What is a 500W HD camera and how can it be 500 million pixels?

[edy]

I was asked to review a product that has a few interesting "specs" which seem to contradict each other and I need some help decyphering it. I can't give too many details yet, but basically one advertisement says it has 500 million pixels, and yet on another video it says 500W HD camera. So what gives?

Doing a quick Google search I noted that "500W" shows up a lot with (1080p) written next to it. What does the 500W mean, as I'm not familiar with this term with respect to camera resolutions. My understanding is that 1080p refers to 1920 x 1080, which is an HD resolution of roughly 2 MegaPixels (2 million pixels). It is NOT 4K, although I've seen some ads even say cameras with 500W-4K.

Certainly the "W" does not mean width, and I can't seem to find any reason why "500" would have anything to do with 1920x1080, or with megapixels. And the "W" can't be referring to Watts either. 

Now with respect to the 500 million pixels reference, I think that's total bull$#!t. A camera with 500,000,000 pixels would have to have a 16:9 resolution of about 29814 x 16771 = 500,010,594. It may be that they are "capturing" a scene and stitching the images together to a maximum stitched image of that 500 MP resolution and they are using misleading marketing.

Another question I had was that when they talk about a 1000 x 1000 pixel camera (for example) the images that come out of that camera are 1000 x 1000 pixels in COLOR. Does that mean that for each color pixel, there are actually 4 sensing elements in the sensor (usually with color masks on top... 1 blue, 1 red and 2 greens). So could a manufacturer try to inflate the number of pixels by referring to the actual CCD sensing elements? Would a 1080p camera, which outputs COLOR images of 1920x1080 resolution, ACTUALLY HAVE a CCD which is 2x the width and height due to the additional color masks - each pixel being a 2x2 array of underlying monochrome sensing pixels beneath color masks)?

[BrokenYugo]

Sounds like Chinese bullshittery to me, a quick Google of "500W HD camera" confirms it.

[Siwastaja]

You are overthinking this.

A company with proper, actual technology will explain the terms they invent.

This is just some con artist putting random numbers and letters together. Like 10000W PMPO boomboxes in 1990's. They come with some gibberish and then they copy each other. Next thing is, someone upgrades their competition to 750W HD camera.

Regarding your question about 1000x1000 pixel color camera, in the typical/normal/default case, this only means 1000x1000 monochrome pixels coated with a so called Bayer filter in which 50% of the pixels sense green, and 25% and 25% sense red and blue. The gaps in color channels are interpolated in camera software (firmware, or when shooting "raw" files, this can be done later). These interpolation algoritms are nowadays much better than just linearly filling in the gaps. After all, you have actual image information at each pixel site, just in "wrong color", but an advanced algorithm can get quite close to the ideal/imaginary case where the same site would capture all three colors, by utilizing the fact that real world often consists of dull colors with lot of correlation between color channels.

This is the widely accepted way of specifying number of pixels, you just need to know this, you are not getting 3x1000x1000 actual pixels, but quite close to it in general purpose imaging. But if you are shooting a scene in deep red lighting which only registers in red pixels, you are going to notice significant reduction in resolution because the camera has only 25% of pixels to work with.

In earlier broadcast video cameras (don't know the situation as it is today), it was common to optically split the image to three separate monochrome CCDs. When they got the registration exactly right, these systems really did have each pixel site registering all three color. I bet this is quite rare today as CCDs (or CMOS sensors, more often now!) can be just manufactured to have ridiculously high number of pixels, like 20 million. The resolution is huge even if you need to interpolate colors. It would be nearly impossible to split beams and register three separate CCDs to that accuracy.

[edy]

Here is a typical example of the "500W HD" Google search result:

https://www.walmart.ca/en/ip/500w-HD-Wifi-Endoscope-Camera-1920P-Flexible-IP67-Waterproof-Inspection-6LEDs-Adjustable-Borescope-Camera-2-meters-hard-line/2JREZVK5UCLB

Nowhere in the specifications for the above Walmart camera do we have any hint of what 500W refers to:

Product Category:Portable camera

Lens Diameter: 5.5 mm
Camera light source: 6 adjustable LED S
Working temperature: 32 to 113 degrees Fahrenheit (0-45 degrees)
Battery: 1800mA Lithium Battery
Working hours: 1.5H
Power input: DC 5V 500 mAh/1A
Waterproof: IP67 (for lens only)
Focal length: 5cm~ infinity
Camera resolution: 2560*1920
Sensor pixels: 5 million pixels
Perspective: 70 degrees
Support system: IOS Android
Photo format: MJPEG
Video format: MP4

And here's an Alibaba camera that has 500W written in the description somewhere:

https://www.alibaba.com/product-detail/4k-Mini-OV5640-Sensor-Full-HD_1600294725966.html

*IF* and when I get the product to review I'll let you all know. For now these review requests are typically bait-and-switch or scam marketing so we'll see. I give them my work address and make sure they pay for all duty/tax/import fees otherwise I won't pick it up, and if they want it back they'd need to supply return postage paid labels and packaging or it's going potentially into the teardown video channel after. 

[Siwastaja]

These are just cheap toys. Typically they kind-of work, but that's about it. If you just want any camera capable of storing some kind of image/video at whatever potato quality, buy one. It's very likely the image sensor isn't even full-HD, it's normal for such cheap toys to employ lower resolution sensors then just upscale in software (which only makes the files bigger, and saving them slower).

Clearly that 500W is now some fairly new trend term that just indicates cheap toy camera. They copy each other so it's not surprising they are suddenly everywhere. Soon they'll come up with something else.

[tooki]

In earlier broadcast video cameras (don't know the situation as it is today), it was common to optically split the image to three separate monochrome CCDs. When they got the registration exactly right, these systems really did have each pixel site registering all three color. I bet this is quite rare today as CCDs (or CMOS sensors, more often now!) can be just manufactured to have ridiculously high number of pixels, like 20 million. The resolution is huge even if you need to interpolate colors. It would be nearly impossible to split beams and register three separate CCDs to that accuracy.

Actually, true broadcast cameras are still 3-sensor devices, albeit CMOS these days and not CCD.

This article sums it up nicely: https://www.redsharknews.com/production/item/6169-whatever-happened-to-the-3-chip-camera

This is the 3-CMOS 8K Sony camera they mention: https://pro.sony/en_GB/products/4k-and-hd-camera-systems/uhc-8300

[Siwastaja]

Thanks, that is quite obvious actually, the advantage of such 3CCD broadcast cameras hasn't gone anywhere; it's the relatively better performance at that mid-range sensor size. The other options would be to go lower performance using the same chip size, or go to larger chips.

But that mid-range sensor size is excellent because it enables effectively very long zoom lenses, yet it enables at least modest amounts of depth-of-field trickery. Great for news, sports etc., and to limited extent, low-to-mid-cost drama production.

I used to own one such 3CCD camera, although it was a mid-range S-VHS model with just 1/2" CCDs. In any case, it already was able to show the benefits compared to single chip cameras of the same age. Although sensors have got better since then, the relative difference should be the same.

[Someone]

In earlier broadcast video cameras (don't know the situation as it is today), it was common to optically split the image to three separate monochrome CCDs. When they got the registration exactly right, these systems really did have each pixel site registering all three color. I bet this is quite rare today as CCDs (or CMOS sensors, more often now!) can be just manufactured to have ridiculously high number of pixels, like 20 million. The resolution is huge even if you need to interpolate colors. It would be nearly impossible to split beams and register three separate CCDs to that accuracy.

Actually, true broadcast cameras are still 3-sensor devices, albeit CMOS these days and not CCD.

This article sums it up nicely: https://www.redsharknews.com/production/item/6169-whatever-happened-to-the-3-chip-camera

This is the 3-CMOS 8K Sony camera they mention: https://pro.sony/en_GB/products/4k-and-hd-camera-systems/uhc-8300

Like a "true" scotsman?
EBU disagree:
https://tech.ebu.ch/cameratests
They are happy to use specifications of performance as the measure rather than the technical way used to achieve it. While it has been difficult to match 3 (often CCD) sensor performance with a single sensor, that is falling away rapidly.

I used to own one such 3CCD camera, although it was a mid-range S-VHS model with just 1/2" CCDs. In any case, it already was able to show the benefits compared to single chip cameras of the same age. Although sensors have got better since then, the relative difference should be the same.

With all the relatively complicated optics needed (either way) the sensors are starting to become the "easy" bit. The performance difference between 3 chip and single chip cameras continues to narrow to the point it will become artistic rather than technical.

[tooki]

In earlier broadcast video cameras (don't know the situation as it is today), it was common to optically split the image to three separate monochrome CCDs. When they got the registration exactly right, these systems really did have each pixel site registering all three color. I bet this is quite rare today as CCDs (or CMOS sensors, more often now!) can be just manufactured to have ridiculously high number of pixels, like 20 million. The resolution is huge even if you need to interpolate colors. It would be nearly impossible to split beams and register three separate CCDs to that accuracy.

Actually, true broadcast cameras are still 3-sensor devices, albeit CMOS these days and not CCD.

This article sums it up nicely: https://www.redsharknews.com/production/item/6169-whatever-happened-to-the-3-chip-camera

This is the 3-CMOS 8K Sony camera they mention: https://pro.sony/en_GB/products/4k-and-hd-camera-systems/uhc-8300

Like a "true" scotsman?
EBU disagree:
https://tech.ebu.ch/cameratests
They are happy to use specifications of performance as the measure rather than the technical way used to achieve it. While it has been difficult to match 3 (often CCD) sensor performance with a single sensor, that is falling away rapidly.

Didn’t bother to read the article I linked, didja!? Sensor performance isn’t the issue. It’s lenses. You can’t make the insane zoom lenses used in broadcast for big sensors. They’re already $50K-300K lenses; to make them for the large sensors that provide similar quality would be prohibitively large and expensive.

So the EBU doesn’t “disagree” because sensor performance was not something I even addressed.

Note also that broadcast ≠ cinema.

[Circlotron]

This is the 3-CMOS 8K Sony camera they mention: https://pro.sony/en_GB/products/4k-and-hd-camera-systems/uhc-8300

Wow! 

And 3 separate sensors would allow for larger pixels for the same sensor size, therefore better s/n and low light performance presumably.

[jmelson]

And 3 separate sensors would allow for larger pixels for the same sensor size, therefore better s/n and low light performance presumably.

The dichroic image splitters use ALL the incoming light.  The splitters send the red photons to one sensor, the blue ones to another, and the rest must be the green ones going to the 3rd sensor.  That is a lot better than blocking most of the light with filters, but more expensive.
Jon

[Someone]

In earlier broadcast video cameras (don't know the situation as it is today), it was common to optically split the image to three separate monochrome CCDs. When they got the registration exactly right, these systems really did have each pixel site registering all three color. I bet this is quite rare today as CCDs (or CMOS sensors, more often now!) can be just manufactured to have ridiculously high number of pixels, like 20 million. The resolution is huge even if you need to interpolate colors. It would be nearly impossible to split beams and register three separate CCDs to that accuracy.

Actually, true broadcast cameras are still 3-sensor devices, albeit CMOS these days and not CCD.

This article sums it up nicely: https://www.redsharknews.com/production/item/6169-whatever-happened-to-the-3-chip-camera

This is the 3-CMOS 8K Sony camera they mention: https://pro.sony/en_GB/products/4k-and-hd-camera-systems/uhc-8300

Like a "true" scotsman?
EBU disagree:
https://tech.ebu.ch/cameratests
They are happy to use specifications of performance as the measure rather than the technical way used to achieve it. While it has been difficult to match 3 (often CCD) sensor performance with a single sensor, that is falling away rapidly.

Didn’t bother to read the article I linked, didja!? Sensor performance isn’t the issue. It’s lenses. You can’t make the insane zoom lenses used in broadcast for big sensors. They’re already $50K-300K lenses; to make them for the large sensors that provide similar quality would be prohibitively large and expensive.

So the EBU doesn’t “disagree” because sensor performance was not something I even addressed.

Note also that broadcast ≠ cinema.

You can pipe a broadcast lens (B4-mount etc) to arbitrary sensors, with some fancy optics as I specifically mentioned. Either 3 chip or single sensor needs some optical processing to make it all work properly, and then along the way though [optical processing block] the image on the sensor can be scaled to match the lens interface specification.

EBU says broadcast is about measurable performance/quality of the system, not how you achieve it (they do go on to list specifically how that is achieved in current examples).

And 3 separate sensors would allow for larger pixels for the same sensor size, therefore better s/n and low light performance presumably.

The dichroic image splitters use ALL the incoming light.  The splitters send the red photons to one sensor, the blue ones to another, and the rest must be the green ones going to the 3rd sensor.  That is a lot better than blocking most of the light with filters, but more expensive.
Jon

Yes, eliminating bayer loss is the major difference in performance. Some groups are trying to do that at the sensor level now.

And 3 separate sensors would allow for larger pixels for the same sensor size, therefore better s/n and low light performance presumably.

Changing the size of the sensor doesn't change the amount of light (as that was set by the lens) so larger pixels aren't always better.

[Siwastaja]

You can pipe a broadcast lens (B4-mount etc) to arbitrary sensors

Of course you can but then you are just using part of the sensor area, and the end result is equivalent to using the smaller sensor (one meant for the lens!) to begin with.

Modest sensor size has obvious set of both advantages and disadvantages. Almost ridiculously long telephoto zoom lenses is one of the advantages. So-called 3/4" sensor size has been a sweet spot for television work including news, sports etc. And it's pretty obvious that a 3CCD * 3/4" performs better than 1CCD * 3/4" in terms of resolution and SNR. If you want the same improvement by just upping the sensor size to compensate for the light and resolution loss of Bayer filter, then you need to make the lenses draw a larger image area, and increase the focal length to match the same field-of-view. This telephoto zoom lens is the costly, large and heavy part.

This is a classic optimization case, moving the complexity, weight and cost between the camera body and lenses.

[Zero999]

How small can the pixels be? I would have thought that diffraction would limit the smallest pixel size to half the wavelength of the longest wavelength of light it needs to respond to. Assuming that would be 700nm, the pixels can be no smaller than 350nm, so a 500M pixel CCD would need to be larger than 350×10^-9√(500×10⁶) = 7.83mm by 7.83mm.

[Siwastaja]

Getting even remotely decent SNR requires counting many thousands of photons for each video frame even assuming 100% photon to electron conversion efficiency and perfect readout and ADC. And if this is for video, you have some dozen milliseconds to collect them.

You can bin the pixels together (even at analog "charge" level) to get more sensitivity at the expense of less resolution (i.e., make the pixel size a runtime parameter), but if you are going to almost always enable binning, then it's only a negative compromise.

This sets the practical pixel size so that you get visually noise-free image in some medium amount of light. Then low-light performance can be boosted by pixel binning (analog or digital), longer exposures, noise reduction algorithms etc. etc. But if you need to rely on those in decent lighting conditions, you made the pixels too small and the high resolution is just wasted, but brings all the negative sides like reduction in fill rate, readout delays / bandwidth requirement, more expensive ADCs and processing, and so on.

Do note that use cases where high resolutions are desired, it's quite normal to require good image quality overall. Some 20-50 megapixels practically "need" the classic 36x24mm sensor size already although you could theoretically do like 1000 megapixels.

[edy]

I found a link to a camera that has both 300W and 500W options. I can't translate the text properly so I am not sure if it says what the difference is, but perhaps this gives some clues:

http://www.newbecca.com/product/42953603959

Here's another interesting example of a Drone Toy which includes 500w in the description, yet under the "color classification" specs later it mentions 30w and then 200w:

https://www.amazon.ca/Headless-Routing-Real-Time-Transmission-Pressure/dp/B08727X4RY

Trying not to overthink this but seems strange that it would be included in so many descriptions. Assuming it has nothing to do with a spec, perhaps it is some chip/sensor model number that is being sourced for these various cameras. I did find this answer on Amazon to some 200w item:

200W=200WAN(Chinese) means 2MP. Thanks.

Imake Seller · May 3, 2015
7 of 7 found this helpful. Do you?
Yes
 
No
 | Report abuse
This is the resolution of the camera which is 200 pixels wide. The higher the pixels, the better the detail in the images.

Daniel G. · May 2, 2015
2 of 8 found this helpful. Do you?
Yes
 
No
 | Report abuse
It's a 2 megapixil cam or 2 million pixels was the way I read it just got mine I'm learning to fly

So I had a look at Wan meaning in Chinese:

What number is Wan in Chinese?
万(wàn): ten thousand.

If w=10,000, then 500w = 500 x 10,000 = 5,000,000 or 5 MP. BINGO! 

Obviously 5MP is a far cry from 500,000,000 (500 MP) but makes more sense. A 5MP camera would be resolution of about 2592 x 1944, certainly better than a 1080p (2.1MP) camera (assuming they are actually using this and not faking the resolution or specs). We'll find out in a few weeks, the company says it will be ready then (not holding my breath). Still not happy their website is claiming 500 million pixels but this could be an *honest* translation error (just like "w" meaning 10,000 maybe they thought in English it means million). It makes for some interesting financial transactions... did you say 10,000 in change? Ok, here's a million. 

[SiliconWizard]

5MP sensors are nothing special these days.
Some smartphones now have 48MP sensors (and maybe even up to 108MP I think.) Doesn't mean they are really worth it, but the pixels are there.

That's still an order of magnitude from the 500MP mark.
Samsung has announced something like this: https://www.techradar.com/news/samsung-working-on-camera-sensors-that-bests-the-human-eye

The product is not there though and you'd never find any camera with something like this at this price point. It's not just the sensor; it would require very powerful image processing chips to make any use of them, and an enormous amount of storage.

[Someone]

You can pipe a broadcast lens (B4-mount etc) to arbitrary sensors

Of course you can but then you are just using part of the sensor area, and the end result is equivalent to using the smaller sensor (one meant for the lens!) to begin with.

FFS, how clearly do I have to say this:

You can pipe a broadcast lens (B4-mount etc) to arbitrary sensors, with some fancy optics as I specifically mentioned. Either 3 chip or single sensor needs some optical processing to make it all work properly, and then along the way though [optical processing block] the image on the sensor can be scaled to match the lens interface specification.

No matter if its singe or 3x sensor, there is a pile of optical processing sitting between a "broadcast" B4 lens and the sensor(s). In some products that includes scaling of the image to fit in a larger or smaller optical format sensor. At that point the choice of image sensor/pixel size is relatively open.

How small can the pixels be? I would have thought that diffraction would limit the smallest pixel size to half the wavelength of the longest wavelength of light it needs to respond to. Assuming that would be 700nm, the pixels can be no smaller than 350nm, so a 500M pixel CCD would need to be larger than 350×10^-9√(500×10⁶) = 7.83mm by 7.83mm.

Light needs some depth to be absorbed as there are boundary/surface effects in the materials. Add some other material to keep light from spilling into neighbouring pixels and it gets hard to make things smaller than the 700nm pixels currently state of the art:
https://www.techinsights.com/blog/part-2-pixel-scaling-and-scaling-enablers

[Siwastaja]

I don't like to discuss with people who are so angry and curse with no reason, but I'll let it slip this time.

Optical scaling between lens mount and sensor, to magnify the image to fit larger sensor? Really?

Of course it's doable. Not saying this isn't possible, but it's going to be a huge compromise and will never work as well as a lens designed for the sensor size.

Any references of such system?

[Someone]

Optical scaling between lens mount and sensor, to magnify the image to fit larger sensor? Really?

Of course it's doable. Not saying this isn't possible, but it's going to be a huge compromise and will never work as well as a lens designed for the sensor size.

Any references of such system?

Its not just possible but relatively common, and well documented/publicized. Yet you came in with nothing more than an outright refusal that it was even possible, you didn't concede it was impractical or doable. Instead took a quote out of the explicit explanation that there are optics involved (only the crappiest of B4 mount cameras would be direct to sensor as the lenses arent designed for that use) and said it would have to be a fixed image size.

So since you cant even do the most basic of research before claiming you know about a field (which clearly the commentators in here don't know much beyond a cursory glance), here is a screenshot from the Panasonic AK-UC3300GJ brochure.

[Simon]

come on children, settle down.

[magic]

How small can the pixels be? I would have thought that diffraction would limit the smallest pixel size to half the wavelength of the longest wavelength of light it needs to respond to. Assuming that would be 700nm, the pixels can be no smaller than 350nm, so a 500M pixel CCD would need to be larger than 350×10^-9√(500×10⁶) = 7.83mm by 7.83mm.

Light needs some depth to be absorbed as there are boundary/surface effects in the materials. Add some other material to keep light from spilling into neighbouring pixels and it gets hard to make things smaller than the 700nm pixels currently state of the art:
https://www.techinsights.com/blog/part-2-pixel-scaling-and-scaling-enablers

The effect of diffraction is actually determined by the F-number of the lens alone.
https://www.cambridgeincolour.com/tutorials/diffraction-photography.htm

Below 1µ pixel size you hit a point of diminishing returns where something may perhaps be recovered with heavy processing if the lens is on the fast side f/1~f/2.

Nokia 808 was said to use the diffraction from its f/2.4 lens as the antialias filter. Pixel pitch was 1.4µ.

[Zero999]

How small can the pixels be? I would have thought that diffraction would limit the smallest pixel size to half the wavelength of the longest wavelength of light it needs to respond to. Assuming that would be 700nm, the pixels can be no smaller than 350nm, so a 500M pixel CCD would need to be larger than 350×10^-9√(500×10⁶) = 7.83mm by 7.83mm.

Light needs some depth to be absorbed as there are boundary/surface effects in the materials. Add some other material to keep light from spilling into neighbouring pixels and it gets hard to make things smaller than the 700nm pixels currently state of the art:
https://www.techinsights.com/blog/part-2-pixel-scaling-and-scaling-enablers

The effect of diffraction is actually determined by the F-number of the lens alone.
https://www.cambridgeincolour.com/tutorials/diffraction-photography.htm

Below 1µ pixel size you hit a point of diminishing returns where something may perhaps be recovered with heavy processing if the lens is on the fast side f/1~f/2.

Nokia 808 was said to use the diffraction from its f/2.4 lens as the antialias filter. Pixel pitch was 1.4µ.

Great, so a 500M pixel full colour CCD would need to be nearly 45mm by 45mm and cost a huge amount of money.

[TimFox]

Not video, but this reminds me of the time I was photographing cityscapes in downtown Chicago with my 8x10 view camera and Ektachrome E100G sheet film, and a tourist asked me how many megapixels I had.
A rough mental computation, and I replied "about 500".  Given the roughly 80 lp/mm resolution of the film, this was about right, since the film area was 51,600 mm².  (Assuming 1 lp = 2 pixels.)

[Circlotron]

I wish there was an easy way to use a flatbed scanner as a camera. ~150 megapixels for everybody!