Ugly Hedgehog - Photography Forum
Home Active Topics Newest Pictures Search Login Register
Main Photography Discussion
Megapixel vs. sensor size?
Page 1 of 3 next> last>>
Mar 15, 2019 13:03:12   #
Spectre Loc: Bothell, Washington
 
Help this old man. I’ve read here that sensor size is needed for better sharpness and also megapixel size is also. How do these two relate?

Reply
Mar 15, 2019 13:14:40   #
PixelStan77 Loc: Vermont/Chicago
 
Spectre wrote:
Help this old man. I’ve read here that sensor size is needed for better sharpness and also megapixel size is also. How do these two relate?
Hope this helps old man.

http://www.laesieworks.com/digicom/MP.html

Reply
Mar 15, 2019 13:18:33   #
repleo Loc: Boston
 
Same number of megapixels on a larger sensor means bigger pixels sites or 'buckets'. Bigger buckets can capture more information per bucket. More buckets on same size sensor means smaller buckets resulting in less information.

Reply
 
 
Mar 15, 2019 13:20:18   #
CHG_CANON Loc: the Windy City
 
Sensors are measured in their pixel resolution. The megapixel measurement is simply the length * width measure (in pixels) / 1,000,000. So, a 24MP Nikon D7200 contains a sensor of 6000 x 4000 pixels.

All other attributes being equal, the more pixels, the sharper (the more finely detailed) will be the image with the higher pixel resolution, when viewed at 100%, whether on your computer screen or printed to large formats such as 20x30-inches and larger. Sensors at 24MP or higher are considered to have now surpassed the resolution of most grades of 35mm film.

Reply
Mar 15, 2019 13:40:27   #
burkphoto Loc: High Point, NC
 
Spectre wrote:
Help this old man. I’ve read here that sensor size is needed for better sharpness and also megapixel size is also. How do these two relate?


Sensor size and megapixel count are interrelated — Sensor size determines the *area* available for sensels. A sensel is what records either red, or green, or blue information (it's a monochrome sensor with a filter over it). Data from adjacent RGGB sensels (yes, there are usually TWO green filtered sensels for each red sensel and blue sensel) is amplified (especially when you turn up the ISO control), digitized from the analog voltages, then merged together with some *very* fancy algorithms to produce an image bitmap (file pixels!) for JPEG output. OR, the digitized information is saved in a raw file... an unprocessed glob of data arrays that is similar by analogy to a latent (undeveloped) image on film.

HOW BIG the sensels are (width/diameter/area) determines how many photons the sensel can convert to electrons during a given exposure. The stronger the signal, the higher the signal-to-noise ratio. Higher signal-to-noise ratios mean better color depth, and better gradation, and less visible noise in the image. The camera can work in lower light levels with the same amount of noise as a camera with a smaller sensor and the same megapixel count.

So... a 24MP full frame sensor has about double the area of an APS-C or DX sensor, which translates to a one stop advantage over a 24MP APS-C or DX sensor. A 24MP APS-C or DX sensor has roughly a one stop advantage over a (20MP) Micro 4/3 sensor. And so it goes... Sensel size varies with both the area of the sensor "chip", and the megapixel count the manufacturer engineers.

Whether this is worth worrying about in real life depends very much on what you are trying to achieve with your camera! It can matter a lot, or not at all.

Reply
Mar 15, 2019 13:41:18   #
Gene51 Loc: Yonkers, NY, now in LSD (LowerSlowerDelaware)
 
repleo wrote:
Same number of megapixels on a larger sensor means bigger pixels sites or 'buckets'. Bigger buckets can capture more information per bucket. More buckets on same size sensor means smaller buckets resulting in less information.


Please define "information".

Reply
Mar 15, 2019 14:05:13   #
repleo Loc: Boston
 
Gene51 wrote:
Please define "information".


....how many photons the sensel can convert to electrons during a given exposure...

(See Burkephoto post above)

Reply
 
 
Mar 15, 2019 14:19:02   #
speters Loc: Grangeville/Idaho
 
Spectre wrote:
Help this old man. I’ve read here that sensor size is needed for better sharpness and also megapixel size is also. How do these two relate?
None of it is correct, yes, more megapixels can translate into more detail, but the most important part in this is the lens. Larger pixels are also better, as they are letting in more light, so a bigger sensor can hold more pixels that can still be quite large. That in return means, a larger sensor is usually better for low light, but then again, the lens does the most important task, recording the image to the sensor. A crappy lens on the best sensor, will still procuce a crappy image, while a good lens on a mediocre or somewhat crappy sensor, can still produce a decent image!

Reply
Mar 15, 2019 14:32:45   #
JD750 Loc: SoCal
 
Spectre wrote:
Help this old man. I’ve read here that sensor size is needed for better sharpness and also megapixel size is also. How do these two relate?


Lots of things affect sharpness, the sensor is one of them.

Reply
Mar 15, 2019 15:16:23   #
photogeneralist Loc: Lopez Island Washington State
 
My response will exceed the range of the question. Read on , McDuff!!

Assume that your sensor is a brick patio of say 10 ft by 10 ft. We'll call that size a full frame patio. If it's made with big bricks (pixels) it takes less bricks to complete the 100 sq ft area. (OR more bricks if smaller bricks are used). If the bricks are different colors, a picture (mosaic) can be made by the bricklayer craftsman/artist. In this analogy the area of patio is the sensor size and the number of bricks is the megapixel "size". The patios come in only 3 or 4 different standardized sizes (Lets call those sizes Full frame, APSC crop, 1 inch and 4/3 ). The number of bricks determines how much detail it is possible to show regardless of what physical size the mosaic is. For given number of pixels, large pixels yield a larger sensor and smaller pixels mean a smaller sensor. Whether larger sensor size equates to greater sharpness (under ideal conditions) is open to debate. It just means that the pixels can be larger. Larger pixels can, but must not necessarily, have greater dynamic range and less noise. NOTE: all current sensors that I know of in major manufacturer's cameras seem to have better dynamic range and noise characteristics from larger pixels. It is up up to the camera buying public to choose:
1. sensor size
2. within the sensor size, the number of pixels that will yield, for their chosen type of photography, the best compromise between number of pixels and pixel size ie max obtainable detail vs dynamic range and noise.

My rant begins here!
Note that early digital cameras had few (3-6) megpixels yet produced some amazingly sharp photos. This brings into question how many pixels are really necessary and where the pixel overkill threshold lies. The current crop of multi mega mega megapixel cameras must of necessity have smaller pixels than they would if they had 1/2 the pixels in the same area of sensor. Because they have so many pixels , they can afford to throw many away through cropping without noticeably effecting the perception of sharpness in the photo. But, because they have such small pixels, they have diminished dynamic range and increased noise. Yet, sensors are arguably becoming capable of capturing more detail than the unaided human eye can perceive. Each new generation of sensor design seems to have better dynamic range and lower noise than does the previous generation at the same number of megapixels. (And at exponentially greater cost). Each of us must determine how much value we receive back from the additional dollars we must divert from our other needs or wants in order to get the latest high megapixel full frame camera body. Excess capabilities are a waste of dollars unless they are actual useful. How little can one spend in order to get a camera whose capabilities barely exceed their ability to utilize those capabilities? Where is the balance between cost and value best achieved?
BTW: I believe that dollars spent on high quality lenses return greater value than if spent on camera bodies. Good glass on a so-so consumer grade body will yield better crispness than mediocre glass on a super duper professional grade body. A good lens is a better long term investment since it will probably be used on several generations of camera bodies. Hope this helps. I also hope my (far too obvious) personal preferences have opened some new ways of looking at wise expenditures for you to consider. I also hope that you feel free to totally ignore my rant.

Reply
Mar 15, 2019 17:34:16   #
repleo Loc: Boston
 
photogeneralist wrote:
..... I also hope that you feel free to totally ignore my rant.....


Thank you.

Reply
 
 
Mar 15, 2019 23:45:00   #
Strodav Loc: Houston, Tx
 
For historical reasons, FF digital sensors are about 35.9 mm x 23.9 mm (1.4" x .94"), the same size as 35mm film. For years, the camera manufacturers tried to get the highest pixel count they could in that format, which can capture higher resolution (sharper) images. Right now the top end is about 46Mp or 8256 x 5504 pixels. There is a trade off between the size of each individual pixel in the sensor and noise. The larger the pixels the more light it captures resulting in less noise. So more (smaller) pixels means more noise, right? Not really because sensor technology is constantly improving and they have constantly reduced noise in while increasing pixel count. So how many pixels do you need? There are 2 answers: 1) as many as you can afford; 2) 24Mp or 6000 x 4000 pixels will do for most any type of photography. Of course you can get away with less, but why? Very good quality dslr 24Mp bodies are going form $600 to $1000. If you want to go much lower go with a smart phone. My iphone 8+ has 2 x 12Mp cameras in it and it won't be too long before they are at 16Mp.

Note: Don't put a low quality lens on a high quality sensor camera. You are wasting a good camera. Get as close to your subject as you can to avoid cropping as you are just throwing away resolution. Practice good technique to avoid camera shake and buy high quality glass with VR (won't need VR for wide angle lenses).

Reply
Mar 16, 2019 06:54:56   #
dave.m
 
I could ramble on about inverse square law and so on but perhaps working in reverse may help.

Imagine you have pixels the size of a grain of sand and 1 million grains. Arrange in a rectangular patter so each touches the other and you will get a rectangle of a certain size.

Instead of grains of sand we now use pixels that are the size of a pea so the rectangular pattern will be much larger.

In both cases we still have 1M pixels so the resolution is the same but the pea pixels are so much bigger so whats the benefit?

Using another analogy: hold a small cup in a shower and in a given time you will catch a certain volume of water. if the time is very short and the shower is sparse you may catch nothing at all. Hold a bucket for the same time and you will catch a much greater volume and even if sparse will probably get something.

Pixels are the same - in reasonable lighting small pixels will work just fine (think of sensors in cellphone cameras which are really small yet still capture good images). But if you want to capture images in very poor lighting or very low contrast, then small pixels struggle to get a clear information when a large pixel will. Lack of information can result in poor colour saturation and noise.

In summary as mentioned above, more pixels = more detail and hence potentially sharper images. Pixel size = more able to work in low/ adverse lighting and get a result.

Does it matter? Again as mentioned above it may or may not depending on your lighting preferences.

Reply
Mar 16, 2019 06:57:16   #
ELNikkor
 
Great responses for a relevant topic! I read them all, learned (and re-learned!) a lot. I believe there is a shifting balance, as technology advances, between the size of the pixels and their abilities to give us the best final images. (Not to mention those amazing PP programs for sharpening, removing noise, increasing resolution.) For now, I still feel I've met the balance for my photo world with my 24x36 24mp D750. Since I recently discovered the amazing zoom ability of my neighbor's 1/2.3 20mp sensor B700, specifically for getting sharp close-ups of little sparrows, I may get a similar camera for when bringing in distant small objects becomes more often important to me.

Reply
Mar 16, 2019 08:48:27   #
chrisg-optical Loc: New York, NY
 
Spectre wrote:
Help this old man. I’ve read here that sensor size is needed for better sharpness and also megapixel size is also. How do these two relate?


Sharpness and general image quality are related to so many factors, but assuming all else equal (same lens, exposure, lighting, subject, etc.) generally speaking larger sensors have better signal-to-noise performance and better dynamic range...but the differences are not that significant in good lighting conditions and unless you pixel peep. The larger sensors perform better because the pixel photo sites are larger, meaning less amplification is needed for weak signals (ie - low light)...think bigger eyeballs behind the window pane, not necessarily a bigger pane. But the bigger pane can fit either more eyeballs or larger eyeballs for the same # of eyeballs. (Ever notice nocturnal creatures have BIG eyeballs relative to their size!) Think about this - the 45 MP D850's crop sensor mode has almost the same pixel pitch as the D500 crop camera...crop mode 19.4MP vs 20.9 MP....the high MP FF cameras approach the density of the crop mode counterparts.

Reply
Page 1 of 3 next> last>>
If you want to reply, then register here. Registration is free and your account is created instantly, so you can post right away.
Main Photography Discussion
UglyHedgehog.com - Forum
Copyright 2011-2024 Ugly Hedgehog, Inc.