5mp practical resolution increase over 1080p

[aurmol 0]

It was ages ago when I was using 5mp camera.. at that time I seemed to remember that you get more dramatic increase in resolution going from 800x600 to 2Mp. I forgot how much better when you get 5mp versus 2Mp.. practically. If Ip cams are justified over hdcvi/hdtvi/ahd.. it is the resolution.

 

Also reviews said aptina sensor has greater dynamic range than sony exmor.. who here has reviewed or checked out the only dahua 5mp bullet cam..

 

http://www.dahuasecurity.com/products/ipc-hfw5502c-711.html

 

Is it an aptina or exmor or others?

how does it behave and image quality?

 

Or maybe we should wait for the 4K (8 megapixel) if the increase not so significant? My cellphone has 12mp and nokia cellphone has 40mp.. so cctv resolution is like reminiscing memories decade ago...

[buellwinkle 0]

Yes, you can do better than 1080P with 5MP, 2.5 times better.

 

Here's a 5MP image from a camera I recently tested, way better than any 1080P camera I ever tested. To see it full size, you have to click on the image, the hover over the image and click the 4 arrows in the lower right.

 

I've very impressed that you phone can do 12MP video. My Samsung S4 can only do 1080P.

 

[aurmol 0]

Yes, you can do better than 1080P with 5MP, 2.5 times better.

 

Here's a 5MP image from a camera I recently tested, way better than any 1080P camera I ever tested. To see it full size, you have to click on the image, the hover over the image and click the 4 arrows in the lower right.

 

I've very impressed that you phone can do 12MP video. My Samsung S4 can only do 1080P.

 

 

What I meant was 12mp still camera.. not video.. I don't think there is a phone that can record 12mp video.

 

Note the dahua 5mp is the same price as 5 pcs of 1080 hdcvi. So unless the 5mp is of superior dynamic range.. won't it make sense to put 5 pcs of 1080 hdcvi spread over the doorways and sides. My use is just outside the house and if I put 5 pcs of hdcvi (same price as 1 5mp cam), the robbers have to come together in 5 to disable them at same time, instead of just one.

 

Also any evidence 5mp cam has higher latency than 1080p ip cam. If the delay is even 2-3 seconds worse than 1080p ip cam, then 5 pcs of hdcvi is a clear answer.. you want real time view outside the house when you are holding the submachine gun trigger.. lol.. just kidding.. but 5 pcs real time analog 1080p may just blow any 5mp cam.. won't it.. unless you want to monitor the neighbors outside across the street.

[Boogieman 1]

Yes, you can do better than 1080P with 5MP, 2.5 times better.

 

Here's a 5MP image from a camera I recently tested, way better than any 1080P camera I ever tested. To see it full size, you have to click on the image, the hover over the image and click the 4 arrows in the lower right.

 

I've very impressed that you phone can do 12MP video. My Samsung S4 can only do 1080P.

 

 

What I meant was 12mp still camera.. not video.. I don't think there is a phone that can record 12mp video.

 

Note the dahua 5mp is the same price as 5 pcs of 1080 hdcvi. So unless the 5mp is of superior dynamic range.. won't it make sense to put 5 pcs of 1080 hdcvi spread over the doorways and sides. My use is just outside the house and if I put 5 pcs of hdcvi (same price as 1 5mp cam), the robbers have to come together in 5 to disable them at same time, instead of just one.

 

Also any evidence 5mp cam has higher latency than 1080p ip cam. If the delay is even 2-3 seconds worse than 1080p ip cam, then 5 pcs of hdcvi is a clear answer.. you want real time view outside the house when you are holding the submachine gun trigger.. lol.. just kidding.. but 5 pcs real time analog 1080p may just blow any 5mp cam.. won't it.. unless you want to monitor the neighbors outside across the street.

If your network is of decent quality you should not be getting any noticeable latency (a few hundred milliseconds at most)...it could be your network or NVR that is causing your issues...

[buellwinkle 0]

The image is from a Bosch 8000 Starlight camera would cost 4-5x what the 5MP Dahua would cost but at least gives you an idea of the detail potential. For outdoor use, you are looking at $1,600ish for camera, lens, housing. At least this give you 30fps, most 5mp cameras, including the Axis P3367 and Dahua you are looking at are only 12fps. Also, most 5mp cameras are 4:3 aspect ratio, what's cool in the Bosch is you can choose 16:9 for what they call 3K or chose 4:3.

 

Hard to put a direct value on it because for me, having say three 1080P cameras to get the same resolution (Bosch has a 6.1mp sensor, so comparable to 3), you would have to wire 3 cameras, 3 PoE ports, 3 windows to watch so there's a value beyond comparing solely on camera price. If there was NVR software that stitched 3 images into one, that would be cool. I know Maxicon is doing this at the front of his home, 3 cameras, each one covering 1/3 of the area. I'm actually in the process of setting this up, have 2 now, need the 3rd to provide the panorama view I want. But I'm doing something a little different, two 12mm looking off to the sides, a 4mm down the middle. I think he has a trio of 4mm cameras.

 

Yes, all IP cameras have some latency, especially compared to analog HDCVI cameras, but has never a been a problem for me.

[aurmol 0]

Yes, you can do better than 1080P with 5MP, 2.5 times better.

 

Here's a 5MP image from a camera I recently tested, way better than any 1080P camera I ever tested. To see it full size, you have to click on the image, the hover over the image and click the 4 arrows in the lower right.

 

I've very impressed that you phone can do 12MP video. My Samsung S4 can only do 1080P.

 

 

What I meant was 12mp still camera.. not video.. I don't think there is a phone that can record 12mp video.

 

Note the dahua 5mp is the same price as 5 pcs of 1080 hdcvi. So unless the 5mp is of superior dynamic range.. won't it make sense to put 5 pcs of 1080 hdcvi spread over the doorways and sides. My use is just outside the house and if I put 5 pcs of hdcvi (same price as 1 5mp cam), the robbers have to come together in 5 to disable them at same time, instead of just one.

 

Also any evidence 5mp cam has higher latency than 1080p ip cam. If the delay is even 2-3 seconds worse than 1080p ip cam, then 5 pcs of hdcvi is a clear answer.. you want real time view outside the house when you are holding the submachine gun trigger.. lol.. just kidding.. but 5 pcs real time analog 1080p may just blow any 5mp cam.. won't it.. unless you want to monitor the neighbors outside across the street.

If your network is of decent quality you should not be getting any noticeable latency (a few hundred milliseconds at most)...it could be your network or NVR that is causing your issues...

 

Are you saying each ip cam has to adjust its cpu to make sure the encoding won't take a certain hundred milliseconds (any standard limit they folloo).. for example if you put slow cpu in the 5mp, it would take 2 seconds to encode into H.264.. are there no manufacturer that do this? or do some do this and would tell you at least it's 5MP and cheap price even if the encoding takes 1.5 to 2 secs...

[Boogieman 1]

Ive never tested a 5mp camera, but the 3mp or 2mp have almost zero latency...You are making a mountain out of a molehill...but if you want cvi, just go for it...but run cat6 with baluns so you can change over to ip later

[aurmol 0]

Ive never tested a 5mp camera, but the 3mp or 2mp have almost zero latency...You are making a mountain out of a molehill...but if you want cvi, just go for it...but run cat6 with baluns so you can change over to ip later

 

But 1080p cvi seemed to have bad contrast that its dynamic range reduction can greatly reduce the resolving power and resolution maybe even lower than 720p. Are you familiar with those black and white bar chart and resolution test with decreasing contrast. Contrast can lower resolution and resolving power. I have scout the net for any reviews of hdcvi, hdtvi, ahd versus ip.. there is none. Why is there not even one reviews of them. If anyone has even encountered, please share it with us. Thanks.

[Boogieman 1]

They are a very small market segment, there is not much interest in it, that is why the reviews are limted. Its not cost effective, its more expensive to install, and you are limited to a few DVR's. You can get an ip camera for a few dollars more than cvi (you were over paying for the ip camera, look at ali express). You are over thinking this issue...just make a choice and install it...

[aurmol 0]

The image is from a Bosch 8000 Starlight camera would cost 4-5x what the 5MP Dahua would cost but at least gives you an idea of the detail potential. For outdoor use, you are looking at $1,600ish for camera, lens, housing. At least this give you 30fps, most 5mp cameras, including the Axis P3367 and Dahua you are looking at are only 12fps. Also, most 5mp cameras are 4:3 aspect ratio, what's cool in the Bosch is you can choose 16:9 for what they call 3K or chose 4:3.

 

By the way. For a $1,600 5mp cctv.. what is your estimate of the real manufacturing cost of the camea. I think it's just $200. So they are making a $1,400 profit. And customers even support them buying them.. so isn't the overpricing somewhat abusive thing already? Or do you honestly believe it takes say $1,400 to manufacturer a 5mp.. who knows.. these sensors may even be overstock from the days when 5mp digital camera reign.. early 2000s. So these manufacturers like Bosch, Axis are making a killing or excessive overpricing. Dahua costs 5 times less and they are already making incredible money. What more those Bosch.

[Boogieman 1]

The image is from a Bosch 8000 Starlight camera would cost 4-5x what the 5MP Dahua would cost but at least gives you an idea of the detail potential. For outdoor use, you are looking at $1,600ish for camera, lens, housing. At least this give you 30fps, most 5mp cameras, including the Axis P3367 and Dahua you are looking at are only 12fps. Also, most 5mp cameras are 4:3 aspect ratio, what's cool in the Bosch is you can choose 16:9 for what they call 3K or chose 4:3.

 

By the way. For a $1,600 5mp cctv.. what is your estimate of the real manufacturing cost of the camea. I think it's just $200. So they are making a $1,400 profit. And customers even support them buying them.. so isn't the overpricing somewhat abusive thing already? Or do you honestly believe it takes say $1,400 to manufacturer a 5mp.. who knows.. these sensors may even be overstock from the days when 5mp digital camera reign.. early 2000s. So these manufacturers like Bosch, Axis are making a killing or excessive overpricing. Dahua costs 5 times less and they are already making incredible money. What more those Bosch.

Huh? Do you know how much it cost for research and development, design, marketing, sales, employees, etc...

Go ahead and try and make your own...if you think they are using a 5mp sensor from cameras made in 2000 you dont understand cameras at all.

You can make the same statement about any device, your cell phone your pc..its simply wrong...

[buellwinkle 0]

You can buy say a Nikon camera that is 16mp for $99 and one for $1,500, does that mean Nikon is overcharging me when I want the $1,500 camera? I don't think the market would support 7-10x markup on a camera. You can count how many companies make a 5MP camera with 30fps on one hand with 4 fingers cut off, so if it's so simple, why doesn't Dahua or Hikvision do it? Heck why don't the other high end companies do it?

[Digiscan 0]

Yes, you can do better than 1080P with 5MP, 2.5 times better.

 

Here's a 5MP image from a camera I recently tested, way better than any 1080P camera I ever tested. To see it full size, you have to click on the image, the hover over the image and click the 4 arrows in the lower right.

 

I've very impressed that you phone can do 12MP video. My Samsung S4 can only do 1080P.

 

Damn, you can make out some pretty good detail on the plates on the cars across the street. Maybe twice the resolution at the most and you could see them with some cleaning up.

 

Have your neighbors ever expressed interest in your cams? Some might think it creepy but on the other hand some might be quite appreciative of having a high end system covering the front of their houses for no cost at all.

 

------

 

Whether that camera costs $200, $1200, or $19.99 to make is kind of immaterial. If you want that level of clarity you pay the money, otherwise wait until somebody else can do it. Such is always the way with tech! When you're the only person making it, you charge what you want and people buy it or they don't.

[MaxIcon 0]

Remember that linear resolution is what counts in resolving details.

 

1920 x 1080 = 2.1MP

2992 x 1680 = 5.0MP, 2.4x the area resolution of 1080p

 

Linear resolution increase from 1080p to this 5MP resolution, assuming the same FOV:

2992/1920 or 1680/1080 = 1.6x increase

 

When it comes to reading text or identifying features, you really only get 1.6x the linear resolution, even though the area resolution increase is 2.5x (which is 1.6 x 1.6).

 

To get 2x the linear increase for 2x the detail, you'd need 3840 x 2160, or 7.5MP; call it 8MP to make it easy.

 

At this point, the latency issues are going to be in decoding the images, since that increases with total resolution, not linear resolution. By the time you have 8 5MP cams, you'll have a much larger decoding load than the same number of 1080p cams, or a lot more 2MP cams for the same total MP.

 

Also, from a practical perspective, 5 1080p cams can give much better coverage than 2 5MP cams, since you can select more areas of interest with more cams. This has always been a flaw in the "1 1MP cam = 4 D1 cams" argument.

 

Like everything in cams, it's a set of trade-offs.

[thewireguys 3]

You can buy say a Nikon camera that is 16mp for $99 and one for $1,500, does that mean Nikon is overcharging me when I want the $1,500 camera? I don't think the market would support 7-10x markup on a camera. You can count how many companies make a 5MP camera with 30fps on one hand with 4 fingers cut off, so if it's so simple, why doesn't Dahua or Hikvision do it? Heck why don't the other high end companies do it?

 

FYI I have the 6MP 30fps Hikvision camera

[buellwinkle 0]

FYI I have the 6MP 30fps Hikvision camera

 

See, you can't name one 5MP that is 30fps, LOL.

[aurmol 0]

I finally got the 5mp dahua bullet. Here are some points.

 

1. When you lower the resolution to 3Mp, 1.3Mp, 1080p, D1, the images keep getting smaller and smaller as if they use only the exact corresponding pixels of the 5Mp sensor.. However in 720p, the vertical are cut with the same horizontal as the 5Mp, but the all the rest of resolution are severely cut in the horizontal and vertical.

 

2. The motorized Zoom and Focus button is not so responsive, it's better to adjust in the setup menu than directly in the unit.

 

3. The colors are very much like the sony exmor, very contrasty. Does it use exmor sensor too? What does it use? I'm not sure of the dynamic range compared to the aptina.

 

4. The field of the view of the 4mm lens is almost close to 90 degrees. Putting at a corner, you can see the adjacent walls... but there is barrel distortion due to the wide angle.

 

5. It has micro sd so not only can you record but you can Playback just like in NVR. But why the 64 Gig limit. Why can't one use 128 Gig or more. Is there data line differences between the 64 gig and 128 gig?

 

6. It has Alarm output and input too and the relays work.. but only accessible via the mobile gDMSS plus. Say, is it possible to port forward and access ip cam directly over the net without using any nvr or dvr?

 

7. In my country. Even banks use camera less than $75. The dahua 5mp costs a whopping $250. No one crazy here to buy them.. so I may return it last minute and get the Hdcvi since my target is not long distant but room size only and the Hdcvi and 5Mp can not be far off in performance.... don't know.. but the need to avoid NVR may make me keep it.

 

8. I'll test it tom how well can my Nvr4204 handle it. The Nvr4204 has incoming bandwidth of 200Mbps. So I wonder if it can do 1 5Mp and 3 1080p.. what do you think based on what you heard or read?

[buellwinkle 0]

Good feedback and don't forget, limited to 12fps at 5MP. I'm not sure why not 128GB cards as the big tech hurdle is at 64GB where it's the smallest of the SDXC. Maybe nobody at Hikvision could afford the $100 for the 128GB card, LOL. You can certainly try it and see. When I reviewed the Bosch, it said 32GB max, but worked fine with a 64GB card, the very same card that would not even work in a Hikvision that claimed to work with 64GB.

 

How was the image quality at 5MP. Was it comparable to their 1080P/3MP cameras but more pixels? How was low light performance compared to the 1080P/3MP cameras?

[MaxIcon 0]

1. When you lower the resolution to 3Mp, 1.3Mp, 1080p, D1, the images keep getting smaller and smaller as if they use only the exact corresponding pixels of the 5Mp sensor.. However in 720p, the vertical are cut with the same horizontal as the 5Mp, but the all the rest of resolution are severely cut in the horizontal and vertical.

 

So the 720p and 1080p have different FOVs? Dahua did that in older cams, but I thought they quit in the newer ones and scaled them to match. Customers complain about this, so most cams join the pixels these days to scale the image.

 

Vendors need to publish the actual FOV for every resolution to avoid these misunderstandings.

[aurmol 0]

If you will analyze the following pictures (which I erroneously posted in another thread in the reply) the 1.3MP, XGA, and D1 version of the dahua 5mp being 4:3 have about 60 degrees fov compared to the 5Mp resolution 87 degrees fov. The 3Mp, 1080p and 720p version have all different sizes. It's as if dahua just zoom the 3mp and 1080p part of the 5Mp frame.. and only the 720p has the same fov as the 5mp. Maybe the designers think since the cam can zoom from 4mm to 9mm and you don't buy a 5Mp to use 1080p, then they just mostly do zooms except for the 720p. So the correct term for the former is direct sampling and the 720p picture scaling? Why.. in other brands.. they have same field of view for the 16:9 and 4:3? And why would anyone use 1080p when they bought 5Mp?

 

Here are the images and field of view taken with the 5Mp fixed in position and target (distance of 1 meter). All resolutions taken 5mp, 3mp, 1.3mp 1080p, Xga, 720p, d1. So from the way dahua cuts them in each resolution.. how do dahua implement their resolution algorithm?

 

5 megapixel

 

..

 

3 megapixel (here dahua just cut the right side)

 

..

 

1.3 megapixel

 

..

 

1080p (dahua seems to cut it right from the 5mp frame)

 

..

 

XGA

 

..

 

720p

 

..

 

d1

 

..

 

Note in all shots, I didn't move the camera or target a single inch.

 

About how the 5mp compares to the 2mp/3mp in low light. It appears to do better than the 2mp sony exmol which as buellwinkle stated, has too great a contrast and some details loss. I don't have the 3mp aptina so not sure how it compares to the 3mp. I still have one day left before I can exchange it for others so I'm wondering if the 3mp aptina has greater dynamic range. I can get two 3mp aptina with the price of the 5mp.

Edited December 22, 2014 by Guest

[buellwinkle 0]

The Aptina has greater dynamic range. It has greater dynamic range with WDR off on the Aptina compared to the Exmor with WDR on, but the image won't be as pretty, meaning it won't have the bright colors and high contrast that people expect from point & shoot cameras. But from a surveillance perspective, I think the Aptina is better.

[aurmol 0]

The Aptina has greater dynamic range. It has greater dynamic range with WDR off on the Aptina compared to the Exmor with WDR on, but the image won't be as pretty, meaning it won't have the bright colors and high contrast that people expect from point & shoot cameras. But from a surveillance perspective, I think the Aptina is better.

 

Is it in the aptina technical literature or other references that it has wider dynamic range than the sony exmor.. or is it your own deduction?

 

Because if the Aptina has really greater dynamic range on spec sheet and the 5MP dynamic range is not as good as the aptina (any theory what sensor the 5Mp use?).. then the aptina may even resolve more.. and I have only 6 hours left before I can exchange the 5mp for two 1080p aptina. What you think?

 

I found out 64 gig micro sd is expensive.. you can get a 2 TB hard disc with two 64G micro sd. And saving 5mp takes up space time. it will take 1.5 days to fill up the 64G at 5Mp 12fps. So the built in Sd lot and even on menu Playback may not give much of advantage

 

The only think I like about the 5Mp is its zoom can be controlled inside the setup menu because it is vari-motorized.. so it acts like ptz zoom (4mm to 9mm or 2.2 optical zoom.. better than nothing)

 

It can accept microphone and audio input too. Is this useful to record sounds at roads.

 

Is there evidence the 5mp can show much more detail than the 3mp? Haven't tested it side by side with the 2mp/3mp because these are already installed on top at another location and the 5Mp is heavy to climb up and put side to side.

 

Please share what you got so I can decide whether to get the 3mp aptina instead or even the hdcvi. With the price of 1 5mp, I can 3 1080p hdcvi complete with recorder.

[MaxIcon 0]

That cam has motorized zoom? That can cause differences in the FOV, of course. Does it remember different zoom levels for different resolution settings?

 

I'd assume they keep the same setting for each resolution change, but assumptions can be wrong. It's always hard to say what Dahua's thinking when they implement their firmware.

 

Assuming no zoom changes, it looks like they mix direct sampling and scaling. I'd call it lazy or uninformed programming, because it's easy enough to scale the images - the sensor typically has scaling modes built in for all the common resolutions. I don't know why anyone would do that on purpose, as it creates customer confusion and unhappiness.

 

I haven't compared the 4300S to my HFW3300C, but the image on the 4300S is definitely less saturated, as BW says. It's also a little less noisy with better detail at night compared to the Hik 2032, which it was mounted next to for a while. Not a huge difference, but noticeable.

 

5MP only gives you 1.4x the linear resolution of 1080p. I'd select 2 or 3 1080p cams over one 5MP, since you can get much better coverage of key areas with 3 cams than with one.

[aurmol 0]

That cam has motorized zoom? That can cause differences in the FOV, of course. Does it remember different zoom levels for different resolution settings?

 

I'd assume they keep the same setting for each resolution change, but assumptions can be wrong. It's always hard to say what Dahua's thinking when they implement their firmware.

 

Assuming no zoom changes, it looks like they mix direct sampling and scaling. I'd call it lazy or uninformed programming, because it's easy enough to scale the images - the sensor typically has scaling modes built in for all the common resolutions. I don't know why anyone would do that on purpose, as it creates customer confusion and unhappiness.

 

I haven't compared the 4300S to my HFW3300C, but the image on the 4300S is definitely less saturated, as BW says. It's also a little less noisy with better detail at night compared to the Hik 2032, which it was mounted next to for a while. Not a huge difference, but noticeable.

 

5MP only gives you 1.4x the linear resolution of 1080p. I'd select 2 or 3 1080p cams over one 5MP, since you can get much better coverage of key areas with 3 cams than with one.

 

I tested it thoroughly a while ago. The zoom level is the same for all resolutions. Meaning those pictures I shared above are all zoom out (maximum fov). If I set it to say 1080p and zoom in middle. And return to 5Mp or other resolutions. It will remain zoom in middle. So the zoom is the same for all resolution. It doesn't remember. So in the 1.3p resolution where dahua did direct sampling making the image biggest and you use the zoom to 9mm... the image is even closer..

 

I also directly compared the fov with the sony exmor.. I noticed it is narrower a bit.. so I think it's right when they spec the 5mp as 82 degrees FOV vs the sony exmor 88 degree fov.

 

Now because the 5Mp is only 12fps and you get much narrower fov when set to other resolutions except 720p.. I plan to return it in exchange for one of the famed 4300s.. but what I'll get is the 4300e.

 

I also tested the dahua nvr4204 with 200 Mbps incoming bandwidth. Well.. the 4 channel can take one 5mp and three 1080p resolution ip cams. The supplier told me the maximum is only two 1080p and they told this to all customers. No wonder no one buys their ip cams and instead get their hdcvi...

 

The problem with the hdcvi is the fov is narrow.. almost the same as the 4300s aptina at 72 degrees. And the image paler than the aptina. But the cleanliness of the hdcvi at wdr set to 50% is close the sony exmor except if you see really close at the shadows and more noise in the hdcvi.

 

So it seems I'm stuck now getting another sony exmor at 88 degrees and a 3mp aptina to test out its famed greater dynamic range but at only 72 in exchange for returning the 5Mp.

 

HiKvision are double the price of the Dahuas. And Avtech is expensive but don't have any P2P. Is there no other brands that are good enough. The 2Mp Jovision have horrible dynamic range. What else good enough with large fov and great dynamic range. Is there a 3Mp Hikvision model that meets this requirement?

[aurmol 0]

To continue the above comment. I noticed one great point about the 5mp though. Direct sampling of the sensor is equivalent to digital zoom. So when you set it from 5mp to 1.3mp. It's like you have a digital zoom of say 2x. And when you use the motorized zoom in from 4mm to 9mm.. it's like optical zoom of 2.25X. So total zoom is like 4.5. See the following image comparison.

 

1.3MP Zoom in all way to 9mm

 

..

 

compared this to the 5MP Zoom out (4mm)

 

..

 

What's great of this vari-focal motorized 4-9mm is you can adjust the zoom right in the setup menu and even at PSS. So it's like having the zoom part of the PTZ in the 5mp. This is the only good thing I noticed of this 5mp. Won't this feature alone be worth it? PTZ are 3X more expensive than the dahua 5mp. At least in the 5mp you get some 4.5 Zoom that can be controlled at computer (mobile gDMSS though can't make you adjust the zoom but at least desktop PSS does). But for zoom to be really worth it.. How do you think should be the minimum magnifying power.. like 10X optical zoom? I only need to know your opinion of this zoom point so I can decide whether to return or keep the 5mp. I have less than a day. Thanks all for the tips.

Edited December 22, 2014 by Guest