gabimaster wrote:
WE have to wait the first TESTS!!! IF the reviews are in favor of 50D,I'LL SELL MY 2 BODIES OF 30D and buy 2 of 50D.(I ALWAIS TAKE 2-BECAUSE YOU NEVER KNOW WHAT CAN HAPEND)
You mean actually wait for some honest-to-god real performance data before deciding to bash a newly announced product to death? Hmmmm... how quaint.
ChrisDM wrote:
So is using f18 on my 1Ds3 "useless" because of diffraction? Of course not, as this example proves. Yes, all lenses (not sensors) suffer from diffraction at narrow apertures. Does this make them useless at narrow apertures. No. We shoot at the aperture required to produce the depth of field necessary to get the shot... Does this mean that we should avoid sensors with high resolving power so we don't see the flaws inherent in the designs of the lenses we choose? Of course not.
Chris M
www.imagineimagery.com
I agree, I shoot hummers from f/18 - f/20 on my 1Ds3 with a multi-flash setup for DOF and the pics are sharp. The "useless past f/16" statement wins post of the month.
Basically, there is a small amount of random 'signal' reported by each sensor pixel. Any light falling on that pixel will be added to the random signal to give you the signal recorded by the camera.
The more light that falls on to one of the pixels, the smaller the fraction of the overall signal that is noise, therefore the less noticable the random variation is.
If you make the capturing microlens bigger, you reduce the amount of light wasted in the gaps between them. In other words, you get more bang for your buck - at any given light level more of the incident light goes into the signal recorded by the camera, making the noise less significant.
In essence there is no actual reduction in noise from having larger microlenses (although they may have done this in some other way), but rather a better signal to noise ratio.
You can also reduce noise by improving that background noise...which will happen as the CMOS technology matures, and I'm sure it's part of this sensor.
This sensor sounds amazing, and I for one am very impressed by this release. Too bad I can't buy a new camera right now.
Good picture ... but also suggests that Canon won't be able to reduce SNR any more using this approach since the microlenses are now gapless ... but maybe they can "increase" the size of the photodiode's by having them go gapless also (?)
Alek Komarnits wrote:
Good picture ... but also suggests that Canon won't be able to reduce SNR any more using this approach since the microlenses are now gapless ... but maybe they can "increase" the size of the photodiode's by having them go gapless also (?)
Yes, what my optimistic mind sees in this diagram is more and/or larger photosites per the same area. That would of course potentially allow us to have our cake and eat it too (more resolution/less noise)... But we can only speculate until some testers get their hands on it and publish findings. Until they we'll just continue to blow smoke...
Someone posted that Rob Galbraith said the new 50D is 1.5 stops cleaner than the 5D (or 40D... not sure, haven't read it myself).
I remember a famous photographer/web host (I forget his name) who said the 40D was cleaner than the 5D when the 40D was first introduced. I've owned both cameras and can confirm that is false.
It would be amazing if the 50D is 1.5 stops cleaner than the 5D, but I will not believe it until a few reputable FM users gets a crack at it and confirm their findings.
It is called progress. Where the "sweet spot" of a sensor vs noise vs pixel size is a moving target as advances in the design and manufactuer of the "chip" advance, coupled with advancing technology of the controlling chips (Digic iv) and the software algorithms being used.
The only thing that is hard and fast about camera, and specifically sensor design, is that nothing is hard and fast.
Only results can show whether the system with all the advancements and compromises were on target, over the top, or short thereof.
Jman13 wrote:
Just think how clean the 7.1 MP sRAW files at ISO 6400 will be, with 2 to 1 pixel binning.
Can someone explain this a little bit. How does 2 to 1 pixel binning compare to just a straight 7.1MP sensor? Is the only difference between sRAW and normal raw the file size... nothing is compressed? So the 50D will be lowest in noise when shooting with sRAW?
For those who just want a low-megapixel camera with excellent high ISO ability, is shooting sRAW equivalent to using such a camera?
Stunnaz wrote:
Someone posted that Rob Galbraith said the new 50D is 1.5 stops cleaner than the 5D (or 40D... not sure, haven't read it myself).
I remember a famous photographer/web host (I forget his name) who said the 40D was cleaner than the 5D when the 40D was first introduced. I've owned both cameras and can confirm that is false.
It would be amazing if the 50D is 1.5 stops cleaner than the 5D, but I will not believe it until a few reputable FM users gets a crack at it and confirm their findings.
Ditto. My 40D was about a half stop noisier than my 5D. In other words, the 40D's 1600 was about equal to the 5D's 1250... If the 50D is equal to or better than the 5D's high ISO performance, I'll be very impressed. Hell, I'll be happy if it matches the 40D at high ISO, considering the big jump in resolution.
Kevin Sherman wrote:
Chris, I am slightly confused as to what diffraction is. Where is the evidence of diffraction in the image you posted?
And I for one am now super stoked about the 50D. Been considering a used 1DMII for sports, but this might really be the camera I am looking for...
Diffraction is just a term for the softness that a lens gets when stopped down to narrow apertures. Most people talk about how soft a lens is "wide open", or at wide apertures, and talk about it's peak sharpness at a certain aperture (typically two stops past wide open, around f5.6 to f8). Then sharpness decreases as the lens is stopped down further, called diffraction.
The point I was making earlier was twofold. First, diffraction has little to nothing to do with the sensor. And more importantly, unfortunately too many "gear oriented" photographers get too concerned with this "maximum sharpness", and foolishly choose apertures that will get them the sharpest image within their depth of field, rather than choosing the aperture that would get them the required depth of field in the first place. An in-focus diffracted area is always going to be infinitely sharper than an out-of-focus area, no matter what aperture it is shot at.
More pixels do mean more noise at hi ISO-assuming equal signal to noise from identical pixel construction. But Canon claims a bigger "light bucket" in comparison to their 40D. Since area varies with the square (and ISO is linear) the elimination of the small gap is significant because 1.4x more light bucket diameter=1 extra stop of light (signal).
With the addition of better on-chip processing with Digic IV (electronic) noise can drop further. With more efficient noise processing (hopefully less aggressive than the D300) Canon can get 1 1/2 stop advantage over their 40D.
A 1.6x crop sensor is only about 42% of the area of a FF sensor. That leaves a small area of glass (whether S or L over coverage) to focus the image. We know that the 1Ds MK III demands the best resolving lenses to take advantage of 21MP. So, it seems, that the 50D will, further, strain the limits of resolution with it's 15Mp sensor (an imputed 30Mp+ for equal size pixels on a FF sensor).
And, yes, that means that further improvements in low noise hi-ISO and IQ have reached the limit for the 1.6x (resolution limited) sensor and edge to edge light bucket pixel. But the next Mk IVs can benefit from bigger light buckets. There's the promise of equal resolution for the 21 Mp FF sensor area but better hi ISO performance while a MK IV could have both better resolution (using a 1.275x sensor) and better hi-ISO performance.
Kevin Sherman wrote:
Chris, I am slightly confused as to what diffraction is. Where is the evidence of diffraction in the image you posted?
And I for one am now super stoked about the 50D. Been considering a used 1DMII for sports, but this might really be the camera I am looking for...
Some after-hours with not-too-many adult beverages reading. In brief, anytime light passes close to an object, it is slightly deflected by the object. In photography, it mostly refers to the diaphragm blades that make up the aperture of the lens. As one stops a lens down (making the aperture smaller), some of the optical flaws inherent in the lens become minimized. Wide open, diffraction is there, but is not really a factor. As one stops the lens down the effect becomes greater. The "sweet spot" of the lens is where a lens is stopped down sufficiently to mask the lens problems, but before diffraction becomes a noticable problem.
please no more newtonian physics! it has everything to do with ieverything, so the arguing could go on for... well it's been going on since the 17th century already.
Stunnaz wrote:
Someone posted that Rob Galbraith said the new 50D is 1.5 stops cleaner than the 5D (or 40D... not sure, haven't read it myself).
I remember a famous photographer/web host (I forget his name) who said the 40D was cleaner than the 5D when the 40D was first introduced. I've owned both cameras and can confirm that is false.
It would be amazing if the 50D is 1.5 stops cleaner than the 5D, but I will not believe it until a few reputable FM users gets a crack at it and confirm their findings.
that is actually not accurate: galbraith's "comments" regarding the 1.5 stop difference are actually attributed to chuck westfall from canon's briefing notes.
post of the month.