September 22nd, 2012, 12:17 PM
Digital Camera Database – feedback wanted
my name is Gregor and I'm the creator of Digital Camera Database (digicamdb.com). It's a new site focused on sensor info and comparison (sensor size, pixel pitch, pixel density, etc.). It offers sensor information for more than 3000 digital cameras.
I would greatly appreciate any feedback you might have. Do you find the information useful? Please tell me what you think?
I would also like to use this opportunity to pose a question to fellow Americans. You see, I'm from Slovenia, EU and I'm all metric. I would like to offer both metric and inch sizes, but I'm a bit confused as to what would make sense.
My question is - which numbers would you like to see in inches? Do micro inches make sense?
Thank you for reading this and special thanks to Amin who was kind enough to allow me to ask you for your opinions.
September 22nd, 2012, 12:37 PM
You've created a very nice website. It functions very smoothly. As a dumb American myself who almost NEVER uses the metric system I don't think you need to use anything other than metric measurements. They are the standard for camera measurements.
September 22nd, 2012, 01:52 PM
Thanks Luke, that's good news. It'll definitely save me some work.
Originally Posted by Luke
September 22nd, 2012, 06:20 PM
I like this site, am bookmarking it. Well done!
September 22nd, 2012, 08:03 PM
Here're my comments, some of them (or maybe all of them) may be anal but I'm just trying to be as critical and precise as possible, that's just the mode I sometimes kick in to when facts and technicalities are concerned
Minor pet peave: maybe you can call the smaller sensors (for instance) 1/1.7" type, rather than 1/1.7", because that way of measuring sensors is no longer really relevant and even when it was, it was an indirect way of measuring the sensor (for instance 1/1.7" = roughly 14.9 mm which won't really tell you anything about the actual sensor size).
Also, the 4/3 sensors are listed as 4/3", but the inch sign shouldn't be there, it's just about the aspect ratio...
In the comparison tool, where you explain the meaning of the different numbers:
I think the 2nd sentence isn't really necessary, since it's the sensor's overall size that determines how much light is captured for the entire photo, not the pixel size (it's just that a larger sensor with the same amount of pixels will also have larger pixels).
Bigger sensors are more effective, because they have more surface area to capture light. To put it another way - bigger sensors have bigger pixels and bigger pixels can capture more photons during exposure. This results in less noise and greater dynamic range.
Also, where you say:
I think it'd be good to add a clarification saying that this is true on a per-pixel level (at a given % of magnification). If you view the image as a whole, for any given absolute size (say 30x20cm, or 1200x800pixels), the individual pixel surface area isn't relevant, but the overall sensor surface area is (at least where noise is concerned; overall dynamic range might improve slightly with cameras using larger pixels).
Pixel or photosite surface area has a direct influence on camera's sensitivity. The larger it is the more light it can capture and the more information can be recorded.
Cameras with larger photosites typically produce better results.
And here basically the same thing:
Lower pixel density is good for per-pixel quality, but if you print the image at a given size or view it on a given size monitor, the pixel density isn't really important (again where noise is concerned, with dynamic range possibly experiencing a benefit from lower pixel density).
Pixel density tells you how many million pixels fit or would fit in one square cm of the sensor.
The smaller the number, the better, because it means that pixels are larger and aren't so tightly cramped together.
September 22nd, 2012, 08:55 PM
I agree that overall the site is great and very helpful. However, I'd go a step further than barteej and say that both of these connections are weak:
Bigger sensors... greater dynamic range.
Sensor size is only a very weak determinant of dynamic range. Amongst current sensors, read noise has a greater impact on DR than sensor size does, which is why a small Sony sensor like the RX100 has more DR than a 35mm format Canon 5D III.
Cameras with larger photosites typically produce better results.
I don't think this is right either. Cameras like the Nikon D800 and D600 as well as the Sony NEX-7 and RX100 all produce results which are at least as good as their larger-pixeled competitors. They may give up a tiny bit of ground in high ISO low light performance (eg D800 vs D4) but they are less susceptible to aliasing, produce more detailed results, and as for overall image quality (by most criteria) seem to lead the pack.
Amongst current cameras, I think the only clear cut advantage for larger pixels is smaller file sizes. It's fair to say that larger sensors tend to have better image quality with the caveat that some sensor technologies can more than make up the difference in certain parameters of image quality (eg, low read noise and DR).
September 22nd, 2012, 09:30 PM
4/3" may be correct for 4/3 and m4/3. The sensor was sized around the old 4/3 tube, with the aspect ratio being coincidental. The Canon G1X is slightly bigger and is classified as having a 1 1/2" sensor, which is also slightly bigger than 4/3".
Last edited by Luckypenguin; September 22nd, 2012 at 09:32 PM.
Nic (Canonite, Olympian, Panasonian, Samsunite) ~flickr~
September 23rd, 2012, 06:15 AM
Thank you for your comments. I have changed 4/3" into Four Thirds. I do plan to rewrite the explanations in the next days and I will certainly take your thoughts into account.
September 23rd, 2012, 01:22 PM
So related to this site, in comparing the latest Rebel T4i to the EOS M.. I get these results. Both cameras are on par with each other. Canon EOS Rebel T4i vs. Canon EOS M
AND when I compare my EOS 45D to the Rebel T4i.. the results are basically telling me my old 450D is better than the T4i Canon EOS 450D vs. Canon EOS Rebel T4i
THUS my 450D compared to the EOS M [what I was considering as a carry around that can also utilize my current lenses] Canon EOS 450D vs. Canon EOS M
Is this lesser IQ because of MP, the 18megapixels vs the 12mp? Spreading the pixels out over a larger image is lessening quality to put it simply?
I know it's not related to site function [or maybe it is site error], but is this correct?
Also related to this site gregorb, on some camera bodies you have lenses listed such as the 450D Canon has a 38-114mm lens-- it seems like that info since lenses can be swapped out, shouldn't even be part of the comparisons.
September 23rd, 2012, 02:10 PM
Kristen: overall sensor size determines how much light falls on the sensor, with sensor efficiency determining how much of the light is actually captured. If you view the image at a given size, it's the overall amount of captured light that determines the quality, not the per-pixel amount.
So if you have two (equally efficient) sensors of different sizes, and print the images at the same size, the image from the larger sensor will be better, regardless of the number of pixels (obviously assuming that both have enough resolution for the viewing size).
If you have two sensors with different numbers of overall pixels, and you view them at a given magnification (for instance 100%) the sensor with the lowest pixel density (largest per-pixel surface area) will have the best per-pixel image quality, regardless of the sensor size.
There is a small issue that more pixels means more electronics/wiring, meaning more noise from the electronics and lower dynamic range (DR is usually measured as the difference between the brightest that a sensor can measure without clipping (which is brighter with a larger pixel), and the darkest that the sensor can measure without signal the noise from the electronics being stronger than the incoming light (which is determined by the state of technology, and is mostly independent of pixel size).
But as Amin said, higher pixel count also has its advantages. So in most cases, for normal viewing (that is, at a fixed overall size instead of a fixed magnification; for instance your monitor's maximum resolution, or 30*20cm print) more pixels doesn't significantly degrade IQ, and if the viewing size is large enough, it can offer significant advantages.
If you have the time to read it, this is an excellent explanation of the issue:
DxOMark Sensor For Benchmarking Cameras
This site uses affiliate programs and referral links for monetization.