3.6mm (90 degree horizontal Fov) fixed 5442 (4Mpx) allows You to Identify at 22ft...
3.6mm fixed 5842 (8Mpx) allows You to Identify at 34ft...
You can try to add to that 30%.. but anything more You need 2 cameras - one overview + one zoomed..
Comparing real pro cameras to home grade. More than resolution specs.
- Thread starter fastline12
- Start date
3.6mm (90 degree horizontal Fov) fixed 5442 (4Mpx) allows You to Identify at 22ft...
3.6mm fixed 5842 (8Mpx) allows You to Identify at 34ft...
You can try to add to that 30%.. but anything more You need 2 cameras - one overview + one zoomed..
I can get dealer pricing on anything. Not saying it would be better than Empire, but....
I am currently looking at the Dahua BF2241 ?
this is the same low end line (Lite) with small thermal resolution (256x192) and small/weak video image sensor and very closed aperture - but previous generation od Eureka.. So lower AI (or no AI) comparing to what Andy have..
Here You have PRO thermal line with bigger thermal resolution (400x300 or 640x512), one fixed model even with 1/1.8" image sensor :
Pro Series - Dahua International
Here You have Ultra series:
As I told You good thermal costs a lot..
And In case of higher thermal cams You should hire someone who have a lot of experience with those type cams to help you choose and implement them in Your case / Yours need (with local inspection)...
My 50.6 degree FOV thermal andycam triggers its IVS on things that a side-by-side 5442 misses. Examples are deer at 160', cat at 100', and small animals like tiny birds and mouses at 25'. Just examples as I haven't done any formal testing. At night many animals that the thermal IVS trips on don't exist on either the visual camera or 5442 image. I can look at the timeline replay at the exact time and location and there's simply nothing there on the visual cameras. During the day I can spot the little critters on either visual camera. The thermal resolution of course sucks, as does the visual camera's IVS, which is worse than with my 8 year old cheap chinese market cameras. It could be that the human/vehicle filters are active even though I have them shut off. But with the thermal camera catching critters the visual cameras can't, that poor visual IVS isn't a big problem. In the big picture I guess it's reasonable to consider it a toy, but this toy catches animals none of my other cameras can, and with very few false triggers too.
Please let us know what your dealer pricing is....I think you will be shocked how much more...
I learned a new feature of the thermal cam, because of a fire I have burning about 100' from the camera.
It retries every few minutes and closes up again in less than a second. Hope I'm not breaking it.
It retries every few minutes and closes up again in less than a second. Hope I'm not breaking it.
Yeah, mine did that when my neighbor was cutting their grass and the mower was too hot.
I will have to eval thermal solutions prob a bit later. For the moment I need to get some new gear running and learn from it. To my next question, help me select the right cameras or narrow this down? I've got several projects on my plate but today I am working on my layout. I am currently playing with This jvsg software, which is pretty helpful in just determining my FOVs, PPF, etc. However, I have found no listings for the rebranded Empire cams? Is there possibly any cross reference sheets to know which Dahua cams they are?
In any case, I think the majority of cams would be OK at 4MP, but at least one, because I'd like it to cover a bigger area at 90* view angle, I am thinking of an 8MP for that? I probably want to stick with vari focal cams as I am just not seasons to know precisely what I need, and want the adjustability. If the software is leading me decent, it looks like about a 5.2mm FL on a 1.2" sensor could get down to 5ft, and stay in the recognition zone out to 50ft. This one cam has an opportunity to monitor people and traffic at many distances between 5-50ft so I feel it would obtain some good shots between those distances.
The ones pointing right at our gates/card readers are places where people would have to stop, would have to get out and engage our readers so the opportunity is good, and I feel 4MP cams will be fine for that. But in all cases, I am mostly seeing places where I need 90* of coverage. I am not sure if that is typically covered with a single cam? I have strategically placed cams at specific choke points, knowing bad guys are likely to approach door openings, with various distances from the cam to hopefully get great coverage.
I am targeting a mount height of 10ft give or take. For nearly all cams, I would have to attach to either a pole, or the side of a steel building so my thought is bullets? I personally like domes better as they seem harder to destroy but I guess part of my strategy is making cams visible and obvious. I know there are two rules of thought on that. With domes, I also like that at a glance, you cannot tell where they are aimed. But seems they need bottom mounted.
I am also trying to understand the models from Empire vs Dahua. Empires are maybe not the latest revisions? Is that software stuff? Optics? Sensor?
I want to stick with 1/1.8 on 4MP or 1/1.2 if we do a couple 8MP. But to keep it simple, It appears most of our shots would be between 20-100ft. Just saying that I may not be able to utilize extreme focal lengths at the moment. I think we need to get cams up and running, then evaluate the next steps.
In any case, I think the majority of cams would be OK at 4MP, but at least one, because I'd like it to cover a bigger area at 90* view angle, I am thinking of an 8MP for that? I probably want to stick with vari focal cams as I am just not seasons to know precisely what I need, and want the adjustability. If the software is leading me decent, it looks like about a 5.2mm FL on a 1.2" sensor could get down to 5ft, and stay in the recognition zone out to 50ft. This one cam has an opportunity to monitor people and traffic at many distances between 5-50ft so I feel it would obtain some good shots between those distances.
The ones pointing right at our gates/card readers are places where people would have to stop, would have to get out and engage our readers so the opportunity is good, and I feel 4MP cams will be fine for that. But in all cases, I am mostly seeing places where I need 90* of coverage. I am not sure if that is typically covered with a single cam? I have strategically placed cams at specific choke points, knowing bad guys are likely to approach door openings, with various distances from the cam to hopefully get great coverage.
I am targeting a mount height of 10ft give or take. For nearly all cams, I would have to attach to either a pole, or the side of a steel building so my thought is bullets? I personally like domes better as they seem harder to destroy but I guess part of my strategy is making cams visible and obvious. I know there are two rules of thought on that. With domes, I also like that at a glance, you cannot tell where they are aimed. But seems they need bottom mounted.
I am also trying to understand the models from Empire vs Dahua. Empires are maybe not the latest revisions? Is that software stuff? Optics? Sensor?
I want to stick with 1/1.8 on 4MP or 1/1.2 if we do a couple 8MP. But to keep it simple, It appears most of our shots would be between 20-100ft. Just saying that I may not be able to utilize extreme focal lengths at the moment. I think we need to get cams up and running, then evaluate the next steps.
So on the 5442, I have seen a few variants. Is this mostly the focal length ranges?
I'm currently looking at a model Color4K-X, 8MP on Empire. If I understand right, the words color/true color/etc just mean they have white lights onboard? Sometimes replacing the IR system, so no IR? At least in one place, the area will be well supported with dusk/dawn flood lighting, but those floods will be facing the camera, but are 200ft away, and 20ft up. I am not sure how a camera will perform with that.
As I understand it, an 8MP can do the full 8, or run reduced resolutions? With the price tag, something like that could be on the menu other than fixed focal. Really need to say variable.
That is correct for most true color cameras.
You do not want to downrez. It still uses the native resolution and downrez 8 to 4MP will usually result in a poorer image than a native 4MP.
Much better to purchase the focal length for the distance you are trying to identify.
You do not want to downrez. It still uses the native resolution and downrez 8 to 4MP will usually result in a poorer image than a native 4MP.
Much better to purchase the focal length for the distance you are trying to identify.
Are you saying a fixed FL would produce a better image than a variable FL?
When you mention "the distance you are trying to identify", that does give me pause as surely there has to be a range? I am trying to learn that detail right now, basically what the focused distances are with a given FL. Again, a very large part of this for me is using the IVS systems to intercept intrusions.
Thsr isn't what I said. I said a native 4MP will typically produce a better image than a 8MP downrezed to 4MP.
But since you asked, many will say a fixed focal length is a slightly better image than a varifocal....depends on your eyes and monitor....but fixed focal lengths are smaller and most will max out at about a 6mm fixed focal length, so if you need 12 or 24 or 32mm, you need a varifocal.
I refer back to this thread where I posted the typical distances most of us are comfortable with based on focal lengths and different cameras.
The Importance of Focal Length over MP in camera selection
But since you asked, many will say a fixed focal length is a slightly better image than a varifocal....depends on your eyes and monitor....but fixed focal lengths are smaller and most will max out at about a 6mm fixed focal length, so if you need 12 or 24 or 32mm, you need a varifocal.
I refer back to this thread where I posted the typical distances most of us are comfortable with based on focal lengths and different cameras.
The Importance of Focal Length over MP in camera selection
I think what gives me pause is realizing that to capture a clean imagine, I am stuck with like this magic 10x10 area. That's it, anymore and all is lost. As in your example in the linked thread, sure enough a guy was captured on video, BUT the difference is you know where to focus the cam because you have a sidewalk with obvious traffic. Criminal behavior is is not always predictable, but rarely 'high level' in the brain activity. They are likely to check doors so those are choke points. But beyond that, like I have mentioned, my hope is that if a cam can pickup reasonable activity at distance, they are likely to walk right into my traps, which are cams placed where you would have to get pretty close to gain access.
But pinching things down to a few feet just to get an ID seems crazy! I'm probably not that concerned that I get cops a perfect picture because they won't do anything anyway. I am more concerned that I can detect an intrusion at a good distance, and I work with the alarm outputs to trigger things to happen that will drive them away. If all these fancy IVS systems are useless unless within the ID range, that is game changing for me.
Now you are getting to the paralysis by analysis stage LOL.
If you want to ID at distance, the field of view gets smaller because you have to optically zoom into it.
Think about binoculars or a telescope. To see something at distance that you have to zoom in and the field of view gets tighter/smaller.
Same concept here.
@Ri22o showed it perfectly in Post #9 of this thread.
Here is an example from @Ri22o of a 3.6mm focal length 180 camera view versus a 2MP varifocal OPTICALLY zoomed in to a pinch point. The person is next to the green line on the left (and would likely trigger for IVS):
And now a 2MP varifocal optically zoomed in to that area in green:
Which one is better for IDENTIFY? There is no way you would have been able to tell it was a male, in a red, flannel shirt, with sunglasses, and a receding hairline in the wide-angle 1st image. But both cameras IVS would trigger, but what can you do with each camera when they trigger.?
Where did we say "these fancy IVS systems are useless unless within the ID range"? We said the closer you get to IDENTIFY distances, the more accurate the IVS will be, but depending on the field of view, lighting, height of camera, the IVS could work within the RECOGNIZE distance, maybe even some OBSERVE distance.
The problem is some try to do too much with one field of view. They wonder why IVS on a 2.8mm camera didn't trigger for a human at the top of the field of view at 200 feet away. That is asking too much from one field of view.
It is another reason why I said in Post #21 that PTZ cameras are a great compliment to an existing static/fixed cam system and using an overview camera to trigger the PTZ to point to an area and start tracking at distances and locations where it doesn't make sense to have IDENTIFY cameras because you have IDENTIFY cameras for the critical capture points that you mentioned you plan to have.
And as I have said, while the thermals won't work at IDENTIFY, if you shouldn't have any movement out in an area, the thermal will have a far better reach at night than any visible camera.
My 2mm thermal is triggering for people at 400 feet. It would be hard pressed for a visual 2.8mm camera to reliably trigger at 400 feet.
As we have said, you will use a 54IR-ZE regardless, so buy one and play with it in your setting and your field of view to learn how the camera works and how far out the IVS will trigger for your setting.
If you want to ID at distance, the field of view gets smaller because you have to optically zoom into it.
Think about binoculars or a telescope. To see something at distance that you have to zoom in and the field of view gets tighter/smaller.
Same concept here.
@Ri22o showed it perfectly in Post #9 of this thread.
Here is an example from @Ri22o of a 3.6mm focal length 180 camera view versus a 2MP varifocal OPTICALLY zoomed in to a pinch point. The person is next to the green line on the left (and would likely trigger for IVS):
And now a 2MP varifocal optically zoomed in to that area in green:
Which one is better for IDENTIFY? There is no way you would have been able to tell it was a male, in a red, flannel shirt, with sunglasses, and a receding hairline in the wide-angle 1st image. But both cameras IVS would trigger, but what can you do with each camera when they trigger.?
Where did we say "these fancy IVS systems are useless unless within the ID range"? We said the closer you get to IDENTIFY distances, the more accurate the IVS will be, but depending on the field of view, lighting, height of camera, the IVS could work within the RECOGNIZE distance, maybe even some OBSERVE distance.
The problem is some try to do too much with one field of view. They wonder why IVS on a 2.8mm camera didn't trigger for a human at the top of the field of view at 200 feet away. That is asking too much from one field of view.
It is another reason why I said in Post #21 that PTZ cameras are a great compliment to an existing static/fixed cam system and using an overview camera to trigger the PTZ to point to an area and start tracking at distances and locations where it doesn't make sense to have IDENTIFY cameras because you have IDENTIFY cameras for the critical capture points that you mentioned you plan to have.
And as I have said, while the thermals won't work at IDENTIFY, if you shouldn't have any movement out in an area, the thermal will have a far better reach at night than any visible camera.
My 2mm thermal is triggering for people at 400 feet. It would be hard pressed for a visual 2.8mm camera to reliably trigger at 400 feet.
As we have said, you will use a 54IR-ZE regardless, so buy one and play with it in your setting and your field of view to learn how the camera works and how far out the IVS will trigger for your setting.
I get the optics. I certainly appreciate all the help, and obviously I am just frustrated because I realize many here are in "play" mode, working with a small-ish residential lot, and we have not only the 5ac to cover, we have several buildings, plus the desire to get prelim into outside the fenced area. And to add more, it is virtually a guarantee we will engage the bad guys. I walked the back city owned area with a city worker 2 days ago to discover the neighbor's fence was cut open. Our threat is real, but we can manage it because this ain't Chicago, LA, etc.
If I were to "cover" every point of interest, I would need 500 cams, ,and I am trying to get a grasp on how I want to do this. I cannot hover around "ID level DORI" for every location. Some, I simply need "trigger that IVS", in which that might invoke some extreme spot lights, sirens, and strobes, etc, etc. A false trigger is not even a big deal there because there will be no one around the area at night! This is not a residential area! No one is either there to complain, or even can because we are allowed certain things by right with the zoning.
It does sound like I need to think more on the PTZ or thermal options BUT I don't even know where the threats might come from yet. My overall intent is the scan well outside our property and alarm/trip before anything ever happens. But I also want a glimpse of what the trigger was. If I know it was indeed a human, solid! I don't even care if it's law enforcement, the system did its job.
If I were to "cover" every point of interest, I would need 500 cams, ,and I am trying to get a grasp on how I want to do this. I cannot hover around "ID level DORI" for every location. Some, I simply need "trigger that IVS", in which that might invoke some extreme spot lights, sirens, and strobes, etc, etc. A false trigger is not even a big deal there because there will be no one around the area at night! This is not a residential area! No one is either there to complain, or even can because we are allowed certain things by right with the zoning.
It does sound like I need to think more on the PTZ or thermal options BUT I don't even know where the threats might come from yet. My overall intent is the scan well outside our property and alarm/trip before anything ever happens. But I also want a glimpse of what the trigger was. If I know it was indeed a human, solid! I don't even care if it's law enforcement, the system did its job.
All 54IR are 5442..
- AS are fixed overview lens (sometimes there are called -ASE) - You choose between 2.8mm (110 degrees) and 3.6mm (90)
- ZE is overview zoom - starts at 114 degrees is going to 47
- Z4E is PROPER tele photo zoom - starts at 45 and going to 15..
If You want 90 view angle, look at fixed 3.6mm lenses (AS/ASE models)
2.8mm 5442/54IR gives You identify up to 20ft and recognition up to 40ft (dahua overestimated the distances in this model).
3.6mm 5442/54IR gives You identify up to 22ft and recognition up to 44ft
2.8mm 5842/58IR gives You identify up to 28ft and recognition up to 56ft
3.6mm 5842/58IR gives You identify up to 35ft and recognition up to 70ft
3.6mm gives You +25% range in comparison to 2.6mm
4K (8Mpx) gives You +50% range in comparison to 4Mpx..
the same You can have on ZE models zoomed to 90degrees horizontal FOV (but static lens models are brighter)...
3.6mm focal gives You less compressed image of people (they are bigger that on 2.8mm focal)..
Fixed 3.6mm is basic model I use most often as overview camera...
Don't look at Color4K at all - this camera model is more for fun that for real CCTV... Too limited and too many problems with them..
In very good lighting You can try to use 180 degrees cams (Color4k-B180). They are full color (no IR!!!!), they create very wide 8Mpx image from 2 sensors 4Mpx...
They work very well in middle on wall, at entrances etc..
But remember they are overview cams... and they have around 3ft half circle dead zone under camera...
there are many models:
- you use mostly/only overview cams around building (90% installations are done this way), in this situation 3.6mm 90 degrees hFoV 8MPx can give You some advantage,
- you use overview cams + zoomed cams directed at strategic places,
- you use overview cams + PTZ..
- you use overview cams + zoomed + PTZ depending of the needs / area..
each model have own pluses and minuses..
54IR, 58IR are the same as latest Dahua 5442/5842... there is no difference..
Color4K is previous generation, latest is not available in the USA..
First, You should order each model 1 pcs and play with them..
especially higher zoom variant (Z4E) which as I understand You don't feel now (You are talking non stop about overview cams)..
Also You should try one 25x PTZ... PTZ5A4M-25X is a killer at night, but with limited budget You can try PTZ425DB-AT or SDT4E425-8P-GB-APV1 (ptz + panorama)..
Second - even with project from cctvdesigntool.com always have 20-40% more cams in different models variations on car, because when You install cam and start watching into monitor with live view, usually things pop up that the project didn't take into account.
you'd be surprised but a few people who responded to you have more knowledge/experience than 90% of installers on the market...
Installations of this size are always a compromise between needs and possibilities and realities.
In most cases, installations of this type are mostly overview cameras, which record images and generate events from IVS. In most cases, they do not provide the ability to identify each person everywhere.
Along with the development of the project, cameras with zoom (telezoom) are added, directed at critically selected places where there is a high chance that a burglar will pass. Alternatively, PTZ cameras, which solve the problem of choosing this place.
P.S. identification is very important, but the reality with such large installations is that it cannot be ensured everywhere. And that worst burglars will have their faces covered anyway.
Last edited:
One more thing:
there is premium version of 5442/5842 called 5442H-ZHE / 5842H-ZHE, which are 5442/5842 with 7442/7842 (Ultra line) optics, chassis and double mic system.
they cost 30-40USD more, are bigger, more rugged and they have junction box in set.
Andy don't have this models in shop, buy You can order by DM (I ordered those models a few times)...
One biggest difference between normal and premium H-ZHE models are more open much brighter aperture for normal varifocal model.
Normal 5442-ZE have aperture which starts at F1.8.
Premium 5442H-ZHE have aperture which starts at F1.2
this is 2x more light (one F-stop) captured by sensor at most open aperture at minimum zoom level.
If you really interested in 5842 (due DORI), you can try to revert to some degree worse night performance by using 5842H-ZHE bullet models..
there is premium version of 5442/5842 called 5442H-ZHE / 5842H-ZHE, which are 5442/5842 with 7442/7842 (Ultra line) optics, chassis and double mic system.
they cost 30-40USD more, are bigger, more rugged and they have junction box in set.
Andy don't have this models in shop, buy You can order by DM (I ordered those models a few times)...
One biggest difference between normal and premium H-ZHE models are more open much brighter aperture for normal varifocal model.
Normal 5442-ZE have aperture which starts at F1.8.
Premium 5442H-ZHE have aperture which starts at F1.2
this is 2x more light (one F-stop) captured by sensor at most open aperture at minimum zoom level.
If you really interested in 5842 (due DORI), you can try to revert to some degree worse night performance by using 5842H-ZHE bullet models..
Last edited: