Go Back   Steve's Digicams Forums > Digicam Help > General Discussion

Reply
 
Thread Tools Search this Thread
Old Sep 30, 2006, 3:41 AM   #1
Senior Member
 
BenjaminXYZ's Avatar
 
Join Date: Jul 2006
Posts: 788
Default

I have learned the fact that largerphoto-detectors in a image sensor, is going to be more sensitive than smaller ones. (This statement I assume, is going to be true)[Correct me if I'm wrong]

In case the claim above is incorrect; it might pose some complications in the following related questions.

Speaking in terms of theory, if we have a very large piece of image sensor with very tiny photo-detectors; and another much smaller image sensor with much larger photo-detectors. Which of the two image sensorsis going to bemore sensitive to lightat the same ISO level?

Toillustrate "more sensitive", imagine both image sensors are being exposed to the same intensity of light at the same ISO level, which onewould then use a faster shutter speedto expose the scene correctly? (The one that is more sensitive, will be using a faster shutter speed; for the same exposure)

The next question;

Is a dSLR in absolute terms, more sensitive to light thana ultra compact camera?

Most of the ultra compact camerashave theF/2.8 aperture at the wide end over their tiny image sensors. If a dSLRcamera hasanF/2.8 aperture at the wide end as well, will it be more sensitive to light than an ultra compact at the same ISO level?

Theabove exampleis assuming that both types of camera are aiming at the same source of light, setat the same ISO level.Onceagain, the more sensitive one will be using a faster shutter speed for the same exposure. (So I hope you understand my questions about sensitivity)








BenjaminXYZ is offline   Reply With Quote
Sponsored Links
Old Sep 30, 2006, 6:18 AM   #2
Moderator
 
Nagasaki's Avatar
 
Join Date: Jan 2004
Posts: 897
Default

If two cameras are set to the same ISO then they will require the same shutter/aperture combination regardless of sensitivity or size. This is what ISO means.

What you will find is that the small sensors on a compact camera will generally have a lower ISO level than a DSLR camera. For example my Nikon D70 has a lowest ISO setting of 200, the new D80 has a setting of ISO 100 whereas many compacts have a low end setting of ISO 50. At the other end DSLRs with their larger sensors have ISOs of 1600 or 3200 where many compacts have a maximum ISO of 400.

The reason is that the smaller photsites on smaller sensors are less sensitive to light and so have more gain applied to acheive higher ISO setting. Upping the ISO introduces noise into the image with small sensors beyond about ISO 400 the noise is unacceptable so higher settings are not used. With a DSLR you can shoot at higher ISOs and still maintain a reasonable image. Again using my D70 as an example at ISO 400 the image is still very good and perfectly usable, with my compact ISO 400 is very noisy and the image is not really good enough for a decent print.

Ken




Nagasaki is offline   Reply With Quote
Old Sep 30, 2006, 7:40 AM   #3
Senior Member
 
BenjaminXYZ's Avatar
 
Join Date: Jul 2006
Posts: 788
Default

Alright, sothe tiny image sensor in a ultra compact camera have to crank up it's sensitivity levels to match a dSLR's image sensor lower ISO levels?

For example, the tiny sensor in a P & S need to push (gain) it's photo-sites up in order to match a dSLR's ISO 100 am I right?

Anyway, does anyone know which of the following image sensors would be more sensitive to light rays?

A very large image sensor having photo-sites as tiny as that of an ultra compact camera'; or a much smaller image sensor having large dSLR sensor photo-detector like sizes?

I am confused as to whether a larger surface area with tiny photo-sites, determines a better sensitivity (reception to light rays); or larger photo-detectors on a much smaller surface area...

Imagine your whole monitor in front of you now is an image sensor with very tiny p&s photo-detectors, and there is another image sensor the size of 35 mm film having much larger pixel pitch than the first one; which one do you think will be the more sensitive sensor???




BenjaminXYZ is offline   Reply With Quote
Old Sep 30, 2006, 8:08 AM   #4
Moderator
 
Nagasaki's Avatar
 
Join Date: Jan 2004
Posts: 897
Default

Your explanation is close enough. I'm not sure if it's 100% accuate because I don't know enough about how the camera sensors work at that level.

Given the same physical sensor size the individual photosites on a 2Mp camera will be more sensitive than the photosites on a 6Mp camera. This assumes similar technologies as sensors are improving with development.

So a larger sensor with smallphotosites will be less sensitive than a small sensor with large photosites, assuming that the differences are enough that the photosites on the small sensor are physically larger than the photosites on the large sensorand the technology is the same.

Remember though taht when you look at Mp it's an area so you need to look at the area of the sensor.

For example a compact may habve a sensor size of 4x3 mm which is an area of 12 square mm. A NikonDSLR has a sensor of23.7 x 15.6 mmwhich is an area of 369.72 square mm. You can fit an a lot more photosites into that area before you begin to approach the small sensor size of a compact. That's an extreme example but a large sensor in a compact is 8.8 x 6.6 mm.


Nagasaki is offline   Reply With Quote
Old Sep 30, 2006, 2:28 PM   #5
Senior Member
 
Join Date: Aug 2006
Posts: 1,066
Default

its an interesting subject i think
generally large sensors have a greater area to place photons on
and there are some features of note
lenses are or should be scaled appropriately to the sensor they service
large sensor cameras have greater depth of field, and generally less noise

you can have a largish sensor crowded with photon sites
or a small sensor, with relatively less per the area
my project alludes to this potential

http://www.stevesforums.com/forums/v...amp;forum_id=2

Add to the mix, there are CMOS and CCD sensors too. CMOS are limited to full frame or near full frame dSLRs while smaller cameras are CCD. The percentage of pixels devoted to collecting light per the area of the sensor is called the 'fill factor'. CCD cameras have 100% fill factor while CMOS are much less.

Low fill factors are compensated by adding a sheath of glass to provide some lenticular properties, virtually a lens above each photosite.

On noise, CMOS have inherrantly more noise than CCD's. This is why full frame cameras with CMOS are unlikely to get all that much bigger in megapixels (Mp). While noise is a much discussed aspect of CCD small sensor cameras, it needs to be remembered that the density of photons, and hence the Mp per the smaller area of the sensor available, would be impossible with CMOS technology.

For small sensor camera producers, improving the technology of CCD's is vital to conquering more of the traditional dSLR market. With the Mp race beggining to settle down some, it is envisaged that the R&D will be moved into this area.

So we come to pixels and Mp and what the limits are. For TV 525 x 320 = 168,000, for asa 25 Kodachrome (used to be the publishing standard) about 25,000,000, for the human eye, 11,000 x 11,000 = 120,000,000. When we get to 120 Mp we match the resolution of the eye, and the pixelation will be unrecognisable.

Riley
Rriley is offline   Reply With Quote
Old Sep 30, 2006, 8:08 PM   #6
Senior Member
 
Bob Nichol's Avatar
 
Join Date: Apr 2003
Location: Eastern Ontario Canada
Posts: 822
Default

Large sensor sites versus small sensor sites can be compared to listening to a strong local radio station compared to a distant weak station. The local station comes through loud and clear and the distant station is hard to hear through the static. This is an example of signal to noise ratio. The smaller sites are like the distant station, the signal is buried in the noise. The larger sites are like the local station where the noise is lost in the signal.

Big sites are always better than small sites no matter how many there are!
Bob Nichol is offline   Reply With Quote
Old Sep 30, 2006, 9:14 PM   #7
Senior Member
 
tmoreau's Avatar
 
Join Date: Dec 2005
Posts: 477
Default

f-stops and iso sensetivity ratings are standardized measurements, if a lens lets in more light or a film emulsion is more sensetive, its rated higher.

In order to get a higher sensetivity to light, all cameras amplify the signal coming from the sensor. If you have a stronger signal, you amplify less. The level of amplification varies based on your set iso speed and how much signal you have to start with, therefore a small sensor requires more amplification as does a higher iso. In either case, more amplification leads invariably to more noise.

Physically larger photosites recieve more light (meaning more signal, meaning more sensetivity in your terms), its very very very simple to understand that. Put a drinking glass and a frying pan out in the rain, each recieves rainwater at the same rate but the larger one accumulates more. Simple. Changing the size of the pan does not change how much it rains.

Life aint simple though, and things like circuit architecture and microlens design amoung other things can have a big effect on the realized efficiency of a sensor. In the end, you can only make so many generalizations before you have to compare product X to product Y on actual performance. Look at the relatively poor performance of turn of the century DSLR's compared to todays breed.

You can consider that the background electrical noise present in any circuit must be low enough in relation to the signal that it does not significantly modify the signal. In the case of high-amplification tiny sensors, you dont have enough signal to marginalize the standard electrical noise, thus you get noise in your image. With bigger pixels, you have a stronger signal. Pressing too many circuits (pixels) into too small a space is likely to compound your noise problems too, so the rope is burning from both ends.

Its been proven that in todays technology you need a physical space of 6 microns per photosite to adequatly marginalize noise, and 8 microns is noticeably better. Thats the pixel size range of current DSLR's, compact cameras go down to the 3 micron range and maybe smaller. See my reply to your "holy grail" post, start with how big you need to make your photosites in order to get an acceptable signal to noise ratio (SNR), then add up as many pixels as you need in your prints, and magically you will arrive at a measurement of how big a sensor you need to accomplish this.

If you really get into all the specifics, you could probably devote your entire life to the issue.
tmoreau is offline   Reply With Quote
Old Oct 1, 2006, 1:03 AM   #8
Senior Member
 
BenjaminXYZ's Avatar
 
Join Date: Jul 2006
Posts: 788
Default

Alright, that means the smaller photo-sites in a P&S image sensor have to always work harder in order to match each larger photo-sites' ISO level in a dSLR image sensor. (It might be like that) **Therefore, by amplifying the signals more; there exist more noise.**

I can imagine that a typically small P&S image sensor, can only go somuch toward ISO 400; before reaching it's limits.

I still cannot imagine what will happen, if we can have an image sensor the size of the Hubble Space Telescope's primary mirror; with very tiny p & sphoto-sites! Comparing it to another 35 mm dSLR, with much larger photo-sites in it's F.F. sensor; which of the two image sensors will be more sensitive to light rays? :idea:

Would I be confident enough to confirm to NASA (If it happens), that my F.F. dSLR is going to be more sensitive to light rays than your Hubble Space Telescope (With P&S like photo-detectors; but on a much much larger image sensor)? :G




BenjaminXYZ is offline   Reply With Quote
Old Oct 1, 2006, 1:41 AM   #9
Senior Member
 
Join Date: Aug 2006
Posts: 1,066
Default

BenjaminXYZ wrote:
Quote:
Alright, that means the smaller photo-sites in a P&S image sensor have to always work harder in order to match each larger photo-sites' ISO level in a dSLR image sensor. (It might be like that) **Therefore, by amplifying the signals more; there exist more noise.**

I can imagine that a typically small P&S image sensor, can only go somuch toward ISO 400; before reaching it's limits.

I still cannot imagine what will happen, if we can have an image sensor the size of the Hubble Space Telescope's primary mirror; with very tiny p & sphoto-sites! Comparing it to another 35 mm dSLR, with much larger photo-sites in it's F.F. sensor; which of the two image sensors will be more sensitive to light rays? :idea:

Would I be confident enough to confirm to NASA (If it happens), that my F.F. dSLR is going to be more sensitive to light rays than your Hubble Space Telescope (With P&S like photo-detectors; but on a much much larger image sensor)? :G
ummm, well not entirely
there are CMOS and CCD cameras. Canon dSLRs are CMOS, pretty much all P&S are CCD. CCD have at their heart 100% or near 'fill factor', I cant find out the ratio for CMOS Canons but have been fielded 75% fill factor. This means that not all the sensor real estate is being used to resolve light to image.

Riley

Rriley is offline   Reply With Quote
Old Oct 1, 2006, 11:13 AM   #10
Senior Member
 
BenjaminXYZ's Avatar
 
Join Date: Jul 2006
Posts: 788
Default

Noticed that the smaller photo-sites/image sensorsalways have problems with dynamic range, and noise. Always blown out highlights and high ISO noise problems...

Quote:

Specific image quality issues
Our biggest complaint about the FZ30 was noise (and noise reduction), and there's no getting round the fact that it is also the FZ50's achilles' heel too. The Venus III noise reduction is undoubtedly effective at reducing measurable noise, but it does so in a way that produces a soft, smeared appearance even at ISO 100 if you view actual pixels on-screen. The excessive chroma noise reduction is particularly evident with low contrast detail such as distant foliage or hair.

That said, with 10 million pixels to play with you're unlikely to see it in standard sized prints or scaled down to fit a typical computer screen or TV. The default contrast is also a little too high, which combined with the limited dynamic range means there is the usual tendency to either clip highlights or to fill in shadows. This means accurate exposure is critical, and the FZ50 is pretty good, although in very bright contrasty conditions the pattern metering seems too easy to fool, and it's worth checking the histogram or bracketing.

I'd advise leaving the noise reduction, sharpness and contrast settings to low if you want the cleanest images and don't mind doing a little post-processing. For the ultimate control switch to raw mode. And of course, remember that 10 million pixels doesn't mean poster sized prints when the sensor is this small; the FZ50 is a camera that demands you understand its limitations if you want to avoid disappointment.

Finally, we found very little of the fringing we've come to expect from long zooms and small sensors - even in areas of extreme overexposure the only problem was a little purple fringing associated with blooming. There is a small (and barely noticeable) amount of chromatic aberration visible if you look very very closely, but it's not significant. It should be noted that the absence of fringes appears to be a result of the Venus engine processing; if you shoot raw you'll see more.
From http://www.dpreview.com

Compared to:

Quote:

Overall Image Quality / Specifics

Overall the D70 performs well, images sharper than the D100 (thanks to a slightly weaker anti alias filter and improved image processing) good tonal balance and a neutral color balance. Noise levels are low and the camera is usable throughout its full range of ISO's, a bonus is that the D70's noise appears to be more monochromatic with less noticeable 'color noise'. We had no major issues with image quality apart perhaps from the occasional brush with moiré / maze artifacts (Bayer interpolation issues, see below). Overall however I liked the look of the D70's images and am impressed with the improvements Nikon has made, Kudos. Thanks to the wider range of selectable image parameters you can get the picture look you like straight out of the camera, in several instances I found the punchy 'Vivid' mode produced some very nice results.



Noticed the stark difference.

We can never fault lower mega-pixel counts and larger photo-diodes/photo-sites. :-)

I give you a hint;

You can go to dpreview's site andjust pick a camera there with alow mega-pixel count; and read a review about it'snoise characteristics. I bet 99% of the time you will be expectingpositive results!

If you also observe for largeimage sensor size,you canexpect a much more better result!

For example, this>>>http://www.dpreview.com/reviews/nikond1h/

That was my first random pick last time, with the choice of a larger APS-C size image sensor in mind!









BenjaminXYZ is offline   Reply With Quote
 
Reply


Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off



All times are GMT -5. The time now is 1:39 PM.