Why don't we get the GOOD data anymore?

I was reading an old story on this site and look at this amazing chart! It lists everything one needs to know about a sensor, no testing needed. Why don't we get this kind of data anymore? Or do we and no one looks at it? I mean, I don't even know what Dark Current Doubling Temperature is, but I'm excited to see it here. And is this level of data useful?

Kodak always offered full technical data on their films too.

I think the companies that make digital cameras now are needlessly secretive about their products. Even the instruction manuals are very limited, as though the question was "Why do you need to know this?"

SmilerGrogan said:

I was reading an old story on this site and look at this amazing chart! It lists everything one needs to know about a sensor, no testing needed. Why don't we get this kind of data anymore? Or do we and no one looks at it? I mean, I don't even know what Dark Current Doubling Temperature is, but I'm excited to see it here. And is this level of data useful?

Click to expand...

That table is just a summary, and the "output sensitivity" should be 25 micro-volts/electron. More details - including the spectral transmission of Bayer and IR blocking filters, and angular acceptance of micro-lenses here: http://www.kodak.com/ek/uploadedFil...utions/Datasheets(pdfs)/KAF-10500LongSpec.pdf

Another Kodak CCD sensor, the KAF 8300 (now made by ON Semiconductor) was used in some Olympus cameras, and is still popular for astro-photography, where dark current doubling temperature is an important parameter. Increasing the sensor temperature by the "doubling temperature" doubles the dark leakage current. Conversely, cooling the sensor reduces leakage current which would otherwise add noise during long exposures - as discussed here: http://www.qsimaging.com/683-overview.html

--
Alan Robinson

SmilerGrogan said:

I was reading an old story on this site and look at this amazing chart! It lists everything one needs to know about a sensor, no testing needed. Why don't we get this kind of data anymore? Or do we and no one looks at it? I mean, I don't even know what Dark Current Doubling Temperature is, but I'm excited to see it here. And is this level of data useful?

Click to expand...

Wow, that's a huge CCD. 6.8 square meters for a single pixel.

ml6 said:

SmilerGrogan said:

I was reading an old story on this site and look at this amazing chart! It lists everything one needs to know about a sensor, no testing needed. Why don't we get this kind of data anymore? Or do we and no one looks at it? I mean, I don't even know what Dark Current Doubling Temperature is, but I'm excited to see it here. And is this level of data useful?

Click to expand...

Wow, that's a huge CCD. 6.8 square meters for a single pixel.

Click to expand...

I think they meant um, not m.

Kodak's data:







But, to get to the OP's point, these are specs for a sensor, not for a camera. I beleive that sensors today are spec'd in at least this much detail. The people who spend their time poring over the spec sheets are, presumably, the camera designers.

This sensor was used in the Leica M8. Leica's spec's for that camera didn't look anything like this. BTW, the M8 was not a full frame camera.

Jim

--
http://blog.kasson.com

JimKasson said:

The people who spend their time poring over the spec sheets are, presumably, the camera designers.

Click to expand...

It seems silly that the camera makers don't provide numerical data that would put an end to all these internet discussions and reviews that are based on non-engineers trying to understand and explain topics that are miles over their heads.

Didn't know it was only sensor specs. But let's say you could get this data for the sensor-processor combo. Or is it already available and we just don't find it because it's only on camera makers' business to business sites?

SmilerGrogan said:

SmilerGrogan said:

The people who spend their time poring over the spec sheets are, presumably, the camera designers.

Click to expand...

It seems silly that the camera makers don't provide numerical data that would put an end to all these internet discussions and reviews that are based on non-engineers trying to understand and explain topics that are miles over their heads.

Didn't know it was only sensor specs. But let's say you could get this data for the sensor-processor combo. Or is it already available and we just don't find it because it's only on camera makers' business to business sites?

Click to expand...

And by today's standards, those specs are not very good.

15 electron read noise, but today's CMOS are under 2.

Dark current at 40 C = 4 millivolt (/ 25 microV/e-) = 160 electrons/second, then at around 8 C, that would be 2^5 less, or 160/32 = 5 electrons/second.

The 7D Mark II is on the order of 0.01 electron/second! (Many other current DSLRs are in the 0.1 to 0.2 electron/second), at ~8C.

And another important factor not on spec sheets is pattern noise levels.

JimKasson said:

But, to get to the OP's point, these are specs for a sensor, not for a camera. I beleive that sensors today are spec'd in at least this much detail.

Click to expand...

I think the underlying problem alluded to by the OP is that these specifications are generally not publicly available.

I should add that the same is true of lens specifications; Zeiss being one prominent exception.

Regards,

bclaff said:

JimKasson said:

But, to get to the OP's point, these are specs for a sensor, not for a camera. I beleive that sensors today are spec'd in at least this much detail.

Click to expand...

I think the underlying problem alluded to by the OP is that these specifications are generally not publicly available.

Click to expand...

I don't think the specs for many sensors are a secret. There are a host of exceptions.

Here's an example of a current datasheet: http://www.cmosis.com/products/product_detail/cmv50000

You can go here and get data for many sensors:


Here is a collection of data sheets from another manufacturer:


Here are specs for many Sony sensors:


Here is Omnivision's site:


With the inclusion of Canon's web site, we have the top five sensor manufactureres, with an aggregate market share of almost 80%.


It is true that the level of detail in the specs is not as detailed as one might want in the case of Samsung and Sony.

It is also true that the sensors in many cameras are missing from these public web sites.

And of course, sensors that are part of a vertically integerated supply chian have no public data sheets.

JimKasson said:

bclaff said:

JimKasson said:

But, to get to the OP's point, these are specs for a sensor, not for a camera. I beleive that sensors today are spec'd in at least this much detail.

Click to expand...

I think the underlying problem alluded to by the OP is that these specifications are generally not publicly available.

Click to expand...

I don't think the specs for many sensors are a secret. There are a host of exceptions.

...

Here are specs for many Sony sensors:

http://www.sony-semicon.co.jp/products_en/IS/sensor0/products/cmos.html

Click to expand...

Well, take Sony for example; those are all indusrial and surveillance sensors; not those used in our digital cameras.

Regards,

bclaff said:

JimKasson said:

bclaff said:

JimKasson said:

But, to get to the OP's point, these are specs for a sensor, not for a camera. I beleive that sensors today are spec'd in at least this much detail.

Click to expand...

I think the underlying problem alluded to by the OP is that these specifications are generally not publicly available.

Click to expand...

I don't think the specs for many sensors are a secret. There are a host of exceptions.

...

Here are specs for many Sony sensors:

http://www.sony-semicon.co.jp/products_en/IS/sensor0/products/cmos.html

Click to expand...

Well, take Sony for example; those are all indusrial and surveillance sensors; not those used in our digital cameras.

Click to expand...

Right you are. I think that many camera sensors are not off-the-shelf.

Jim

SmilerGrogan said:

SmilerGrogan said:

The people who spend their time poring over the spec sheets are, presumably, the camera designers.

Click to expand...

It seems silly that the camera makers don't provide numerical data that would put an end to all these internet discussions and reviews that are based on non-engineers trying to understand and explain topics that are miles over their heads.

Didn't know it was only sensor specs. But let's say you could get this data for the sensor-processor combo. Or is it already available and we just don't find it because it's only on camera makers' business to business sites?

Click to expand...

But the trouble is that providing numerical data to people who don't have the training to understand what those specs mean only creates more confusion. The internal working specifications for any complex consumer device are extremely detailed, but only of use to those building and maintaining the product. If you're a geek, you get off on the numbers, but if you're honest, they probably would be of some service for on-line arguments with other enthusiasts, but they would not help you take a better picture. Cameras are such complicated tools that the attention of the consumer (and of the manufacturer) should more appropriately be placed on how well the camera enables you to capture an image, and how pleasing that image is to you, and how easily that image can be shared and manipulated after capture. Thus, published specifications tend to be more about features and problems solved rather than hard numbers, which are difficult to translate into use-case competencies outside of a few special metrics.

The spec sheet you refer to is from the very early days of digital. I can think back to when CDs were brand new, exciting technology. Sony went on for pages with elegant cutaway drawings of mechanisms and we debated endlessly whether this or that reconstruction filter adequately preserved phase differences essential, we thought, to sound stage reproduction accuracy. Only a few short years later, no one cared...the technical problems had been largely solved and we got on with listening to the music.

The same is true in the camera industry. A sensor spec - or even a sensor and processor spec - is largely meaningless to the utility of a camera anymore. The ability to capture easily an image is, and so our geeking seems to concentrate on things like ergonomics, AF competence, and connectivity more than anything. Almost any modern sensing chain is supremely competent...it's workflow that now is of most interest...or at least it is to me.

I agree there is a limited audience for specifications at this level.
I do sometimes miss them as a sanity check for the measurements I perform.

Regards,

rnclark said:

And by today's standards, those specs are not very good.

15 electron read noise, but today's CMOS are under 2.

The 7D Mark II is on the order of 0.01 electron/second!

Click to expand...

I have to say that I doubt this number if it is at RT, and I think it is at least 100x higher.

Do you have some sort of reference for this data point?

(I saw your data point of 0.016e-/s at 10C so I mean corroborating measurements)


This is 15fA/cm^2 and is unlikely, even at 10C.

Eric Fossum said:

rnclark said:

And by today's standards, those specs are not very good.

15 electron read noise, but today's CMOS are under 2.

The 7D Mark II is on the order of 0.01 electron/second!

Click to expand...

I have to say that I doubt this number if it is at RT, and I think it is at least 100x higher.

Do you have some sort of reference for this data point?

(I saw your data point of 0.016e-/s at 10C so I mean corroborating measurements)

http://www.clarkvision.com/reviews/evaluation-canon-7dii/

This is 15fA/cm^2 and is unlikely, even at 10C.

Click to expand...

Yeah, my number is about 1.5e- (remember the y-axis is log2)



--
Bill ( Your trusted source for independent sensor data at http://www.photonstophotos.net )

By "Nobody" I mean 99.8% of consumers. Of course some of us care, and a subset are competent to understand the details of the technical information documents.

My suspicion is that unlike the old days where sensor makers OEM'd a sensor and the camera makers were integrators, the sensor maker is now producing a bespoke sensor for each camera maker, so publishing the specs is less important.

I also have a social-evolutionary theory that the Y-chromosome harbors genes which code for proteins that manipulate brain development in a way that we ( males ) care about these deep technical details rather than just pressing the shutter button and taking great pictures.

-- Bob
http://bob-o-rama.smugmug.com -- Photos
http://www.vimeo.com/boborama/videos -- Videos
http://blog.trafficshaper.com -- Blog

bclaff said:

Eric Fossum said:

rnclark said:

And by today's standards, those specs are not very good.

15 electron read noise, but today's CMOS are under 2.

The 7D Mark II is on the order of 0.01 electron/second!

Click to expand...

I have to say that I doubt this number if it is at RT, and I think it is at least 100x higher.

Do you have some sort of reference for this data point?

(I saw your data point of 0.016e-/s at 10C so I mean corroborating measurements)

http://www.clarkvision.com/reviews/evaluation-canon-7dii/

This is 15fA/cm^2 and is unlikely, even at 10C.

Click to expand...

Yeah, my number is about 1.5e- (remember the y-axis is log2)

Click to expand...

sorry, I was not clear. I am referring to the dark current number of 0.01e-/s as being unlikely. 1.5e- rms read noise sounds quite likely.

bclaff said:

Eric Fossum said:

rnclark said:

And by today's standards, those specs are not very good.

15 electron read noise, but today's CMOS are under 2.

The 7D Mark II is on the order of 0.01 electron/second!

Click to expand...

I have to say that I doubt this number if it is at RT, and I think it is at least 100x higher.

Do you have some sort of reference for this data point?

(I saw your data point of 0.016e-/s at 10C so I mean corroborating measurements)

http://www.clarkvision.com/reviews/evaluation-canon-7dii/

This is 15fA/cm^2 and is unlikely, even at 10C.

Click to expand...

Yeah, my number is about 1.5e- (remember the y-axis is log2)



--
Bill ( Your trusted source for independent sensor data at http://www.photonstophotos.net )

Click to expand...

Bill, your are citing read noise above. We are talking dark current.

Eric Fossum said:

bclaff said:

Eric Fossum said:

rnclark said:

And by today's standards, those specs are not very good.

15 electron read noise, but today's CMOS are under 2.

The 7D Mark II is on the order of 0.01 electron/second!

Click to expand...

I have to say that I doubt this number if it is at RT, and I think it is at least 100x higher.

Do you have some sort of reference for this data point?

(I saw your data point of 0.016e-/s at 10C so I mean corroborating measurements)

http://www.clarkvision.com/reviews/evaluation-canon-7dii/

This is 15fA/cm^2 and is unlikely, even at 10C.

Click to expand...

Yeah, my number is about 1.5e- (remember the y-axis is log2)

Click to expand...

sorry, I was not clear. I am referring to the dark current number of 0.01e-/s as being unlikely. 1.5e- rms read noise sounds quite likely.

Click to expand...

And I measured a low read noise of 1.6 electrons (per my above web page).

Eric Fossum said:

bclaff said:

Eric Fossum said:

rnclark said:

And by today's standards, those specs are not very good.

15 electron read noise, but today's CMOS are under 2.

The 7D Mark II is on the order of 0.01 electron/second!

Click to expand...

I have to say that I doubt this number if it is at RT, and I think it is at least 100x higher.

Do you have some sort of reference for this data point?

(I saw your data point of 0.016e-/s at 10C so I mean corroborating measurements)

http://www.clarkvision.com/reviews/evaluation-canon-7dii/

This is 15fA/cm^2 and is unlikely, even at 10C.

Click to expand...

Yeah, my number is about 1.5e- (remember the y-axis is log2)

Click to expand...

sorry, I was not clear. I am referring to the dark current number of 0.01e-/s as being unlikely. 1.5e- rms read noise sounds quite likely.

Click to expand...

My bad. I misread