Forgot Password
Pentax Camera Forums Home
 

Closed Thread
Show Printable Version Search this Thread
09-01-2009, 09:05 PM   #31
Senior Member




Join Date: Jun 2008
Location: Atlanta, Georgia
Photos: Gallery
Posts: 182
Original Poster
QuoteOriginally posted by dosdan Quote
According to Equivalence the difference in sensitivity between a APS-C & a FF is about 1.33 stops, everything else being equal.

Dan.
I just read that article, and it actually states the same thing as I said
"For the same perspective, AOV, DOF, shutter speed, and apparent exposure, all sensors, regardless of size receive the same total amount of light. Thus, under these conditions, for the same efficiency of sensor, all formats will have the same amount of total apparent image noise" (I added bold)

Best regards,
Haakan

09-01-2009, 11:22 PM   #32
Veteran Member




Join Date: Jun 2008
Location: Borås, Sweden
Photos: Gallery
Posts: 3,165
So how much impact does the more compact/dense circuitry of an APS-C or 4/3 sensor have on noise compared to FF (assuming same pixel count)?

(edit: after getting a good night's sleep I reread this thread and it's a pretty interesting topic actually. )
09-02-2009, 01:32 AM   #33
Pentaxian
falconeye's Avatar

Join Date: Jan 2008
Location: Munich, Alps, Germany
Photos: Gallery
Posts: 6,862
QuoteOriginally posted by pingflood Quote
So how much impact does the more compact/dense circuitry of an APS-C or 4/3 sensor have on noise compared to FF (assuming same pixel count)?
This is another famous debate

I prefer to discuss this topic independent of sensor size because the size of a pixel and the size of the sensor are not linked to each other. It is just to simplify the discussion as in this thread that one keeps constant the pixel count constant.

If you take the D300, you see two FF derivatives: D3 (same pixel count) and D3X (same pixel pitch).

So, the question really is how much impact does the pixel pitch have on noise, independent of sensor size?


Here, Haakan's nice reinterpretation of DxO means that the impact in none.


In general, to compare noise, you must first normalize it to a given print size (factor out the different number of pixels resulting from the different pixel pitch). E.g., in order to normalize to 10 MPixels, you either resize all samples to this size, or multiply the ISO of measurements by SQRT(pixels/10mp) in ISO (which is a simple shift in ISO on a logarithmic ISO scale). I.e., You compare a 14.6mp/ISO800 pixel measurement with 10mp/ISO1000.

This normalization removes the first order effect of pixel pitch, i.e., the scale dependence of noise.

The second order effects are more subtle. On the one hand, you have more circuitry overhead per pixel (a larger shaded area), smaller circuitry (more electronic noise), etc. On the other hand, you have more pixels to average out the noise coming from, e.g., dark current or gain amplifiers. And microlenses reduce the effect of shading.

The second order effects are such that there exists a sweet spot (optimum pixel size) resulting in minimum normalized noise, which depends on the silicon process, chip and microlens design, and sensor temperature. Experiments with RAW files from P&S cameras suggest that this sweet spot is at somewhat smaller pixel sizes than in current use in dSLRs.

Normally, large pixel pitch FF sensors are expensive enough to have carefully designed gain amplifiers and ADC circuitry not adding to normalized noise.

If you compare D3 and D3X in DxO mark's SNR plot (the "Print" does the scale normalization and must be looked at), then you see almost identical results (less than 1dB difference).


So, to summarize: The pixel pitch can be ignored in this discussion. Fortunately.


--
Personally, I like the resolution advantage coming from high pixel pitch FF (like D3X). But it is expensive from Nikon (somewhat more affordable with 5D2 and esp. A850). Where high pixel pitch FF currently means 6µm vs. 5µm in K-7 and 4.3µm in 7D. And a high pixel pitch FF makes video expensive to implement if you don't want a noise penalty in video.
09-02-2009, 02:23 AM   #34
Veteran Member




Join Date: Jun 2008
Location: Borås, Sweden
Photos: Gallery
Posts: 3,165
So Falk, basically you are saying that the people clamoring for lower resolution "lower noise" sensors are chasing a mirage? There've been some very heated debates on FM over this, with folks firmly entrenched in either camp.

One thing though... if you have a noise floor in the circuitry, once you get to a certain pixel pitch won't simply the low # of photons hitting each photosite cause the signal to entirely disappear in the noise? If you disregard the photon being a singular entity and simply look at it as a distribution of light then yeah, I am with you the entire way; however since we are ultimately down to individual photons then there has to be a breaking point somewhere. Right?

E.g. if the photosites receive say 200 photons each (no clue what the real numbers are) and the circuitry noise is equivalent to 20 photons, things are fine, but increase the pixel density significantly and each photosite might get 20 or 30 photons, and at that point, how do you sort it out from the noise? Or is the noise "orderly" enough that you can effectively get past it?

09-02-2009, 03:40 AM   #35
Pentaxian
falconeye's Avatar

Join Date: Jan 2008
Location: Munich, Alps, Germany
Photos: Gallery
Posts: 6,862
QuoteOriginally posted by pingflood Quote
the circuitry noise is equivalent to 20 photons, things are fine, but increase the pixel density significantly and each photosite might get 20 or 30 photons, and at that point, how do you sort it out from the noise?
You ask about shot noise vs. read-out noise.

You assume that read-out noise is independent from the pixel size. It is not. Larger pixels have larger dark currents (because they have larger well capacities).

If you have 10 cells measuring 20 photons each and having the equivalent of 20 photons "circuitry noise" (50% read-out noise), then you have a mathematically equivalent situation to 1 cell ten times as big measuring 200 photons and having the equivalent of 63 photons "circuitry noise". So, would a larger real cell have less or more than 63 photons "circuitry noise"? You don't know. It depends on which side of the sweet spot you are.



And yes, 99.9% of people fighting over noise vs. pixel pitch had something better to do than listen at this statistics lessons in high school ... The remaining 0.1% are in the know and seriously care about the remaining 1dB difference. Like speaker wires in high end audio ...

Last edited by falconeye; 09-02-2009 at 03:52 AM.
09-02-2009, 04:05 AM   #36
Senior Member




Join Date: Jun 2008
Location: Atlanta, Georgia
Photos: Gallery
Posts: 182
Original Poster
QuoteOriginally posted by falconeye Quote

And yes, 99.9% of people fighting over noise vs. pixel pitch had something better to do than listen at this statistics lessons in high school ... The remaining 0.1% are in the know and seriously care about the remaining 1dB difference. Like speaker wires in high end audio ...
When I read dpreview, I often wonder in which category their reviewers fall. The quote below is taken from their conslusion in the Sony A900 review:
"especially when you factor in the rather disappointing high ISO performance that is the inevitable consequence of such a high pixel density" (I added bold)

Best regards,
Haakan
09-02-2009, 04:19 AM   #37
Veteran Member




Join Date: Jun 2008
Location: Borås, Sweden
Photos: Gallery
Posts: 3,165
QuoteOriginally posted by Haakan Quote
When I read dpreview, I often wonder in which category their reviewers fall. The quote below is taken from their conslusion in the Sony A900 review:
"especially when you factor in the rather disappointing high ISO performance that is the inevitable consequence of such a high pixel density" (I added bold)

Best regards,
Haakan
It is rather funny when people start ranting about how 20-25MP on full frame is too high density, then turn around and praise 10MP APS-C sensors.
09-02-2009, 04:36 AM   #38
Senior Member




Join Date: Jun 2008
Location: Atlanta, Georgia
Photos: Gallery
Posts: 182
Original Poster
QuoteOriginally posted by pingflood Quote
It is rather funny when people start ranting about how 20-25MP on full frame is too high density, then turn around and praise 10MP APS-C sensors.
I agree, it seems to lack somewhat in consistency.

Best regards,
Håkan

09-02-2009, 06:00 AM   #39
Inactive Account




Join Date: Dec 2008
Location: Ames, Iowa, USA
Photos: Albums
Posts: 2,965
QuoteOriginally posted by Haakan Quote
I just read that article, and it actually states the same thing as I said
"For the same perspective, AOV, DOF, shutter speed, and apparent exposure, all sensors, regardless of size receive the same total amount of light. Thus, under these conditions, for the same efficiency of sensor, all formats will have the same amount of total apparent image noise" (I added bold)

Best regards,
Haakan
What happens as the sensor size approaches zero, ie. no sensor at all?

Seriously; there's something amiss in the original statement, diffraction limits, etc aside.

Dave in Iowa
09-02-2009, 06:16 AM   #40
Pentaxian
Wheatfield's Avatar

Join Date: Apr 2008
Location: The wheatfields of Canada
Posts: 10,043
I love it when people throw in meaningless parameters to create false equivalencies. In this case, the whole DOF thing. If you need the extra low light advantage that a full frame camera can give, you are going to shoot in such a way to ensure that the advantage is utilized.
Throwing in an arbitrary requirement that the "depth of field has to be the same or else there is no advantage" is not a meaningful parameter 99% of the time.
09-02-2009, 07:25 AM   #41
Veteran Member
nostatic's Avatar

Join Date: Feb 2008
Location: socal
Photos: Gallery
Posts: 1,576
QuoteOriginally posted by Wheatfield Quote
I love it when people throw in meaningless parameters to create false equivalencies. In this case, the whole DOF thing. If you need the extra low light advantage that a full frame camera can give, you are going to shoot in such a way to ensure that the advantage is utilized.
Throwing in an arbitrary requirement that the "depth of field has to be the same or else there is no advantage" is not a meaningful parameter 99% of the time.
For me it is very simple. I take a shot under conditions that I need to shoot. If I'm lucky, I can compare two different cameras shooting under said conditions. I look at the results. If I'm really lucky, I do it again. If under real world conditions on camera outperforms the other, then that is the one I want.

Theory is nice to understand a phenomenon, but it is important to remember that the math/science is created to try and explain something you see. Not the other way around.
09-02-2009, 07:35 AM   #42
Veteran Member
WMBP's Avatar

Join Date: Dec 2006
Location: Dallas, Texas
Photos: Gallery
Posts: 2,496
QuoteOriginally posted by deadwolfbones Quote
All I know is I've been looking at D700 ISO12800 shots all afternoon and they're way cleaner than my K20D ISO1600 shots.
and

QuoteOriginally posted by Wheatfield Quote
I love it when people throw in meaningless parameters to create false equivalencies. In this case, the whole DOF thing. If you need the extra low light advantage that a full frame camera can give, you are going to shoot in such a way to ensure that the advantage is utilized. Throwing in an arbitrary requirement that the "depth of field has to be the same or else there is no advantage" is not a meaningful parameter 99% of the time.
I just read this whole thread. I am indeed interested in the technicalities and I do my best to understand them. I want to understand them better than I do.

But the comments above from deadwolfbones and Wheatfield strike me as valid and important. deadwolfbones's quip reminds me of the old joke, "Who you gonna believe - me or your lying eyes?"

And Wheatfield's comment really speaks to the point for me. My Pentax DSLRs produce dramatically less noisy photos at, say, ISO 800-1600, than those that I used to take with my Canon S5 IS several years ago. And like deadwolfbones, I've seen pictures from Nikon D3's taken at absurdly high ISOs that rival or surpass mine at ISO 1600. I don't care to "normalize" the comparisons, since I'm not interested in the technicalities personally, I'm only interested in the results available from real cameras in real shooting situations. I understand - I consider it quite obvious - that, in order to get the noise advantage of a given camera, you have to take the camera as a package. A Nikon D3 is a lot bigger than my old Canon S5 IS and there are tons of other technical differences. That's okay. What would matter to me (if I had a D3) is that I can actually take usable pictures with a D3 in much lower light than I could with any point and shoot. Perhaps I'd lose a little depth of field in some shots, but I'd get the shot.

Now, I hasten to add that I am not wracked with full-frame envy. I would love someone to GIVE me a D3X or even a lowly D700. If somebody wants to give me a digital Hasselblad, I'll take that, too. But in the real world where I actually have to BUY my own cameras, everything I do is a compromise, and I'm pretty happy with the compromise represented by the K20D. A Canon 5D might have 1+ stop better high ISO performance, but the Pentax has built-in shake reduction, so I can shoot slower and still hope to get a usable result. Where the Canon or a Nikon is plainly superior in technical ways, I have to work a little harder. And perhaps, in extreme lighting conditions, there will be times when my results won't be as good as those of a much more expensive camera.

But the lighting isn't extreme most of the time, and in the vast middle ground, I find that it's very hard to tell if a photo was taken by a K20D or a D3X.

Woops, there I go again, believing what I see. ;-)

Will
09-02-2009, 08:07 AM   #43
Pentaxian
falconeye's Avatar

Join Date: Jan 2008
Location: Munich, Alps, Germany
Photos: Gallery
Posts: 6,862
QuoteOriginally posted by Wheatfield Quote
the whole DOF thing.
You are right, the question asked by the OP isn't very relevant.

However, I didn't bother because it can be restated as following:

"What if I keep the physical size (diameter, weight, cost) of the lenses I use constant?" That may be more relevant than asking "What if I keep DoF constant?".

As it turns out, both questions are equivalent.

QuoteOriginally posted by newarts Quote
What happens as the sensor size approaches zero, ie. no sensor at all?
If the sensor size approaches zero so does the focal length. And the required f-stop number will approach zero as well, like requiring a f/0.001 lens
And no sensor at all means requiring an impossible lens.
09-02-2009, 11:04 AM   #44
Senior Member
Mister Guy's Avatar

Join Date: Jan 2009
Location: Raleigh, NC
Posts: 244
QuoteOriginally posted by WMBP Quote

And Wheatfield's comment really speaks to the point for me. My Pentax DSLRs produce dramatically less noisy photos at, say, ISO 800-1600, than those that I used to take with my Canon S5 IS several years ago. And like deadwolfbones, I've seen pictures from Nikon D3's taken at absurdly high ISOs that rival or surpass mine at ISO 1600. I don't care to "normalize" the comparisons, since I'm not interested in the technicalities personally, I'm only interested in the results available from real cameras in real shooting situations. I understand - I consider it quite obvious - that, in order to get the noise advantage of a given camera, you have to take the camera as a package. A Nikon D3 is a lot bigger than my old Canon S5 IS and there are tons of other technical differences. That's okay. What would matter to me (if I had a D3) is that I can actually take usable pictures with a D3 in much lower light than I could with any point and shoot. Perhaps I'd lose a little depth of field in some shots, but I'd get the shot.
Well, that's something of the point isn't it?

The whole of the statement can easily be generalized as a blanket statement that Full Frame cameras are not automatically better than APS-C cameras by sheer virtue of their sensor without any sense of context. They do not, in toto, out perform all smaller sensors.

The point of normalizing in theory is, in fact, a way to account for the real world. It is simply a way of addressing the fact that in the real world, composition is more than a hard line matter of pushing ISO. Normalizing is a way to account, in discussions, for the real world conditions that change when you change any piece of the exposure triangle.

The question becomes: If Pentax, today, made a FF camera with a K mount, would it provide better performance than the K-7?

I believe, if I'm understanding this discussion correctly, the answer becomes yes with an "if", or no with a "but".

Yes, IF you put the same FA lenses (assuming they cover, of course) on a FF body as a APS-C body, and IF you take the same picture, from the same vantage point, with the same shutter and aperture settings, you'd end up being able to use a lower ISO to take the picture, if it was available, IF you are willing to crop it to get the same field of view, but your depth of field will be slightly different, so IF your subject is within subset of the FF DOV, you will end up with a better picture.

OR

No, you can try and take the same picture, with the same field of view, and same depth of field, BUT you won't be able to take the same picture with the same setting from the same location, and thus your picture is always going to be slightly different, which will probably result in you needing to adjust your ISO to match your new settings, which will negate the advantage.

That's a very real world way of looking at it, because it prompts the more practical question: Is it REALLY the sensor size that is preventing me from taking the picture I want to take? Many people have convinced themselves the answer to that question is yes, but a careful examination indicates the circumstances when that answer really IS yes require much more nuanced construction than is often realized.
09-02-2009, 11:28 AM   #45
Veteran Member
Torphoto's Avatar

Join Date: Mar 2008
Location: Trinidad W.I.
Photos: Gallery
Posts: 620
Ok so let me get this, this thread says if I shoot my k20D vs the mk2 at 1600 iso and compare with shots from the k7 there is no difference in noise?

but comparing raw shots in similar if not identical conditions my mk2 out performs either by 1-2 stops, the noise from the mk2 at 1600 iso looks like or better than 800 from the k20d and close to the k7's 400 iso, so explain this all again to me?

Hell I can shoot the mk2 to 4000 iso and it looks better than 1600 iso from the pentaxes.

what am I doing worng here?
Closed Thread

Bookmarks
  • Submit Thread to Facebook Facebook
  • Submit Thread to Twitter Twitter
  • Submit Thread to Digg Digg
Tags - Make this thread easier to find by adding keywords to it!
aps-c, camera, dof, dslr, ff, fov, images, noise, parameters, performance, photography, shutter, time
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
K5 Low ISO Noise ----- Not! JeffJS Pentax K-5 52 06-24-2011 12:07 AM
Achieving low-noise footage? inferno10 Video and Pentax HDSLRs 13 03-03-2010 11:10 AM
Poll - $2500 low end FF or Hi-spec APS-C? - Please read initial post before voting Richard Day Pentax News and Rumors 126 02-15-2010 03:08 PM
Noise at low ISO in k200d Pusiek Pentax DSLR Discussion 10 08-24-2008 02:53 PM
Low light and noise Substitute Photographic Technique 11 11-09-2007 10:31 AM



All times are GMT -7. The time now is 11:05 AM. | See also: NikonForums.com, part of our network of photo forums!
  • Red (Default)
  • Green
  • Gray
  • Dark
  • Dark Yellow
  • Dark Blue
  • Old Red
  • Old Green
  • Old Gray
  • Dial-Up Style
Hello! It's great to see you back on the forum! Have you considered joining the community?
register
Creating a FREE ACCOUNT takes under a minute, removes ads, and lets you post! [Dismiss]
Top