Luminous Landscape Forum
Equipment & Techniques => Digital Cameras & Shooting Techniques => Topic started by: ejmartin on May 22, 2008, 09:51:35 pm
-
... for those with a technical bent:
http://theory.uchicago.edu/~ejm/pix/20d/te...oise/index.html (http://theory.uchicago.edu/~ejm/pix/20d/tests/noise/index.html)
Put on your propeller beanies
Seriously, though, fans of this site's disquisition on ETTR might be interested in the analysis on page 3 of the article, among other things.
-
Actually the article seems to be reasonably accessible - not a whole lot of nasty math.
After a quick read (particularly of the ETTR section) it seems the author is positing that the benefits of ETTR accrue due to the fact that more photons are captured as the exposure is migrated to the right, improving SN as opposed to to the quantization benefit of subsequently re-mapping the midtones of an ETTR image down into the shadows during post processing (assuming the move to the right is accomplished by holding the shutter open longer - or aperture wider - rather than boosting the ISO) - did I get that right?
Having said that, if I remember correctly, it seems to me that Jonathan W.'s sample of a "broken" image reflected damage to tonal gradations rather than noise issues.
And maybe it's just the impact of Canon's marketing, but I think I can see better shadow quality in some circumstances with the 14 bit over the previous 12 bit models. But I might be fooling myself.
-
Actually the article seems to be reasonably accessible - not a whole lot of nasty math.
After a quick read (particularly of the ETTR section) it seems the author is positing that the benefits of ETTR accrue due to the fact that more photons are captured as the exposure is migrated to the right, improving SN as opposed to to the quantization benefit of subsequently re-mapping the midtones of an ETTR image down into the shadows during post processing (assuming the move to the right is accomplished by holding the shutter open longer - or aperture wider - rather than boosting the ISO) - did I get that right?
Yes that is an accurate summary. Actually, there is a slight benefit to ETTR by ISO boost, but it is much milder than doing it by decrease in Tv or Av and almost non-existent at high ISO.
Having said that, if I remember correctly, it seems to me that Jonathan W.'s sample of a "broken" image reflected damage to tonal gradations rather than noise issues.
I'd be interested in the link if you have it...
And maybe it's just the impact of Canon's marketing, but I think I can see better shadow quality in some circumstances with the 14 bit over the previous 12 bit models. But I might be fooling myself.
[a href=\"index.php?act=findpost&pid=197482\"][{POST_SNAPBACK}][/a]
The higher shadow quality in newer Canons (which I agree is real) has mostly to do with improvements in controlling patterned read noise and a slight lowering of overall read noise. It has nothing to do with the passage from 12-bit to 14-bit tonal depth.
-
Actually, there is a slight benefit to ETTR by ISO boost, but it is much milder than doing it by decrease in Tv or Av and almost non-existent at high ISO.
Emil, I agree that improvement when reaching high ISO values is very slight, but just doing ETTR using ISO200 instead of ISO100 can be much more than a slight improvement:
Canon 350D shots setting constant aperture/shutter:
(http://www.guillermoluijk.com/article/iso/luces.jpg)
(http://www.guillermoluijk.com/article/iso/versus.jpg)
-
... for those with a technical bent:
http://theory.uchicago.edu/~ejm/pix/20d/te...oise/index.html (http://theory.uchicago.edu/~ejm/pix/20d/tests/noise/index.html)
Put on your propeller beanies
Seriously, though, fans of this site's disquisition on ETTR might be interested in the analysis on page 3 of the article, among other things.
[a href=\"index.php?act=findpost&pid=197353\"][{POST_SNAPBACK}][/a]
This is the best treatise on the subject that I have seen online and anyone with an interest in these technical issues should study it.
The analysis of ETTR is informative and confirms what I have felt for some time: the number of raw levels in the brightest f/stop of a 12 or 14 bit digital capture is not that important (because of noise), but the increased signal to noise ratio with ETTR is significant. While ETTR is important for optimal image quality, it should not be overdone since clipping does result in data loss. Since the S:N varies as the square root of the exposure, an increase of one full f/stop in exposure only improves the S:N by a factor of 1.414.
-
Emil I freely took your plots figures to make a DR comparision between 5D, 40D and D3. For simplicity I used the criteria SNR>2EV which was easy to estimate over your plots and yields DR values arounf 8-9 f-stops which seem logical in photographic usability:
(http://img301.imageshack.us/img301/2410/rangoseg4.gif)
5D plot crosses 40D's, which means while 5D is better at high ISOs even if it's an older camera, for static scenes where there is no problem to ETTR at the lowest electronic ISO 40D is around 1/3 f-stops better than 5D. For this kind of scenes D3 is 2/3 f-stops better than 5D.
Considering ISO values are correctly normalised on the three cameras, at the same ISO, D3 has about 1/2 f-stop additional DR over 40D which makes it a less noisy camera for action shooting. For ISO below 800 D3 is also better than 5D, but at ISO800 and ISO1600 5D and D3 behave the same.
Considering again static scenes, D3 is also better than 40D but for less than 1/3 f-stop.
______________________________
I have a question however: your SNR measures refer to measures of signal vs noise over individual pixels, so the effect that a larger amount of pixels (more resolution) could have in the final noise perception was not entered in the equations, right?
For instance, according to your measures, two cameras with the same SNR per pixel but having camera A twice as many pixels as camera B, would result in the same plot; but in practice camera A will have less visible noise when rescaled to the same final size in pixels due to noise averaging.
Is that right?
-
... for those with a technical bent:
http://theory.uchicago.edu/~ejm/pix/20d/te...oise/index.html (http://theory.uchicago.edu/~ejm/pix/20d/tests/noise/index.html)
Put on your propeller beanies
[{POST_SNAPBACK}][/a] (http://index.php?act=findpost&pid=197353\")
Considering the scope and quality of Emil's post, I am quite surprised that there have not been more comments. Perhaps readers need additional time to digest the article. To get discussion started, here are a few observations.
"one might think that read noise is a fixed cost per pixel in recording an image, again assuming that the same quality of circuit elements are employed."
This does appear to be true, but should one measure the cost in terms of noise expressed in data numbers (DN, raw pixel values) or electrons? In [a href=\"http://www.clarkvision.com/imagedetail/digital.sensor.performance.summary/#read_noise]Clark Fig 3[/url] of a post, Roger Clark notes that read noise expressed in electrons is similar in sensors of the same generation but of differing pixel sizes. Read noise varies with ISO, but read noise quoted in manufacturer's spec sheets is typically for unity gain, which complicates the analysis since unity gain is correlated with pixel size.
One can increase signal to noise at the expense of resolution by pixel binning as Emil discusses. The results of binning differ when binning is done in hardware prior to digitization or in software by downsizing. Consider hardware binning in which 4 pixels in a 2 by 2 array are combined and read out together, as described here (http://www.photomet.com/library/library_encyclopedia/library_enc_binning.php). The four pixels are combined into a superpixel, which is read out with the same read noise (in electrons) as for a single small pixel. This implies that when other factors are held constant, read noise is independent of pixel size. If one performed binning in Photoshop after the fact, the resulting superpixel would have incurred 4 read noise contributions, not a single one as with hardware binning.
If the read noise is similar for large and small pixels, the large pixel camera still has an advantage when read noise is measured in DNs rather than electrons. This is because of camera gain as Roger explains in this example: Clark Table 3 (http://www.clarkvision.com/imagedetail/does.pixel.size.matter/index.html#Example_Noise). See the text just following table 3.
"The realization from Table 3 is even though the read noise is similar in terms of electrons, the effect on the image is huge because of the gain factor. Thus shadow detail on a small sensor is severely compromised as the gain factor drops."
Seriously, though, fans of this site's disquisition on ETTR might be interested in the analysis on page 3 of the article, among other things.
[a href=\"index.php?act=findpost&pid=197353\"][{POST_SNAPBACK}][/a]
Any comments by the resident "experts" who have expounded on the "levels theory", where half the levels are in the brightest f/stop? The improved signal to noise ratio with ETTR is a definite benefit, but how does it scale to human perception? Considering only shot noise, the S:N increases by a factor of sqrt(2) with doubling of the exposure. Looking at Emil's S:N graphic demonstration, I perceive only slight image improvement with a doubling of S:N. Is this a log function or what?
Bill
-
Emil, I agree that improvement when reaching high ISO values is very slight, but just doing ETTR using ISO200 instead of ISO100 can be much more than a slight improvement:
Canon 350D shots setting constant aperture/shutter:
[a href=\"index.php?act=findpost&pid=197523\"][{POST_SNAPBACK}][/a]
Well it is a substantial improvement in this example -- but if the ISO 1600 image was properly exposed, the ISO 100 shot here is 4 stops underexposed, a stop or more than that in the red and blue channel if these were shot under indoor lighting, then you are way down into the read noise. My comment about mild improvement had in mind ISO 200 properly exposed vs ISO 100 one stop underexposed. But you are right the biggest improvement will come at ISO 100 vs 200 since the read noise in Canons is almost constant between the two. Going all the way from 100 to 1600 on some cameras is about a factor of 4 or two stops improvement in shadow noise comparing in-camera to post-processing gain-up.
-
My comment about mild improvement had in mind ISO 200 properly exposed vs ISO 100 one stop underexposed. But you are right the biggest improvement will come at ISO 100 vs 200 since the read noise in Canons is almost constant between the two.
That's perfect, thank you Emil.
BTW I sent you a message, just to find out if you have some reference url with links to all your articles. I think they are incredibly valuable.
BR
-
I have a question however: your SNR measures refer to measures of signal vs noise over individual pixels, so the effect that a larger amount of pixels (more resolution) could have in the final noise perception was not entered in the equations, right?
Correct. Data are for individual pixels; comments on how to scale the data to account for varying pixel size are in the section on pixel size on page 3.
For instance, according to your measures, two cameras with the same SNR per pixel but having camera A twice as many pixels as camera B, would result in the same plot; but in practice camera A will have less visible noise when rescaled to the same final size in pixels due to noise averaging.
Is that right?
[a href=\"index.php?act=findpost&pid=197796\"][{POST_SNAPBACK}][/a]
Two comparisons that I think are valuable are noise scaled to the image size, and noise scaled to fixed spatial scale -- that is, referred to a scale in lines/picture height, or to a scale in lines/mm. Noise scales linearly rather than as an area (as a result of combining as sum of squares rather than a simple sum), and so noise scales with the pixel spacing for instance if referring to fixed scale in lines/mm, or inversely with the number of pixels per picture height if referring to the frame size.
-
or with the number of pixels per picture height if referring to the frame size.
I think this is the more valuable performance indicator for the photographer.
-
Considering the scope and quality of Emil's post, I am quite surprised that there have not been more comments. Perhaps readers need additional time to digest the article. To get discussion started, here are a few observations.
[a href=\"index.php?act=findpost&pid=198063\"][{POST_SNAPBACK}][/a]
I think it's an excellent exposition of the subject. Unfortunately, the time I first attempted to download the article, the connection was very,very, very slow. The mouseovers didn't work at all and it just took too long to move from one heading to the next. It's now a bit faster and the mouseover image is working.
What I find particularly interesting are the plots of S/N ratio at various ISO's for the 5D, D3 and 40D.
It's interesting because I happen to own both a 5D and 40D and always got the impression that the 5D provided noticeably better image quality at high ISO. These plots confirm that impression.
Whilst I'm holding off buying a Nikon D3, I did take the trouble to do a few test shots in a store in Bangkok (comparing it with my 5D), but all my test shots were at ISO 3200 and above. I never thought of comparing the D3's noise performance at lower ISO's, and I didn't really have the time if I had thought of it.
I'm surprised that according to these plots the Nikon D3 from ISO 200 to ISO 1600 has the same read noise as the 5D. But I'm puzzled how Guillermo's plots comparing dynamic range can show the D3 being better than the 5D at ISO 100 to the same degree that the 5D is better than the Nikon D3 at ISO 800. That seems really surprising.
-
I'm surprised that according to these plots the Nikon D3 from ISO 200 to ISO 1600 has the same read noise as the 5D. But I'm puzzled how Guillermo's plots comparing dynamic range can show the D3 being better than the 5D at ISO 100 to the same degree that the 5D is better than the Nikon D3 at ISO 800. That seems really surprising.
Hi Ray, I just plotted DR according to the chosen criteria SNR>2EV on the SNR plots referred close to Emil's Fig. 12. According to the article these plots combine read and shot noise (perhaps that's the reason).
Values are aproximated, just visually inspected:
(http://img181.imageshack.us/img181/4927/5daj9.gif)
(http://img301.imageshack.us/img301/2410/rangoseg4.gif)
However 5D is never better than D3 using this SNR criteria for DR.
-
I think it's an excellent exposition of the subject. Unfortunately, the time I first attempted to download the article, the connection was very,very, very slow. The mouseovers didn't work at all and it just took too long to move from one heading to the next. It's now a bit faster and the mouseover image is working.
What I find particularly interesting are the plots of S/N ratio at various ISO's for the 5D, D3 and 40D.
It's interesting because I happen to own both a 5D and 40D and always got the impression that the 5D provided noticeably better image quality at high ISO. These plots confirm that impression.
Whilst I'm holding off buying a Nikon D3, I did take the trouble to do a few test shots in a store in Bangkok (comparing it with my 5D), but all my test shots were at ISO 3200 and above. I never thought of comparing the D3's noise performance at lower ISO's, and I didn't really have the time if I had thought of it.
I'm surprised that according to these plots the Nikon D3 from ISO 200 to ISO 1600 has the same read noise as the 5D. But I'm puzzled how Guillermo's plots comparing dynamic range can show the D3 being better than the 5D at ISO 100 to the same degree that the 5D is better than the Nikon D3 at ISO 800. That seems really surprising.
[a href=\"index.php?act=findpost&pid=198191\"][{POST_SNAPBACK}][/a]
John Sheehy pointed out that the read noise figures for the 5D were out of line with those from 5D's he had measured. While I trust the source of the data I used, the body may have not been a typical copy. Some data from Peter Ruevski (whose 1D3 data I used) are more in line with the figures John quoted, so I may update the 5D plots to reflect this. Current data seem to be about 15-20% low for 5D read noise at high ISO.
This is one shortcoming of current testing by the people who do this sort of thing -- there is considerable sample variation, and so it is difficult to put reliable error bars on the data.
Here is the plot of the data from the other 5D:
(http://theory.uchicago.edu/~ejm/pix/20d/tests/noise/5d_noisesq_vs_iso-ruevski.gif)
-
... for those with a technical bent:
http://theory.uchicago.edu/~ejm/pix/20d/te...oise/index.html (http://theory.uchicago.edu/~ejm/pix/20d/tests/noise/index.html)[a href=\"index.php?act=findpost&pid=197353\"][{POST_SNAPBACK}][/a]
Emil,
as one of those with a technical bent (also a one-time Cargese participant!), thanks very much for your excellent essay. To answer Bill Janes' question, I have not commenting so far because I am happy reading, thinking and learning.
One comment though: I am reassured to see further evidence that 12-bit A/D converters depth are not in practice a limit to current DSLR image quality. For one thing it adds to the promise of the Sony/Nikon approach of massively parallel on-sensor A/D conversion (by removing concerns about its being 12-bit only). That approach could help to eliminate the frame rate disadvantage that digital SLR's still often have compared to film SLRs, and finally offer cutting edge high resolution and high frame rates together.
-
"one might think that read noise is a fixed cost per pixel in recording an image, again assuming that the same quality of circuit elements are employed."
This does appear to be true, but should one measure the cost in terms of noise expressed in data numbers (DN, raw pixel values) or electrons?
For the noise coming from sensor readout (what I called noise "upstream" of the ISO amplifier) electrons makes sense to me -- the noises are directly related to counting electrons; for subsequent contributions to noise (what I called "downstream" noise) the electron count has long since been converted into a voltage, the noises are voltage fluctuations, and I don't see any great benefit to continuing to think of the noise in terms of electrons. I chose ADU for a couple of reasons; first, converting the data to electrons compounds measurement errors -- those of the read noise measurement, which is directly in ADU, and the gain measurement, are included when one refers the noise to electrons. Second, it seems more intuitive to me to set the noise in ADU, since it grows with ISO due to the amplification of the sensor readout noise, and this accords with the increased visual appearance of noise at higher ISO (rather, decreased S/N).
In Clark Fig 3 (http://www.clarkvision.com/imagedetail/digital.sensor.performance.summary/#read_noise) of a post, Roger Clark notes that read noise expressed in electrons is similar in sensors of the same generation but of differing pixel sizes. Read noise varies with ISO, but read noise quoted in manufacturer's spec sheets is typically for unity gain, which complicates the analysis since unity gain is correlated with pixel size.
If the read noise is similar for large and small pixels, the large pixel camera still has an advantage when read noise is measured in DNs rather than electrons. This is because of camera gain as Roger explains in this example: Clark Table 3 (http://www.clarkvision.com/imagedetail/does.pixel.size.matter/index.html#Example_Noise). See the text just following table 3.
"The realization from Table 3 is even though the read noise is similar in terms of electrons, the effect on the image is huge because of the gain factor. Thus shadow detail on a small sensor is severely compromised as the gain factor drops."
Bill
[a href=\"index.php?act=findpost&pid=198063\"][{POST_SNAPBACK}][/a]
Well, this ties in with the previous comments -- it depends which noise contributions one wants to focus on as staying fixed. By referring back to electrons, downstream noise contributions to read noise are falsely downplayed in the DSLR relative to the digicam (since they are more fixed in ADU; only upstream read noise and photon noise are properly referred to electrons).
There is no doubt that small pixels gather fewer photons, have lower gain/pixel, and so have lower SNR/ph; but they do NOT, as I tried to emphasize, have a substantially lower SNR/mm. One should be clear in these sorts of comparisons whether one is talking about small vs large format sensors, and the different pixel sizes they entail -- or about merits/deficiencies of higher pixel densities on a given format; that makes a huge difference in the conclusions to be drawn.
-
For the noise coming from sensor readout (what I called noise "upstream" of the ISO amplifier) electrons makes sense to me -- the noises are directly related to counting electrons; for subsequent contributions to noise (what I called "downstream" noise) the electron count has long since been converted into a voltage, the noises are voltage fluctuations, and I don't see any great benefit to continuing to think of the noise in terms of electrons. I chose ADU for a couple of reasons; first, converting the data to electrons compounds measurement errors -- those of the read noise measurement, which is directly in ADU, and the gain measurement, are included when one refers the noise to electrons. Second, it seems more intuitive to me to set the noise in ADU, since it grows with ISO due to the amplification of the sensor readout noise, and this accords with the increased visual appearance of noise at higher ISO (rather, decreased S/N).
Well, this ties in with the previous comments -- it depends which noise contributions one wants to focus on as staying fixed. By referring back to electrons, downstream noise contributions to read noise are falsely downplayed in the DSLR relative to the digicam (since they are more fixed in ADU; only upstream read noise and photon noise are properly referred to electrons).
[a href=\"index.php?act=findpost&pid=198359\"][{POST_SNAPBACK}][/a]
I agree that expressing noise in ADUs rather than electrons is most logical, since ADUs are what one measures and what are in the raw file. However, since electrons and ADUs are related by a linear scaling factor (the gain, expressed as electrons/ADU), one may use either in calculations. The point of my post was that shot noise and read noise scale differently with respect to pixel size.
The following chart represents a noise model for three different pixel sizes for sensors with otherwise identical characteristics. Calculations are shown for the maximum signal and for the signal 8 stops down from saturation. With the 8 um pixel, shot noise predominates at all signal levels, whereas with the 2 um pixel read noise becomes important in the shadows. The chart demonstrates the effect of gain on the read noise for the different pixel sizes. Although the read noise is the same when expressed in electrons (5 e-) for all pixel sizes, the read noise expressed in ADUs (DN in the chart), is much higher for the smaller pixels.
(http://bjanes.smugmug.com/photos/302902478_pipo2-O.gif)
-
I found this interesting:
"On the other hand, if shooting raw, it makes little sense to use the extended ISO's since they are simply mathematical manipulations of the raw data post-capture, and their main effect is to throw away one or more stops of highlight headroom as the doubling, quadrupling etc of the raw values pushes more and more of them beyond the maximum recordable value of 4095 for 12-bit, or 16383 for 14-bit data. Setting the highest analog ISO amplification keeps the headroom, and one can always do as much additional software amplification as is needed afterward during raw conversion."
Does this also apply to the "low end"? For the D3 and D300, ISO 200 is the lowest "real" ISO but there is an L1 ISO 100..is something "thrown away" with this setting?
Steve
-
I found this interesting:
Does this also apply to the "low end"? For the D3 and D300, ISO 200 is the lowest "real" ISO but there is an L1 ISO 100..is something "thrown away" with this setting?
Steve
[{POST_SNAPBACK}][/a] (http://index.php?act=findpost&pid=198541\")
For additional analysis, you might want to look at Roger Clark's article on testing the Canon 1D MII in the conclusions section, where he discusses ISO 50 on the Canon cameras:
[a href=\"http://www.clarkvision.com/imagedetail/evaluation-1d2/index.html]Clark Conclusions[/url]
In the case of low ISO, additional exposure is given and the camera amplifier gain is decreased. The S:N is increased, but the dynamic range is decreased and there is danger of blowing the highlights.
Active D lighting is essentially the reverse. Exposure is decreased, giving more headroom for the highlights, and the amplifier gain is increased.
Bill
-
For additional analysis, you might want to look at Roger Clark's article on testing the Canon 1D MII in the conclusions section, where he discusses ISO 50 on the Canon cameras:
Clark Conclusions (http://www.clarkvision.com/imagedetail/evaluation-1d2/index.html)
In the case of low ISO, additional exposure is given and the camera amplifier gain is decreased. The S:N is increased, but the dynamic range is decreased and there is danger of blowing the highlights.
Active D lighting is essentially the reverse. Exposure is decreased, giving more headroom for the highlights, and the amplifier gain is increased.
Bill
[a href=\"index.php?act=findpost&pid=198550\"][{POST_SNAPBACK}][/a]
For the Nikon, I am assuming that ISO 200 is the lowest AD amp gain and that the L1 (100) ISO is an in camera software manipulation. If so, does the L1 "ISO" provide any benefit for a RAW image that can't be done better on a computer. Also, is there a "downside" of L1 as there seems to be with the H1 and H2 "ISO" on the high end.
This question came to mind after reading that Michael used "ISO" 100 to expose his "Maelstrom" photo.
Steve
-
For the Nikon, I am assuming that ISO 200 is the lowest AD amp gain and that the L1 (100) ISO is an in camera software manipulation. If so, does the L1 "ISO" provide any benefit for a RAW image that can't be done better on a computer. Also, is there a "downside" of L1 as there seems to be with the H1 and H2 "ISO" on the high end.
This question came to mind after reading that Michael used "ISO" 100 to expose his "Maelstrom" photo.
Steve
[a href=\"index.php?act=findpost&pid=198559\"][{POST_SNAPBACK}][/a]
I have no data on the D3, but on the D300, the raw data of the camera I tested shows that the L1 low ISO extension is actually ISO 125. One sees this from the gain measured in photons/raw level, which is not quite twice that of ISO 200 on the D300; if you work out what it corresponds to you find ISO 125. The raw data goes up to the same saturation level as it does at higher ISO. Amusingly, there is no false advertising -- when you turn on LO ISO extension, the ISO field in the metadata is left blank
On current Canons (Mark 3 1D/1Ds and 40D) ISO 100 is truly ISO 100, but the raw saturation point is somewhat below the 14-bit maximum of 16383; this means that the "true" base ISO is somewhat higher than 100, in that the sensor is maxing out early -- one would have to use a slightly higher ISO to have the sensor saturate at the 14-bit max. The same was true of ISO 50 on the 1D2; raw saturation occurred below the max 12-bit level for this ISO. What used to occur at ISO 50 in the older generation, now occurs at ISO 100 in the newer generation of Canons. This is a good thing, it means that the camera is more efficient in capturing photons (more sensitive to light) since photosite saturation occurs at a higher ISO than before. In the current models, ISO 50 is not a true ISO at all, it is ISO 100 overexposed by one stop "under the hood". You have one less stop of highlight headroom to play with, but you get better SNR because you've exposed ISO 100 more to the right.
So on current Canons, there is no difference in the raw data between ISO 50 on the one hand, and ISO 100 with the same Tv/Av and the raw data pulled by one stop in post-processing. On the Nikon D300, the low ISO extension is not quite ISO 100, and so one should take that into account if shooting manual mode. It would be nice to have data from a D3 to examine.
-
I would like to see how the Sigma SD14 would fare, considering that it has a few less factors to deal with.
-
I would like to see how the Sigma SD14 would fare, considering that it has a few less factors to deal with.
[a href=\"index.php?act=findpost&pid=198587\"][{POST_SNAPBACK}][/a]
Someone with an SD14 supplied me with some raw files for just this purpose; unfortunately, the software tools I have been using process the raw data to an extent that I was unable to extract the needed figures for read noise and sensor gain. I believe that is more than a problem with those software tools, apparently the Foveon data undergoes some serious massaging before it becomes an exposure level for each color channel, and that massaging affects the noise measurements in ways that I don't yet know how to account for.
-
So on current Canons, there is no difference in the raw data between ISO 50 on the one hand, and ISO 100 with the same Tv/Av and the raw data pulled by one stop in post-processing. On the Nikon D300, the low ISO extension is not quite ISO 100, and so one should take that into account if shooting manual mode. It would be nice to have data from a D3 to examine.
I reached the same conclusion when analysing 2 Canon 5D RAW files at ISO50 and ISO100. Keeping aperture/shutter RAW histograms were identical:
ISO50, 1/128.0s, f/8.0
(http://www.guillermoluijk.com/article/isos5d/iso50.gif)
ISO100, 1/128.0s, f/8.0
(http://www.guillermoluijk.com/article/isos5d/iso100.gif)
I did the RAW extraction with dcraw -v -D -T -4.
When developing the image the result was obviously the same on both RAW files:
(http://www.guillermoluijk.com/article/isos5d/escena1.jpg)
The funny thing comes when the owner of the RAW files reported that when opening both RAW files in Adobe Camera Raw and setting all adjustments (specially exposure) equal for both, the ISO50 RAW file displayed 1 f-stop darker.
Does this mean ACR knew the fake ISO50 particularity in the 5D and displayed the image accordingly, and not just looking at the pure RAW levels?
If that is true I still trust less commercial developers than before.
BR
-
I have no data on the D3, but on the D300, the raw data of the camera I tested shows that the L1 low ISO extension is actually ISO 125. One sees this from the gain measured in photons/raw level, which is not quite twice that of ISO 200 on the D300; if you work out what it corresponds to you find ISO 125. The raw data goes up to the same saturation level as it does at higher ISO. Amusingly, there is no false advertising -- when you turn on LO ISO extension, the ISO field in the metadata is left blank
[a href=\"index.php?act=findpost&pid=198572\"][{POST_SNAPBACK}][/a]
I performed some tests with the D3 by exposing a Stouffer wedge at f/8 and varying the shutter speed in 1/3 EV increments at both ISO 200 and L 1.0 (ISO 100) and measuring the resulting pixel level in the green1 channel of the raw files using Rawanalyze. As a check, I also measured the levels in Iris and got the same values. The data are shown graphically. For ISO 200, the proper exposure is 1/25 s at f/8, which places the green1 channel just short of clipping. For ISO 100, 1/25 s at f/8, light falling on the sensor is the same as for the ISO 200 exposure and the same pixel values are recorded in the raw file. This indicates that the camera amplifier gain is the same for both exposures.
However, the proper exposure for ISO 100 would be 1/13 s at f/8. Again, the same raw data values are obtained for ISO 200 exposed at 1/13 s at f/8. The green1 channel is saturated in both instances.
-
I'll upload some images on usendit and post the URLs in a separate reply in this thread. Right now my head is swimming in confusion after this exercise.
[{POST_SNAPBACK}][/a] (http://index.php?act=findpost&pid=198773\")
Img007, f/8.0, 1/13 sec, ISO 200:
[a href=\"http://download.yousendit.com/1DCB1F446BB7D9F6]http://download.yousendit.com/1DCB1F446BB7D9F6[/url]
Img012, f/8.0, 1/25 s, ISO 100:
http://download.yousendit.com/8142C2AA3397556C (http://download.yousendit.com/8142C2AA3397556C)
Img015, f/8.0, 1/13 s, ISO 100:
http://download.yousendit.com/8AAC1A4219A6209C (http://download.yousendit.com/8AAC1A4219A6209C)
Img004, f/8, 1/25 s, f/8, ISO 200:
http://download.yousendit.com/A978F9C3545E442C (http://download.yousendit.com/A978F9C3545E442C)
Bill
-
I performed some tests with the D3 by exposing a Stouffer wedge at f/8 and varying the shutter speed in 1/3 EV increments at both ISO 200 and L 1.0 (ISO 100) and measuring the resulting pixel level in the green1 channel of the raw files using Rawanalyze. As a check, I also measured the levels in Iris and got the same values. The data are shown graphically. For ISO 200, the proper exposure is 1/25 s at f/8, which places the green1 channel just short of clipping. For ISO 100, 1/25 s at f/8, light falling on the sensor is the same as for the ISO 200 exposure and the same pixel values are recorded in the raw file. This indicates that the camera amplifier gain is the same for both exposures.
However, the proper exposure for ISO 100 would be 1/13 s at f/8. Again, the same raw data values are obtained for ISO 200 exposed at 1/13 s at f/8. The green1 channel is saturated in both instances.
[a href=\"index.php?act=findpost&pid=198773\"][{POST_SNAPBACK}][/a]
I think this first test is the most telling -- the fact that the raw values are more or less the same for ISO 200 and the LO extension, for the same exposure, says to me that the LO iso setting is secretly ISO 200.
The rest of what you did seems to me telling more about how ACR reacts to the ISO metadata tag than it tells us about the actual ISO of the LO setting. You can't reliably discern what the true ISO is from how much compensation is needed from "0" to get the wedge values to match; for instance, HTP on Canons and ADL on Nikons use a lower ISO than is set in the camera to record the image. The raw converter reads that tag and applies a different exposure compensation than it does with HTP off (about one stop). So you couldn't tell from ACR that ISO 400 with HTP on is the same gain as ISO 200 with HTP off, but you can tell it by seeing that the raw histograms are the same.
-
I have completed the DR comparision, introducing now the Mark III cameras, and applying the SNR correction suggested by Emil in 'BIG PIXELS vs. small pixels', i.e. SNR scales with the image resolution in pixels. I took 40D as a fixed reference so SNR and hence DR improved for the rest of cameras with more pixels:
Now 5D clearly takes its place over the 40D. The 1D3 yields very good results specially taking into account it's a 1.3 sensor. D3 beats the 1D3 and 1Ds3 beats D3 thanks to those huge 20Mpx.
(http://img77.imageshack.us/img77/677/graph2ys3.gif)
-
I have completed the DR comparision, introducing now the Mark III cameras, and applying the SNR correction suggested by Emil in 'BIG PIXELS vs. small pixels', i.e. SNR scales with the image resolution in pixels. I took the 40D as a fixed reference so SNR and hence DR improved for the rest of cameras with more pixels:
[a href=\"index.php?act=findpost&pid=198858\"][{POST_SNAPBACK}][/a]
That's a very well-presented chart, Guillermo. I struggle with a lot of the in-depth technical discussion here, but simple graphs are easy .
What I find interesting here is that at the high ISOs of ISO 800 and 1600, that are probably used most without too much compromise in image quality, there's virtually no difference in DR between the 5D, the 1D3 and the Nikon D3.
It appears to be at ISO 200 that the Nikon D3 has the greatest DR advantage over the Canon competition, with the exception of the 1Ds3 as a result of its greater pixel count.
-
I reached the same conclusion when analysing 2 Canon 5D RAW files at ISO50 and ISO100. Keeping aperture/shutter RAW histograms were identical:
[a href=\"index.php?act=findpost&pid=198607\"][{POST_SNAPBACK}][/a]
Guillermo,
I observed similar behavior with the D3. The raw file has the same values at ISO 200 (the lowest normal setting) and at ISO 100 (the L 1.0 setting).
The funny thing comes when the owner of the RAW files reported that when opening both RAW files in Adobe Camera Raw and setting all adjustments (specially exposure) equal for both, the ISO50 RAW file displayed 1 f-stop darker.
Does this mean ACR knew the fake ISO50 particularity in the 5D and displayed the image accordingly, and not just looking at the pure RAW levels?
If that is true I still trust less commercial developers than before.
[a href=\"index.php?act=findpost&pid=198607\"][{POST_SNAPBACK}][/a]
With the D3, the fake ISO 100 shots are also rendered 1 stop darker than the ISO 200 shots with the same exposure (f/stop and shutter speed). I think that this is the intended behavior, so that the ISO is taken into account when the file is rendered. Unfortunately, the raw channels can be blown and there would be no indication of this in ACR.
Bill
-
... for those with a technical bent:
http://theory.uchicago.edu/~ejm/pix/20d/te...oise/index.html (http://theory.uchicago.edu/~ejm/pix/20d/tests/noise/index.html)
Put on your propeller beanies
Seriously, though, fans of this site's disquisition on ETTR might be interested in the analysis on page 3 of the article, among other things.
[a href=\"index.php?act=findpost&pid=197353\"][{POST_SNAPBACK}][/a]
Emil,
I´m not yet involved with digital photography but about to make the switch. So my following question is due to my minor knowledge. In "Aspects of Noise Reduction in Raw Data" you state that
"Unfortunately, none of the noise reduction methods outlined above can be conveniently applied to raw images using current commercial raw converters such as ACR, DPP, Capture NX, Capture One, Bibble, Aperture, etc. The methods involve the preparation of a noise reduction template, which must then be applied to the raw image before raw conversion. None of the mentioned raw converters have the ability to apply such a template..."
So how can the techniques you describe then be applied? Is it possible to use them manually in Photoshop?
Best regards!
Marc Wossner
-
Emil,
I´m not yet involved with digital photography but about to make the switch. So my following question is due to my minor knowledge. In "Aspects of Noise Reduction in Raw Data" you state that
"Unfortunately, none of the noise reduction methods outlined above can be conveniently applied to raw images using current commercial raw converters such as ACR, DPP, Capture NX, Capture One, Bibble, Aperture, etc. The methods involve the preparation of a noise reduction template, which must then be applied to the raw image before raw conversion. None of the mentioned raw converters have the ability to apply such a template..."
So how can the techniques you describe then be applied? Is it possible to use them manually in Photoshop?
Best regards!
Marc Wossner
[a href=\"index.php?act=findpost&pid=199258\"][{POST_SNAPBACK}][/a]
Currently it is not possible to do properly in Photoshop dark frame subtraction or other noise suppression of the sort discussed in the article. It is important to do the subtraction in the raw state, before the image is white balanced, demosaiced, gamma corrected, etc. The noise templates give the specific information about what to subtract from each pixel, before the pixel values are mixed up by the interpolation of colors and nonlinear transformations of gamma correction and such. Attempting to do the subtraction after raw conversion is pretty much hopeless.
The sorts of noise subtractions I described are standard procedure in astrophotography, and there are a number of software tools that are used in that community (IRIS, ImagesPlus, etc) that can do the job properly; but they have their own shortcomings (for instance IRIS has no color management AFAIK, and the demosaicing algorithm is pretty basic).
I find it hard to believe that dark frame subtraction is going to be implemented in ACR for instance, probably not a feature for which there is high demand, and let's face it, many if not most images are not strongly affected by a little extra noise in the deep, deep shadows. That's why I think there is room for a little application that allows one to take the output file of a little preprocessing on the side in the raw state in eg an application like IRIS, and then stuff the result into a raw dng file together with the metadata of the original raw file, that one could process with one's regular raw converter. For the occasional problem image or an image that one wants to lavish a lot of extra careful attention on.
-
Currently it is not possible to do properly in Photoshop dark frame subtraction or other noise suppression of the sort discussed in the article. It is important to do the subtraction in the raw state, before the image is white balanced, demosaiced, gamma corrected, etc. The noise templates give the specific information about what to subtract from each pixel, before the pixel values are mixed up by the interpolation of colors and nonlinear transformations of gamma correction and such. Attempting to do the subtraction after raw conversion is pretty much hopeless.
...
But a lot of articles on the web promote the idea of doing at least dark frame subtration in Photoshop using .jpeg and layers.
http://www.astropix.com/HTML/J_DIGIT/JPG_DFS.HTM (http://www.astropix.com/HTML/J_DIGIT/JPG_DFS.HTM) even states that "It is a common misconception that dark frame subtraction can only be beneficial on Raw images." While I cant say weather this is true or not because Im not digital yet and have no images to compare I find it interesting to think about the various possibilities. So what about a split workflow using .raw and .tif files. For pattern noise that might look like that: Take a couple of dark frames with short exposure and average their raw files together in IRIS or ImagesPlus to get a bias frame. Save the bias frame as 16-bit .tif and subtract it in Photoshop from the image using layers and blending mode difference. The same workflow should work for dark frames with long exposures to reduce thermal noise (and pattern noise and amplifier glow).
Has anyone experience with this kind of image manipulation and can tell if it works without introducing too many additional errors? And what about the reduction of pixel response non-uniformity? To do so one must divide the raw values of the image by the corresponding pixel raw values of a so called flat field. This can be done in IRIS (http://www.astrosurf.com/buil/us/iris/iris.htm. But will it give good results using 16-bit .tif files for both the image and the bias frame?
Best regards!
Marc Wossner
-
The following chart represents a noise model for three different pixel sizes for sensors with otherwise identical characteristics.
Can you point me to the technical specs of these sensors? I ask because the research I've done (across Kodak sensors, for example) shows that, within a family of sensors, the number of electrons is actually less than proportionate for smaller pixels. That is to say that the electron count of 4 smaller pixels doesn't add up to the count of 1 pixel of equal size. So I’d be interested viewing the specs of a family of sensors that do show electron counts being proportional to size. Thanks.
-
Can you point me to the technical specs of these sensors? I ask because the research I've done (across Kodak sensors, for example) shows that, within a family of sensors, the number of electrons is actually less than proportionate for smaller pixels. That is to say that the electron count of 4 smaller pixels doesn't add up to the count of 1 pixel of equal size. So I’d be interested viewing the specs of a family of sensors that do show electron counts being proportional to size. Thanks.
[{POST_SNAPBACK}][/a] (http://index.php?act=findpost&pid=201495\")
No, my analysis was from a mathematical model. In the real world, the full well capacity vs pixel size relationship is not linear and the slope of the curve generally increases as pixel size increases, as you mention. Roger Clark has constructed a model for this relationship.
[a href=\"http://www.clarkvision.com/imagedetail/digital.sensor.performance.summary/]Sensor Performance Summary[/url]
CCDs have a relatively high fill factor, and I would expect the relationship to be more linear than with CMOS, where the on chip image processing transistors take up a certain percentage of the pixel area. As the pixel area increases, I would think that a higher proportion of the pixel area could be dedicated to light capture.
Bill
-
What an excellent article! And I'm thrilled to see so many people on this forum with a technical bent. Back in 2002 I found out about these concepts, and it totally changed the way I think about photography (and the way that I do photography, as well).
My first camera with RAW capability was an Olympus E-20N. For me, this was a doorway into experimentation. I started out merely taking bias, dark, and flat frames, and writing programs to average them, subtract the bias, scale and subtract the dark, and divide by the flat. I used this to refine some heavily-stacked astrophotos. (The E-20N was a quasi-DSLR, with a beam-splitter instead of a mirror, a fixed zoom lens with a leaf shutter, and a 2/3" sensor. Needless to say it was much noiser than "true" DSLRs, and astrophotography required a whole bunch of stacked frames to beat down the noise.)
I came up with a way to plot variance (standard deviation squared) versus data level (ADU) with a huge amount of precision. The conventional technique is to measure noise locally within a region of solid color in one photo — this is not only subject to PRNU but assumes that the region doesn't have any tonal variation; and the number of data points is limited to the number of solid color swatches (or exposures at different light levels).
I decided I wanted to compare each pixel only with itself, across multiple exposures — camera on tripod, triggered remotely, photographing a scene with as much dynamic range as I could squeeze in, defocused to minimalize the effects of cumulative camera movement (ideally there'd be none of this, but even the teensiest subpixel movement threw off my data otherwise).
There was a major problem: when I took multiple exposures of the same thing, the brightness in each frame was different. I didn't know whether this resulted more from variations in my light source or in my camera's shutter lengths. It didn't matter, because I had no way of eliminating either variation.
I realized that with only one light source, and constant or negligable thermal noise, the frames differing brightnesses could be averaged together to make a reference frame that still had a monotonic tone curve. So I took this "reference frame", with much lower noise than each frame that went into it, and sorted the pixels by brightness.
The sorted list is merely a list of x,y coordinate pairs, going from darkest to lightest. The upshot of this is that each individual source frame can then be processed, and pixels grouped by brightness; simply take a subregion of the sorted list, and all the pixels at those coordinates will be very close in average brightness (using knowledge gleaned from the full set of frames) but will follow a Poisson distribution within the one frame. So the mean and the variance provide a data point for the graph; then the program can move on to the next set of pixels adjacent in the list ("adjacent" in the list, not adjacent spatially). (Actually, for extra precision my program pools the data from all the frames, using the sorted pixel list, before doing any calculations of mean and variance; but the way in which this is done is analogous to doing it on one frame.)
Some simple algebra gave me a formula to get a graph of gain from the graph of variance. Integrating this yields a graph of linearity.
It turns out that the graph of gain shows some interesting features when mapped with great precision. Here are the graphs I came up with:
Olympus E-20N @ ISO 80 (http://www.pbase.com/image/8464261/original)
Olympus E-20N @ ISO 320 (http://www.pbase.com/image/9279719/original)
Notice that at ISO 80, the gain varies quite significantly, going from 7.3 to 8.6 electrons per ADU. This results in a significantly nonlinear "tone curve". When I used this tone curve in my astrophotography stacking program, the division-by-flat-frame became more accurate, and the last trace of PRNU disappeared!
Here is the astrophotography experiment (http://www.pbase.com/image/7490631/original) where I put all this to use; I was limited to photographing the sky around Polaris, because I don't have a tracking mount (I made my stacking program rotate the frames according to timestamp). PRNU was quite prominent in this, because that region of sky is quite light-polluted at my home. (I subtracted the light pollution using a windowed median function.)
Back when I first did this experiment, I tried to find other people to do it on their own cameras (especially true DSLRs) by posting on a dpreview.com forum. I didn't get many responses, and the one person I recruited to do it on his own camera didn't follow through. I think it will be different, now, on this forum!
Three weeks ago I finally got a DSLR, and have begun to do these experiments on it:
Canon Rebel XSi (450D) @ ISO 100 (http://www.pbase.com/image/98966771/original)
(I was also shocked to realize that my new DSLR has different read noises at different ISOs. I expected it to have the same read noise (in electrons) at all ISOs, just like my old Olympus E-20; I expected to be able to stay at ISO 100 always, and simply underexpose when necessary, never having to worry about highlights being blown. But I suppose that'd be too easy!)
My experiment could use lots of refinement. For one thing, the way I arrange a highly-dynamic-ranged scene is pretty much seat-of-the-pants; I use my computer room as the scene (photographed at 18mm), and move objects around to "sculpt" the histogram. What is needed is a standard way to sculpt the histogram into being utterly linear. Photographing a print doesn't come close to being good enough, because the blacks aren't black enough. I'm thinking perhaps a 3-dimensional shape could have the proper reflective characteristics, but I'm not sure how to design it.
Sources of error also have to be wrangled. There's miniscule camera movements, which I don't think can be entirely eliminated; defocusing should be done in the direction that yields soft bokeh, and of course mirror lockup is a must. Dust in the air is a source of error I have no idea how to measure, and no idea of its significance; the only way to know would be to do an experiment both in a vacuum and at normal air pressure, with everything else equal. Variations in thermal noise are an issue if dim lighting is used (i.e., long exposures); using a flash as lighting may solve this, but it may also complicate things (there may be vertical variances in lighting due to a DSLR's focal plane shutter and possibly imperfect flash sync). Note that thermal noise is fine and dandy if it stays constant; excluding hot/dead pixels, it is just another source of shot noise which is indistinguishable from photon shot noise as far as my experiment is concerned.
There's a slight snag in my calculations, at the left edge of the graph of gain. Currently I have to input an "initial gain" to get it to come out nicely, but I doubt that this initial gain is accurate. Getting deeper shadows into the photographed scene (or intentionally varying the exposure) may improve this, but it may also require modifications in my model. In the meantime, I can only approximate the read noise in electrons, by using the average gain a little ways into the graph rather than at the beginning of it.
Also, I need to implement some curve-fitting to handle the near-saturation levels. As it is now, my graph cuts off there. This will of course also be necessary for Nikon cameras, which subtract the bias level resulting in some shadow clipping (do all Nikons do this?) I was actually hoping my new Canon would not do any highlight clipping but would instead allow the sensor to saturate naturally. Sigma Foveons do this, and as a result their RAWs have some noise in the highlights. That'd be great for my experiment.
For the preliminary Canon 450D graph I've posted, I took a subset of 36 contiguous frames from a series of 136 frames to get good data; something must have changed over time, as the full set of frames had much more variance. My room was lit with incandescent lighting — the frames were 10 second exposures; the level of thermal noise may have changed over time. Also, I defocused the lens in what I now realize was the wrong direction, giving hard bokeh. Nevertheless, the subset of 36 frames appear to be giving good data.
I tried a series of flash shots (off-camera external bounce flash), and while the shape of that graph is similar, it is a bit steeper (enough to change the projected saturation point from about 32000 electrons to about 31000). I'm not sure why, but I suspect that at 1/200 second flash sync, the shutter curtains may not be entirely clear at the moment the flash fires, resulting in slight vertical variations in sensor illumination.
Despite the complications described above, this experiment works as-is, giving interesting results. If properly interpreted, the bumps in the graph of gain may provide an intimate look into the inner workings of cameras' ADC units — I'd be extremely curious to see where the bumps are on other cameras (especially comparing ADCs of differing bit depth), and whether they all have such a flat gain graph (unlike my old Olympus E-20).
Who wants to do this experiment on your own RAW-capable camera(s)? I'm quite willing to share my source code and calculations, and go into more depth in my explanation. (I hope this post isn't too long already!) Comments and questions are quite welcome. Also, I'll bet some of you may have ideas on how to improve/refine the experiment and methodology!
EDIT: Emil, I just realized that you are both the topic-starter and the writer of the article! It's awesome to have you here on the forum, and I would really appreciate your input on my experiment.