400 Leica photographers agree: we love CCD!

[BerndReini]

Advertisement (gone after registration)

On the whole, I think dynamic range is extremely useful ... not everything is done under ideal lighting.

 

You have more dynamic range with the M9 than with slide film. Dynamic range is useful, but again what happens within the five stops that your usual subject has is the most important. I work in Hollywood and all the time I see people taking pictures on the south side of Hollywood Blvd. standing in the shade with the Chinese Theatre in full sun in the background. This will NEVER be a good picture, no matter how much dynamic range your camera can capture, simply because the emphasis of the lighting is completely wrong. The M9 separates mid-tones very nicely, and I believe this is what people love about their CCD cameras at base ISO.

[TonyField]

You have more dynamic range with the M9 than with slide film. Dynamic range is useful, but again what happens within the five stops that your usual subject has is the most important. I work in Hollywood and all the time I see people taking pictures on the south side of Hollywood Blvd. standing in the shade with the Chinese Theatre in full sun in the background. This will NEVER be a good picture, no matter how much dynamic range your camera can capture, simply because the emphasis of the lighting is completely wrong. The M9 separates mid-tones very nicely, and I believe this is what people love about their CCD cameras at base ISO.

 

I very seldom am interest in the Chinese theatre interpretation of dynamic range. What I am more interested in is how to record highlights in the way that negative film (not slides) with it's characteristic curve handled the shadows and highlights.The highlights "roll off" gracefully and you can easily recover detail in the darkroom.. With digital, you do not have this luxury -- being more like slide film With more dynamic range of the sensor, you can induce this roll-off easily.

 

Fortunately, Photoshop(ACR)/Lightroom now have reasonable control of most of the bit depth of the sensor and some of this "roll off" can be accomplished even with the M9 -- however I have run into many situations where 3,6, or 9 DB additional dynamic range would be very useful ... for example, recording the snow-capped mountains on a lovely spring day or a well tanned blond.wearing aggressive clothing with strong dark and light textures on a sunny day.

 

With the present implementation of the M9 CCD, the shooting rule to follow is "expose for the highlights and pray that the dynamic range will accommodate the shadows --- or expose for the shadows and to hell with the highlights"... just like shooting slides. Wider dynamic range will mitigate this problem.

 

Of course, for nominal shooting under reasonable conditions, the M9 works extremely well at base ISO.

Edited February 16, 2015 by TonyField

[BerndReini]

And that roll-off is no prettier with CMOS based cameras like the M240, and that's the problem.

[TonyField]

You miss the point ... with the extra dynamic range you can, in ACR/Photoshop be able to induce the equivalent of the H&D curve rolloff and achieve highlight detail with separation.

[BerndReini]

Some people believe that the pleasant look of the M9 at base ISO is a result of its more limited dynamic range. Whatever the reason, a lot of people seem to like it, and yet others are trying to relocate that look in Photoshop with superior sensors.

 

I own a medium format back with a CCD sensor, and I have looked at a lot of photographs from the new Pentax medium format camera with a CMOS sensor. I have yet to see a photograph from the new Pentax that suits my taste as much as the results from older Phase One backs, and Michael Reichmann is certainly someone who knows how to get good results in post-processing. It all comes down to a matter of taste.

 

If you don't clip your highlights, you get very nice highlight transitions from the M9 and even better from the MM.

[TonyField]

...

If you don't clip your highlights, you get very nice highlight transitions from the M9 and even better from the MM.

 

Exactly the point !!! If you have the DR, you don't clip the highlights and can pull them appropriately into the image.

[jaapv]

Advertisement (gone after registration)

A slightly diffuse point, though. One must differentiate between specular highlights and highlights with content. There is no dynamic range that can get detail into specular highlights or for instance the sun if you include it in your image - nor is there a need.

 

"Normal" highlights - that is obvious, and in regular situations it can easily be encompassed in the present-day DR of the better digital cameras - which, btw is similar to film and far better than slide film-

Only in more extreme situations like snow scenes or contre-jour do we need more DR and it would be welcome, I agree. However, I sometimes get the impression that it is also a wish inspired by sloppy exposing.

[Lenshacker]

Does it really? or does IR affect the test results by brightening the shadows with the M8?.

 

I was surprised as well. I'm reading the Data Sheets, these are numbers with the IR cover glass in place- but measurements of the sensor, not the camera.

 

From the Manufacture's Data Sheet for the KAF-18500 (M9 sensor):

Saturation Signal 42 ke-

Linear Dynamic Range (f = 24 MHz, T = 60 °C) 68.1 dB

 

For the KAF-10500 (M8) Sensor:

Saturation Signal 60 ke-

Linear Dynamic Range (f = 24 MHz, T = 40 °C) 71.5dB

 

The Temperatures given are different for the two, sensors warm up as they get read-out and there are more pixels to read out on the M9 sensor. But it is difficult to know why the manufacturer chose to change the temperature for the spec. I am speculating that the sensor had to be thinned down due to the geometry of a full-frame sensor for an M-Mount camera. The other sensors in the 6.8um family have Saturation signals close to the M8 sensor.

 

I spent much of the 1980's writing image processing code, including code to convert data from digital infrared imagers to radiometrically calibrated data. The Physicist that I worked with set the saturation count of the sensor for the Sahara and the "Black-Level" for the North Pole, 15-bit pixels with ~1 bit of noise. He hated to see clipped data. He passed away recently, was the quintessential Scientist.

Edited February 16, 2015 by Lenshacker

[NB23]

So, any news from leica regarding their fear of losing 1 or 2 serious customers (if not ZERO) if they don't start producing a ccd camera involving a few millions of dollars in R&D?

 

Curious.

[Lenshacker]

Is anyone taking bets that Leica will be the first to market a Full-Frame BSI Sensor?

 

I believe that their need to do so is greater than those of other manufacturers.

 

Just to add- I still love CCD's. I have a vested interest. It's hard not to be unbiased.

Edited February 16, 2015 by Lenshacker

[BerndReini]

So, any news from leica regarding their fear of losing 1 or 2 serious customers (if not ZERO) if they don't start producing a ccd camera involving a few millions of dollars in R&D?

 

No, not really, but I bet you a lot of fear about competing with companies like Sony and other CMOS sensor manufacturers on their own turf considering their quality and discount pricing.

[jaapv]

Not really . Those are by far not the only companies in the world. As everybody knows Leica explored the market and came up with CMOSIS, there are others European companies like Dalsa, etc. Absolutely no need to look over their shoulder.

Even Sony has no problem supplying competitors with sensors, like Nikon or Leica.

It is clearly an open and competitive market.

[Lenshacker]

I'm repeating myself but- the engineers at CMOSIS designed the 14MPixel full-frame CMOS sensor for the Kodak DCS Pro 14n (later called SLR/n) and SLR/c over 10 years ago. Kodak selected them for making the sensor for their full-frame camera. The company that the CMOSIS engineers worked at previously was call "FillFactory".

Edited February 16, 2015 by Lenshacker

[Rick]

I believe there is zero chance that you will see Leica go back to CCD technology. They do not see CCD as the future of cameras at Leica. They view a pixel as just a way to convert photons to electrons and the "look" everyone thinks they see is in the processing.

 

 

"Many people think there is a big difference in the touch and feel and the look and feel of the CCD vs CMOS. We think a pixel just renders light or transforms light into electricity. And the look and feel is done in the image processing. On the other hand, the CMOS sensors have a lot of advantages such as video and live view and we therefore think that the CMOS have the future at Leica." - Stefen Daniel, February, 2014

 

 

I hope this answers the question regarding the likelihood of Leica producing a CCD based camera.

 

Rick

[jaapv]

I'm repeating myself but- the engineers at CMOSIS designed the 14MPixel full-frame CMOS sensor for the Kodak DCS Pro 14n (later called SLR/n) and SLR/c over 10 years ago. Kodak selected them for making the sensor for their full-frame camera. The company that the CMOSIS engineers worked at previously was call "FillFactory".

Err... no. CMOSIS is not a rename. It is a company founded in 2008 by a group of people that are former employees of Fill Factory. I doubt that they own any of the Fill Factory patents, for instance. Everybody, except fresh from school, worked somewhere else before.

 

http://www.cmosis.com/news/press_releases/new_supplier_of_image_sensor_solutions_created_cmosis

Edited February 17, 2015 by jaapv

[Lenshacker]

http://www.leica-microsystems.com/fileadmin/downloads/Leica%20DFC550/Brochures/Leica_DFC550_Brochure_EN.pdf

 

You made me look. Now I want one. Leica Microsystems seems to have put some work into their new generation CCD cameras.

 

Nikon has a Monochrome version of the sensor used in the Df and the D4- used in a microscope camera.

[jaapv]

You want a 2/3rd 1.4 MP sensor that gets its 12.5 MP resolution from multiple partial exposures, and has to be cooled by Nitrogen as well?

Edited February 17, 2015 by jaapv

[Lenshacker]

Err... no. CMOSIS is not a rename. It is a company founded in 2008 by a group of people that are former employees of Fill Factory. I doubt that they own any of the Fill Factory patents, for instance. Everybody, except fresh from school, worked somewhere else before.

 

New supplier of image sensor solutions created: CMOSIS - CMOSIS - CMOS Image Sensors

 

Again- I stated that the engineers at CMOSIS worked at FillFactory. FillFactory was bought by Cypress, which was bought by On Technology. On Technology recently acquired Kodak's Imaging Division, Truesense. The engineers that were formerly at Fill Factory founded CMOSIS. I knew that CMOSIS is a new company, founded by the engineers that left Fill Factory after the acquisition. Whether or not they owned the patent, as any engineer that has worked on such a project they bring the expertise with them.

 

The 14MPixel CMOS array used in the Kodak camera is an analog device, did not have onboard A/D converters. I did not think they owned the patents, but they certainly carried the knowledge and experience of engineering the sensor. The part number is CYII4SM014KAA-GEC and the main patent mentioned in the sensor data sheet that covers it is US

patent No. 6,225,670.

 

Patent US6225670 - Detector for electromagnetic radiation, pixel structure with high ... - Google Patents

 

I downloaded the sensor long sheet years ago. It uses the same Schott S8612 IR cover glass used in the KAF-18500.

Edited February 17, 2015 by Lenshacker

[Lenshacker]

You want a 2/3rd 1.4 MP sensor that gets its 12.5 MP resolution from multiple partial exposures, and has to be cooled by Nitrogen as well?

 

I work in a Research Lab.

 

I was using digital cameras that were cooled using liquid nitrogen 35 years ago. We have a very large supply of liquid nitrogen for the building.

 

"In addition to all common microscope

modes, such as incidental and transmitted

light, the Leica DFC550 digital camera is

ideal for weak fluorescence and poorly

illuminated samples. With combined

cooling from a Peltier element and fan, the

camera’s CCD sensor provides perfect

image and color quality even from long

exposures. The 2/3" CCD Sensor is encapsulated

in a sealed nitrogen chamber to

avoid condensation on the optics, even

with difficult-to-image samples."

 

But this particular camera does not need that- they are just sealing the CCD in a Nitrogen chamber to isolate it. It uses a TEC for cooling. I need a new scope camera. Nikon makes a 16MPixel monochrome scope camera, a monochrome version of the sensor in my Df. They are much cheaper than 1MPixel cameras were 20 years ago. These things used to be expensive.

Edited February 17, 2015 by Lenshacker

[Advertisement]

Hallo,

Look here: Leica M

[yeeper]

Expose the M9 or the MM well in good light, and you will get fantastic results.

 

Yup. Just like Velvia. But I love versatility: Meaning low-light ability. I think CMOS is great. I didn't buy into Leica digital until the M came out.

 

But neither sensitivity or resolution is more important than light and composition.