different.

Another thing we have all seen is the DORI is a good starting point, but no camera at this wide of a zoom is going to perform well at 50 feet, regardless of what DORI says.

Most of us here would not use this camera at this zoom/focal length and field of view to try to IDENTIFY someone at this distance, which he stated was 50 feet.

We would optically zoom in to a much tighter field of view.



Then the next thing he does is takes this field of view above and then he DIGITALLY zooms in to about what most of us would OPTICALLY zoom into.

You may get away with this digital zoom in the day, but certainly not at night. If we wanted to IDENTIFY someone at this distance, we would OPTICALLY zoom in so that our field of view is what he shows here digitally, we would not DIGITAL zoom after the fact.



And then at night time, we see how difficult that is. Again, most of us would not use this field of view to IDENTIFY someone at that distance of 50 feet.



and then when DIGITALLY zoomed, it is an even worse mess. Wrong camera for the distance to IDENTIFY



Finally, I question was he was running default/auto settings or changed them but still too slow of a shutter?

Again at this zoom and distance, we would not be trying to IDENTIFY, but this is too blurry. Had to be default shutter. See my next post as an explanation and example.

 

[wittaj]

Let me demonstrate with my EmpireTech cameras (commonly referred to here as Andy cams by @EMPIRETECANDY) to show you the difference between using default/auto that I think the Hookup was running at (or if not default, then the wrong parameters) versus dialing in the cameras to your field of view and lighting.

So here is a screenshot of the Hookup running forcing the camera in color and some DIGITAL zoom for him running at 25 feet:



Aside from the obvious blurring due to digital zoom, he has too much motion blur and he is almost see thru as a result of too much gain.


My 5442 3.6mm (4MP) overview camera has roughly the same wide field of view as his with the street at around 50 feet, so someone across the street would be in the 80ish foot range.

To try to make a comparable comparison, I wanted to find a runner as a walker wouldn't be a fair comparison. I was able to find an 11pm jogger, but unfortunately they were not at 25 feet, but were on the other side of the street at about 80 feet, so if anything my performance should be worse.

So of course I am not expecting to IDENTIFY at 80 feet with a DIGITAL zoom, but my image looks better at 80 feet with a lot less light than his at 25 feet. There is blur in mine obviously due to the digital zoom, but it is not motion blur or see thru the person because of too much gain; whereas his blur is from both motion and digital zoom along with too much gain:



Now let's compare that to an OPTICAL zoom of the Z12E (2MP) where it is OPTICALLY zoomed in to be able to get an IDENTIFY at that distance:



This my friends is why you do not chase MP and instead purchase a camera on the correct MP/sensor ratio and focal length for the area to cover!

You need to identify the areas you want to cover and pick a camera designed to cover that distance. In some instances, it may be a 2MP or 4MP that is the right camera. DO NOT CHASE MP!!!

You need to get the correct camera for the area trying to be covered. A wide angle 2.8mm to IDENTIFY someone 40 feet away is the wrong camera regardless of how good the camera is. A 2.8mm camera to IDENTIFY someone within 10 feet is a good choice OR it is an overview camera to see something happened but not be able to identify who.

One camera cannot be the be all, see all. Each one is selected for covering a specific area. Most of us here have different brands and types, from fixed cams, to varifocals, to PTZs, each one selected for it's primary purpose and to utilize the strength of that particular camera.

So you will need to identify the distance the camera would be from the activities you want to IDENTIFY on and purchase the correct camera for that distance as an optical zoom.

If you want to see things far away, you need optical zoom, digital zoom only works in the movies and TV (especially at night)...And the optical zoom is done real time - for a varifocal it is a set it and forget it. You cannot go to recorded video and optically zoom in later, at that point it is digital zoom, and the sensors on these cameras are so small which is why digital zoom doesn't work very well after the fact.

 

[dudemaar]

should be called default settings shoot out

 

[wittaj]

Is he a member here?

 

[TonyR]

I think it's safe to say that both "John Perrone" and "Battlecampfan" have never been on IPCT and performed a search with the term "Reolink."

 

[mat200]

Is he a member here?

View attachment 121905

Indeed, Good to see solid comments in the thread on the topic, and good to see The Hookup improve on the discussion and finally bring up DORI concepts, as well as being more open and engaging in a useful discussion. Happy to see this happen and it just not be a 100% Reolink is the best for everyone...


Brian Simmons
I'll also note that 4k cameras in general are not going to perform as well at night as lower resolution cameras (all other things being equal). This is why people who care about low light performance will usually skip 4k cameras and stick with a 4mp camera with a large sensor (like 1/1.8" or larger). If you have a 4k camera (8mp) and a 4mp camera with the same size image sensor, the pixels on the 4k camera are half the size of the pixels on the 4mp camera. There are twice as many of them on the same size image sensor after all. This means the 4k camera will require TWICE as much light to provide the same exposure as the 4mp camera or the 4k camera will require a shutter speed twice as slow as the 4mp camera, or the 4k camera will require twice the gain as the 4pm camera - or a combination of all three variables. Slower shutter speeds and higher gain settings result in worse low light performance. Now I will say that there are starting to be 4k cameras with 1/1.2" sensor sizes that can perform about the same in low light as a 4mp 1/1.8" sensor size (because the 1/1.2" sensor is much larger than the 1/1.8" sensor). But even better would be a 4mp camera with a 1/1.2" sensor size (which I don't believe they are making yet).

The longer discussion in this:




Brian Simmons
In a residential setting, low light/nighttime performance should be very high on the list of "must haves". (I'd argue it is the most important element). Let's face it, while crime can certainly occur at anytime of the day, it happens more often in low light conditions. You joked about a criminal simply having to "move fast" to avoid any usable footage, but criminals aren't stupid and they do understand that low light provides them the advantage - both from being detected by actual humans, but also making it harder to get identifiable footage from CCTV systems should they be captured by cameras.

Brian Simmons
@The Hook Up My comment about movement wasn't that you were right that criminals try to move fast to avoid CCTV coverage (which was a joke you were making), it's that they know that low light gives them the advantage - whether they are moving fast or slow. I don't know about your area, but in my area 100% of door checkers occur at night. You are right that they are not trying to bring attention to themselves which is why they aren't out roaming neighborhoods during the day when they will be seen (and therefore reported) by lots of people. This (not just door checkers, but most criminal activity) is why nighttime performance should be at the top of people's "must haves" when researching a CCTV system.




Håkon Håkonsen
4 hours ago
Criminals and cars rarely stand still….so be advised that still images is mostly "useless". And most crimes happen at night. So….for best low light performance, use a cam with a large (HQ) sensor, low F-factor (HQ) lens. Empire Tech has an great 200$ cam for low light (not included in this test) if that is important. If you can support the cam with some extra light in the area, the result will be much better - and you can save money on the camera.

Also….Not a fan of the built in light on some of the cams. The small LED creates a small hot spot, making correct exposure difficult. Its just a "better than nothing" solutions (sometimes), but its easy to make the picture look worse.

Most chinese cheap cameras reduce the shutter speed to make the (still) image appear (!) ok at night. Maby one can just ask the intruder to stand still for a picture?

Håkon Håkonsen
@The Hook Up I was thinking about the Color4K-X. Its a 4k bullet cam, F1,0 and 1/1,2" CMOS.




Blake Blackshear
The Empire Tech 5442 will be even better at night. The 4mp sensor has larger pixels on the sensor and each needs less light to get the same image. This also improves the quality of moving objects. This is the tradeoff of high megapixel cameras.

The Hook Up
Correct, but it also cuts the DORI distances in half.


Comment:
DORI distance .. technically does not cut it in 1/2 .. but reduces it by 25% or 1/4 if the aspect ratio remains the same .. ( DORI distance is calculated from the H FOV )

2MP = 1/2 the DORI distance of a 8MP with the same aspect ratio.
4MP = 3/4 the DORI distance of the 8MP with the same aspect ratio.

See the DORI section of the cliff notes.

 

[EMPIRETECANDY]

Hookup's testing, i don't think he will dial too much, more will be at default as newbie. For the professional models auto setting make the pic work bit good than small brand, but more settings will make the camera workig much better. For exactly camera's function there is no compare, most of the video just show the pic quality.
From his posting, reolink still work very well in the daytime.

 

[wittaj]

So people watch that video and still want reolinks...

 

[wittaj]

We do have to give the Hookup props though on this answer with someone trying to compare these cameras to DSLR cameras. We get someone here every once in awhile with a DSLR background and they are expecting the same image quality.

 

[Flintstone61]

It's an Insane value though @wittaj

 

[mat200]

We do have to give the Hookup props though on this answer with someone trying to compare these cameras to DSLR cameras. We get someone here every once in awhile with a DSLR background and they are expecting the same image quality.

View attachment 121942

Oh yes! I remember .. we had a lively discussion on DSLR vs security cameras thanks to a Dsmtweaker ..

we have a business system that consisted of all illustra 2mp mini domes connected to an exaqvision server at 1080p.

The owner was not impressed with the resolution, so I replaced 1 camera with a
DS-2CD2185FWD-1 4K dome

The owner is still not happy with the performance as he doesn’t think it looks 4K(like tv show or movies)

The camera overlooks a parking lot, and you can’t even make out a license plate 50-75 feet away.

Are the expectations to high? Or the product quality to low?

see

Expecting to much

we have a business system that consisted of all illustra 2mp mini domes connected to an exaqvision server at 1080p. The owner was not impressed with the resolution, so I replaced 1 camera with a DS-2CD2185FWD-1 4K dome The owner is still not happy with the performance as he doesn’t think it...

ipcamtalk.com

@J Sigmo did some nice analysis on some of the issues brought up ..

Expecting to much

Well now I feel like a sap for giving the guy something like the benefit of the doubt earlier, and even wasting my time trying to explain, to some degree, why a typical DSLR shooting stills should outperform a typical security camera streaming compressed video. I guess it takes a few years of...

ipcamtalk.com

 

[Flintstone61]

It's one test. I don't have florida for a back drop. We got ultra white tundra on the ground 5- 5 1/2 months a year. Every placement has its challenges. Oh then it's dark like 14 hours a day LOL....until march...So if i had $3000 dollars and a 2 x 12 with 10 camera's glued to it facing North, from my garage, I'd have different results. my drive way is twice as long, so i'd really need one of the test camera's as the Overview cam, and then a 5241z12 to get the far away stuff.
One camera shouldn't be expected to cover all the area he's testing. You need to talk about camera's watching camera's. multiple view angles of the same location perhaps. To optimize your chance of getting usefull " footage". I don't have adequate coverage in my Condo's back lot. Ideally it'd be nice to have an over view camera with the 5442 6mm on LPR duty. and it'd be nice to zoom that 2MP Jidetech PTZ looking at the red Honda in so it can read plates, But it isn't as good at motion as the 5241z12e is, and it you miss stuff. So it's best off as an overview cam. Could use another 5241z12 coming from the courtyard looking down the hill zoomed into the lot catching plates and or faces. None of the camera's tested could stand alone out in this lot day or night and give you everything you need. It'll take 4-5 cams back here to nail these fuckers who steal converters. I'm nailing em, But not a decisively as i'd like.

 

[KenAllen15]

One camera shouldn't be expected to cover all the area he's testing. You need to talk about camera's watching camera's. multiple view angles of the same location perhaps. To optimize your chance of getting usefull " footage".

He talks about all that in this video:

I was surprised how well the axis performed in this most recent video, I haven't seen much mention of axis on this forum.

 

[wittaj]

He talks about all that in this video:

I was surprised how well the axis performed in this most recent video, I haven't seen much mention of axis on this forum.

Yeah, he talks about it in that other video, and yet in his newest video he goes against everything mentioned in the previous video!

Axis isn't mentioned much here because did you see the cost LOL. It is a premium brand. Most folks here don't want to spend that kinda cash on a varifocal camera. It would be like going to a Ford Mustang forum and not seeing many mentions of a Ferrari lol.

So even with this video, comments in the video show people don't want to spend $237 for the most affordable cost/benefit ratio EmpireTech camera and are going to purchase the $110 reolinks based on that video. They aren't willing to spend $137 more for an improvement in quality, so they aren't going to spend $990 more for one camera that is even better quality.

The majority of folks on this site are DIYers and we tend to look more at the cost/performance ratio. The sweet spot for most is under $250 for a varifocal, so most are not looking at a $1100 camera.

And based on this video, I still believe (and my post 6 shows it), that one is better off trying not to get too much out of the camera as the video was doing and instead have multiple cameras.

For the same amount of money, I could purchase one Axis camera or 4 EmpireTech cameras, each one with more optical capabilities and end up with a better overall result. My Empiretech camera OPTICALLY zoomed to the street 50 feet away will beat an Axis camera with an optical sweet spot of 15 feet that someone then tries to DIGITALLY zoom out to 50 feet to try to identify something.

 

[Flintstone61]

I think He's been watching the forums, and digesting the feedback for sure. And he's now more squarely placing the Reo's in the middle of the pack as far as winners and losers. I mean who else would come up with an EmpireTech from all the other Dahua Oem's on Amazon. At least he's not Putting Jidetech, or Louwice, or Vyklin, or YoLuke, HIK and Dahua knock offs in the tests, but since they are on the Amazon Site, maybe he could buy up all those Brand names and tell eveybody what they're really getting. That's where people get the screwjob. They read a bunch of reviews, and shop and shop and surf, and pick( based on a budget) some of these Camera's that lack sought after features, or are promised features that are underwhelming. ( Like My Jidetech)
Noob mistakes.
Without the forums existence, I wonder how many of these goofy named cameras I would have had to buy, to figure out which direction I need to move.
I started out knowing 0. Got a job, in Maintenance, and they were like, " and here's the security camera computer". Best of Luck ( you poor bastard). It was a semi-proprietary, EYE-Max PC with Eye Max surveillance software written in the late Neolithic Period just before DOS. With black and white 4:3 ratio 480 by whatever the fuck camera's. and a few vari-fuckled ( screwdriver adjust) domed 960 x whatever the fuck.
So I saw "swann" cams at Menards( BNC coax) and paid like $70 dollars for a 2.8mm bullet cam. Thus began the journey of a Noob into Camera world. Then you learn that there is an industry trend towards IP. Then you realize in order to step out of the Pleistocene age and stop video taping Neanderthals, you need to scrap the system and start from scratch.
Jump ahead 3 years later, and the legacy system has been cut down to 9 cams.( from 12) and i have 18 IP cams in a BI box in running in a Craigslist server rack ( $50) And I make much better camera choices today as a result of the forum, and You tube and of course The Hook up , and Nelly's, and lots of other You tubers.

 

[wittaj]

We finally made the comment section LOL. Apparently this person hasn't seen this thread where we show the limitations of this video test and trying to do too much with one camera field of view...