dosdan

dosdan

Lives in Brisbane
Joined on Dec 17, 2007

Comments

Total: 89, showing: 1 – 20
« First‹ Previous12345Next ›Last »
On Let me try to address that... article (59 comments in total)

The comparison of the noise of the D810 and the A7S in the 2nd page of the article was not a good choice, as the A7S is likely to be using Aptina's (now ON Semiconductors) DR-Pix technology.

http://home.comcast.net/~NikonD70/GeneralTopics/Sensors_&_Raw/Sony_A7S_DR-Pix_Read_Noise.htm

There is a step in the read noise between ISO1600 and IS2000+, with the higher ISOs show less RN than expected. DR-Pix technology allows the changing of the gain in the sensor, so the sensor designer's trade-off between Full Well Capacity, Conversion Gain & Read Noise can be altered in-camera.

A better pixel-size comparison choice would have been between the Pentax K-5 (16MP) & K-3 (24MP), both APS-C cameras.

http://www.dxomark.com/Cameras/Compare/Side-by-side/Pentax-K-3-versus-Pentax-K5___914_676

On the DR graph you can use the Screen tab to seen the pixel-level performance, while the Print tab shows the result using the same output size e.g. same print size.

Dan.

Dan.

Direct link | Posted on May 17, 2015 at 22:32 UTC as 2nd comment
On Sources of noise part two: Electronic Noise article (229 comments in total)
In reply to:

coyot3: I tested on my nikon d3100 and the results are at iso 100 1" f4 underexposing 4 ev and later recovering in Darktable(linux) give less noise than iso 3200 1/30 f4 no exposure changes.

I dont know if i understand right the article xD im going to re read ._.

¿Some thoughts on my results?

You'll find an ISO100 + 4-stops boost VS ISO1600, using a K-5, here:

http://www.pentaxforums.com/forums/61-post-processing-articles/234154-investigation-reported-colour-change-when-ev-boosting.html

The exposure and sensor was the same in both cases: same scene luminance, same shutter speed & same f-stop. So the shots have the same amount of shot noise. But the ISO100 image was sent to the ADC at 1/16th of the level of the ISO1600 shot, so it was "under-brightened" in the stored image file.

Dan.

Direct link | Posted on May 15, 2015 at 03:38 UTC
On Sources of noise part two: Electronic Noise article (229 comments in total)
In reply to:

DuncanDovovan: Richard / Rishi: What I would really love to see is a shot noise simulation. Rishi looks to be the guy who can calculate the number of photons that a pixel gets.

From the number of photons caught, couldn't you calculate the S/R ratio?

And from the S/N ratio, couldn't you visualise/simulate the shot noise how it would look like in a 100% crop (pixel level) and a crop that would show the scene at the same scale (more pixels scaled down to the same magnification)?

Wouldn't it be possible to visualise the shot noise, other upstream noise and downstream noise separately that way?

I would love to see (as in picture crops / not tables) and compare the shot noise, upstream and downstream noise components under several conditions. Like higher vs lower ISO. Shorter vs longer exposures. Overexposure and compensating, underexposure and compensating vs normal exposure.

And all of that for 5 different tonal values from dark to bright and in different situations from cloudy to bright daylight.

"I would expect the bright areas to have less noise."

No, the bright areas have more shot noise because they are the result of more photons hitting the sensor in this area. (Shot noise increases at the sq root of the number of photons.) It's just that the Signal-to-Noise Ratio will be better in these areas, so you probably won't notice the extra noise.

Two examples:
A: 1,000,000 photons.
B: 10,000 photons.

A. Shot noise = sq-root(1,000,000) = 1,000.
SNR = 1,000,000/1,0000 = 1,000:1

B: Shot noise = sq-root(10,000) = 100.
SNR= 10,000/100 = 100:1

BTW, SNR is sq-root(number of photons), but I've worked it out here to show the amount of the shot noise.

Dan.

Direct link | Posted on May 14, 2015 at 22:36 UTC
On Sources of noise part two: Electronic Noise article (229 comments in total)

www.sensorgen.info has been mentioned a few times in these comments. It is a site run by Prof. Bob Newman, which shows the Full-Well Capacity of sensor pixels and Total Read Noise, as calculated from DxOMark measurements.

I'm interested in examining the Total Read Noise, input-referenced in -e (photo-electron charges), curve-fitted to estimate the Sensor read noise (upstream of the PGA) and the ADC noise (downstream).

Here's a sheet that shows a few cameras.

https://www.dropbox.com/s/1h67vzosluaxz8x/Noise%20Components%20Contribution%20v4%20.xls?dl=0

There are more cameras on the RN (Read Noise) Contributions tab, and it's easy to use it as a template to add more cameras.

SNR (dB) is my calculated value. You can compare it against SNR (DxO) to see how well the curve-fit performs. On the RN Contributions page you also see this in the "Diff square" value.

The 16x Ratio compares the Total RN at base ISO & 16x base ISO. The closer this is to 1, the more "ISOless" the camera.

Dan.

Direct link | Posted on May 14, 2015 at 21:31 UTC as 25th comment
On Sources of noise part two: Electronic Noise article (229 comments in total)
In reply to:

DuncanDovovan: Richard / Rishi: What I would really love to see is a shot noise simulation. Rishi looks to be the guy who can calculate the number of photons that a pixel gets.

From the number of photons caught, couldn't you calculate the S/R ratio?

And from the S/N ratio, couldn't you visualise/simulate the shot noise how it would look like in a 100% crop (pixel level) and a crop that would show the scene at the same scale (more pixels scaled down to the same magnification)?

Wouldn't it be possible to visualise the shot noise, other upstream noise and downstream noise separately that way?

I would love to see (as in picture crops / not tables) and compare the shot noise, upstream and downstream noise components under several conditions. Like higher vs lower ISO. Shorter vs longer exposures. Overexposure and compensating, underexposure and compensating vs normal exposure.

And all of that for 5 different tonal values from dark to bright and in different situations from cloudy to bright daylight.

Duncan, there is already a shot noise simulation in Wikipedia:

https://en.wikipedia.org/wiki/Shot_noise

Dan.

Direct link | Posted on May 14, 2015 at 13:04 UTC
On Sources of noise part two: Electronic Noise article (229 comments in total)
In reply to:

SolidMetal: Great article! Though I still cant really understand why high ISO images have less dynamic range. That orange line which is overamplified in the picture wouldnt be out of the raw file even if its straight not curved? I mean its still brighter than the highlights marked in yellow on the low ISO picture.

SolidMetal, 6 stops less DR if there was no DR reduction at low ISO due to the ADC contributing some noise. But the ADC contributes either a little noise (Sony, Nikon, Pentax) or moderate noise (Canon). This means that the straight-line relationship of 1 stop less DR for each stop increase in ISO is usually not reached until the mid or high ISO ranges.

See my DR graph and DR reduction explanation in:

http://www.dpreview.com/forums/post/55736829

That's the 2nd part of an explanation of the role of ISO Sensitivity control in a digital camera. The 1st part is here:

http://www.dpreview.com/forums/post/55734917

Dan.

Direct link | Posted on May 13, 2015 at 22:29 UTC
On Sources of noise part two: Electronic Noise article (229 comments in total)
In reply to:

SolidMetal: Great article! Though I still cant really understand why high ISO images have less dynamic range. That orange line which is overamplified in the picture wouldnt be out of the raw file even if its straight not curved? I mean its still brighter than the highlights marked in yellow on the low ISO picture.

SolidMetal, there are two ways that hightlights get blown:'

1. Sensor pixels saturating (filling up to overflowing).
2 ADC clipping (running out of bits in a 12-bit/14-bit digital number).

The gain stage between the sensor & the ADC is set up so, at base ISO, the sensor saturates just before/at the same time as the ADC clips. This gives max. DR.

With weaker & weaker exposures, less photons are captured, so the sensor pixels are "less well-filled". So once you start boosting the gain to compensate for weak exposures and make the rendered image brighter, it becomes less & less likely that any pixels will have saturated.

But with more & more gain, any relatively high pixel values are now more likely to clip the ADC. From ISO100->ISO1600 is 16x gain or 4 stops. If a pixel saturates at 48,000e- (photo-electrons) at ISO100 and the ADC also just clips at this level, then at ISO1600 the pixel will only reach 3,000e- before the ADC clips instead.

So the DR decreases with increasing ISO.

Direct link | Posted on May 13, 2015 at 20:28 UTC
In reply to:

Muqdad: If RC of L.Rentals is so insistingly positive about Canon's response and reaction, and the general impression on Canon's customer service is almost of the very best out there among manufacturers, shouldn't an honest will to be helpful to Canon in this matter made him more willing to give them a supposedly deserved chance to show their good support action expected by R. Cicala himself?

Rather, he went trigger-happy and did this HUGE damage to the company we applaud for caring for and actually respecting their customers?

I can only think he believed he may lose the chance to be the one who first tells about it. This is exactly the perfect abuse of authority.

"The spots may be part of the fabrication, e.g. a film coating that is required to limit contact between the laminates, and once the laminates relax for a period after fabrication, the spots go away. So in that case this is not a problem"

Interesting conjecture. So the cameras come from the factory under-matured and should not be used for serious work until they've had time to age?

Haven't heard that one before. Isn't the internet wonderful?

Dan.

Direct link | Posted on May 2, 2015 at 03:37 UTC
In reply to:

Muqdad: If RC of L.Rentals is so insistingly positive about Canon's response and reaction, and the general impression on Canon's customer service is almost of the very best out there among manufacturers, shouldn't an honest will to be helpful to Canon in this matter made him more willing to give them a supposedly deserved chance to show their good support action expected by R. Cicala himself?

Rather, he went trigger-happy and did this HUGE damage to the company we applaud for caring for and actually respecting their customers?

I can only think he believed he may lose the chance to be the one who first tells about it. This is exactly the perfect abuse of authority.

I don't see it the way. If an individual has this problem they can get fobbed off with a line like "you must have done something to the camera" (e.g. stored it inside a hot car). But a rental company with a bigger sampling goes public about it so, if others have an issue, they're less likely to be fobbed off.

Manufacturers have to take the heat when their Quality Assurance process fails. Why should companies and individuals who buy their equipment worry about the manufacturer's reputation? That's a fanboi reaction.

The manufacturers spend enough telling us how wonderful their products are and how our lives will be better if we buy them. Some negative feedback from owners helps to rebalance this a little in the direction of reality.

Dan.

Direct link | Posted on May 1, 2015 at 20:57 UTC
In reply to:

D Bowcut: I have posted an open-sky shot in the Gallery at f22, ISO 150, from my new T6s: This is at 100%, but re-sized to fit the Galleries rqmt.

I can't tell if the problem exists here or if this reveals only expected pixelization? My camera s/n is 022031000385.

Otherwise, camera is running OK.

Lock up the mirror for cleaning and look at the top of the sensor with a loupe or magnifying glass.

Dan.

Direct link | Posted on May 1, 2015 at 20:44 UTC

This looks like a defect in the "toppings" (anti-refection layer/IR-filter/ perhaps even microlenses) which are often added by specialist firms. Don't know if Canon adds the toppings in-house or sends their sensors outside for these. (They make their own DSLR sensors.)

You wonder why this type of obvious defect wasn't picked up in the visual QA stage? Perhaps it's a chemical reaction that takes some time to form bubbles?

Dan.

Direct link | Posted on May 1, 2015 at 03:08 UTC as 126th comment
In reply to:

Earth Art: Thermal conditions make a huge difference in shot noise as well. Big difference shooting at night in 30 deg F and 80 deg F.

That's not shot noise. Rather it's dark current & thermal blooming if parts of the sensor (the columnar ADCs situated on the edge) are operated in a long enough exposure.

Shot Noise aka Photonic Noise is related solely to the number of photons in the capture. It's an intrinsic property of light, so it can't be avoided.

Dan.

Direct link | Posted on Apr 28, 2015 at 03:08 UTC
In reply to:

thk0: "There are three factors that affect how much light is available for your sensor to capture: your shutter speed, the size of your aperture (not f-number) and the size of your sensor."

Yes, f-number. If you use the physical aperture size, then the sensor size does not matter (assuming the same sensor coverage of the light circle). The amount of light passing through the lens, and hence "available for your sensor", only depends on the area of the entrance pupil, transmission efficiency and the duration of the exposure. If you use f-number, then the physical aperture scales with the sensor size and sensor size becomes part of the equation.

(edits for typos)

Another way of stating this: At the same exposure level and similar sensor technology, the shot noise SNR is proportional to the sq root of the sensor area. So 4x bigger sensor = 2x better shot noise SNR.

Dan.

Direct link | Posted on Apr 27, 2015 at 22:42 UTC
In reply to:

thk0: "There are three factors that affect how much light is available for your sensor to capture: your shutter speed, the size of your aperture (not f-number) and the size of your sensor."

Yes, f-number. If you use the physical aperture size, then the sensor size does not matter (assuming the same sensor coverage of the light circle). The amount of light passing through the lens, and hence "available for your sensor", only depends on the area of the entrance pupil, transmission efficiency and the duration of the exposure. If you use f-number, then the physical aperture scales with the sensor size and sensor size becomes part of the equation.

(edits for typos)

"There are three factors that affect how much light is available for your sensor to capture: your shutter speed, the size of your aperture (not f-number) and the size of your sensor."

The "Exposure Triangle" is Shutter Speed, F/number, Scene Luminance. That's why you use a flash or a light reflecting panel: to increase the scene luminance component of exposure. These 3 determine the exposure (the photonic density captured per mm2 during the exposure period). So if these 3 remain the same, with larger sensors you get the same exposure, but more Total Light (more photons) being captured. For the same exposure, there's more photons collected, so there's also more shot noise photons in a bigger sensor capture. But, since the Shot Noise SNR is proportional to the sq root of the number of photons, it's still better with a bigger sensor.

Example. Small sensor: 10,000 photons captured, SNR = 10,000/100= 100:1

2.3x bigger sensor: 23,000 photons captured: SNR = 23,000/151.7 = 151.7:1

Dan

Direct link | Posted on Apr 27, 2015 at 19:45 UTC
In reply to:

dosdan: "we wish Adobe went a bit further and allowed for simple image averaging with Raw DNG output."

Photo Acute has had this ability for years (2007), as well as a lot more. The parent company, Almalence, has been around since 2005.

Dan.

Here's a stacking test of 14 handheld K20D raw files (the max. I could do in a burst before the shot buffer filled up), each taken at ISO3200, from 2009:

http://www.pentaxforums.com/forums/32-digital-processing-software-printing/74707-photoacute-studio-nr-stacking-test.html

Direct link | Posted on Apr 21, 2015 at 21:39 UTC

"we wish Adobe went a bit further and allowed for simple image averaging with Raw DNG output."

Photo Acute has had this ability for years (2007), as well as a lot more. The parent company, Almalence, has been around since 2005.

Dan.

Direct link | Posted on Apr 21, 2015 at 21:22 UTC as 97th comment | 2 replies

Well done, Ricoh. http://www.ricoh.com/csr/savethememory/ is a good read.

Anyone interested in learning about Ricoh's ethical approach to the environment, society and governance can find a list of topics here:

http://www.ricoh.com/sustainability/

Dan.

Direct link | Posted on Apr 16, 2015 at 02:03 UTC as 12th comment

DPR, while the ruckus about the omitting of "Ricoh" from the original title of this article seems excessive to you guys, you may have noticed in the comments about your Samsung NX-1 review, that the topic of apparently unethical behaviour from Samsung was raised.

I believe the responses in the current thread, the thread "Making 'Art': We go inside Sigma's lens factory" and in the NX-1 review thread indicate an increase sensitivity in your readership to both perceived bias and to manufacturers' ethical behaviour.

Favourable comment was made by your readers to Sigma's attitude to its workforce.

Ricoh seems to have earned a good reputation in ethical behaviour:

http://www.prnewswire.com/news-releases/ricoh-named-as-a-2015-worlds-most-ethical-company-by-the-ethisphere-institute-for-the-6th-time-300048883.html

While cynicism abounds in journalism, I think your readership really appreciates it when you highlight the ethical reputation and behaviour of manufacturers.

Dan.

Direct link | Posted on Apr 16, 2015 at 00:57 UTC as 14th comment
On Samsung NX1 Review preview (1229 comments in total)
In reply to:

ttran88: Lots of praises here and I'm sure the camera is great but with a company like Samsung and its rich history of "astroturfing", which praises are real?

http://www.theverge.com/2013/10/24/5023658/samsung-fined-340000-for-posting-negative-htc-reviews

And as recent as last week..

http://m.theinquirer.net/inquirer/news/2403094/samsung-allegedly-hired-500-fans-for-galaxy-s6-launch-in-china

I find this company to be extremely unethical and I personally wouldn't want to support this behavior with my money.

@ttran88: "I find this company to be extremely unethical and I personally wouldn't want to support this behavior with my money."

Fair enough. But even if you don't want to support the Samsung company, I think we can still agree that the release of the NX1 can be seen in a positive light as it should rouse the marketplace up and hopefully improve the offerings from the other manufacturers.

Dan.

Direct link | Posted on Apr 10, 2015 at 23:43 UTC
On Samsung NX1 Review preview (1229 comments in total)
In reply to:

tom1234567: comments on poor JPEG quality, was the camera tested with the update mentioned at the end of the shooting experience

As I understand it, Samsung's NR in this camera is similar to Sony's: Area-specific NR in some of its models, rather than Global same-level NR. Is this correct?

Both brands are criticised for the unnatural-looking results in high-ISO JPEGS. So perhaps this type of NR needs more refinement.

Dan.

Direct link | Posted on Apr 10, 2015 at 19:59 UTC
Total: 89, showing: 1 – 20
« First‹ Previous12345Next ›Last »