Leica M9 FF-CCD corrosion - A most satisfactory conclusion

Indefinitely? At least until the sensor stock is depleted? Is this sensor still being manufactured?

If they can't replace sensor while it's covered by warranty they should replace camera, so easy. With comparable next model if current isn't in stock anymore. Customer pays, Leica delivers.
 
I really love the Nikkor glass and even have a beauty of a 105 ltm for my M cameras. I tried to find a clean 35 1.8 ltm but struck out. I think Nikkor glass is first class. The millinium 50 1.4 is one of if not the best all round 50 Ive ever used. It's every bit as sharp as the 50 asph summilux and smoother in tone. It's a real winner.

X-ray,

I was lucky to find a clean 35/1.8 in LTM. I agree that the Nikkor glass is special. On my Monochrom the 35/1.8 is kinda like a magic lens that offers the best contrast (midrange) that resembles medium format. My glass is nuclear hardened and displays a built in yellow filter of about a half stop of yellow. While not as sharp as modern Leica ASPH glass plenty sharp and detailed.

Cal
 
In my mind, I find the CCD vs CMOS bit hilarious, considering that CCD rendering in the M9 is a product of its limited dynamic range. In fact, as CCDs go by, the one in the M9 is far behind the best in CCD technology. CCDs used in scientific apparatus are far better than the one used in the M9.

People complain that pictures out from CMOS look "flat" but that's because the files record more color information and are far more malleable than the ones out from the M9. I could achieve similar "look" just by upping the contrast, which in a way, clamps the dynamic range.

Fidelity I guess is how you guage superiourity.

I think film is a product of its limited dynamic range, yet many like its rendering.

I offer no bias: I plan on keeping my Monochrom; and if the MM-240 rumor is true I will also get one. No doubt they are different, but to say one is inferior or superior due to how it handles its dynamic range I would not say. Highly subjective is it not?

Cal
 
Fidelity I guess is how you guage superiourity.

I think film is a product of its limited dynamic range, yet many like its rendering.

I offer no bias: I plan on keeping my Monochrom; and if the MM-240 rumor is true I will also get one. No doubt they are different, but to say one is inferior or superior due to how it handles its dynamic range I would not say. Highly subjective is it not?

Cal

We each have our own criterion for good.

IMO the M9 has a pretty sorry dynamic range compared to the current top end CMOS cameras like the D810 and D610 with over 14 stops. I'd guess the M9 might have 8-9 stops but that's a guess.

I wound up selling my Hasselblad digital because my D800 had over two stops better range. I do a lot of architectural work for commercial contractors and architectural firms. That extra range was quite a help when doing interiors.

As to dynamic range of film, many color neg films will give 14 stops and many B&W can do the same with proper development. It's color transparency that's has very limited range at 6 - 8 stops at best and Velvia has even less.

That snap in the image with the M9 comes from the contrast / limited dynamic range and internal processing. The thing about a broad dynamic range is you can compress it and get that snap but if you have limited range you can never expand it. The same thing is true with a contrasty film vs flat and a contrasty lens vs flat. Give me flat any day for more control of my image.
 
We each have our own criterion for good.

IMO the M9 has a pretty sorry dynamic range compared to the current top end CMOS cameras like the D810 and D610 with over 14 stops. I'd guess the M9 might have 8-9 stops but that's a guess.

I wound up selling my Hasselblad digital because my D800 had over two stops better range. I do a lot of architectural work for commercial contractors and architectural firms. That extra range was quite a help when doing interiors.

As to dynamic range of film, many color neg films will give 14 stops and many B&W can do the same with proper development. It's color transparency that's has very limited range at 6 - 8 stops at best and Velvia has even less.

That snap in the image with the M9 comes from the contrast / limited dynamic range and internal processing. The thing about a broad dynamic range is you can compress it and get that snap but if you have limited range you can never expand it. The same thing is true with a contrasty film vs flat and a contrasty lens vs flat. Give me flat any day for more control of my image.

X-ray,

Thanks for the detailed explaination.

I guess that I do have a bias after all because I like how the M9 renders a lot.

Cal
 
When my M9 was working I did like the look. It produced images that looked much like the Hasselblad. The Hasselblad has a considerably broader to al range and color depth. The color depth was second to none which comes from a true 16 bit capture. Honestly I don't know why manufactures of 35 format digicams don't capture in 16 bit. The difference is like watered down paint vs concentrated pigments. It's more than just saturation, there's something that's hard to put into words. And yes, there is a different look to a CCD capture vs a CMOS.
 
That is exaclty what I mean. The more we move on with digital the more trouble we seem to have, or maybe we are just discovering the start of the problem after a couple decades of digital sensors on the market.

But only for one sensor assembly design out of thousands. I am not aware of any other "trouble" or "problem". Since I switched to digital I have less trouble. To be fair I do not expect a sensor to last 20 years (although some certainly will).

I do agree that a digital body was not designed to function indefinitely. This is certainly different than a purely mechanical camera where it is possible for skilled craftsmen to manufacture replacement parts from scratch. Of course many excellent film cameras depend on electronic parts as well. For these the electronic parts don't last indefinitely either.
 
Which design elements contribute to more flare with m240, I cannot say.

Here is a pro who is shooting the 240 all the time describing it.

http://www.fredmiranda.com/forum/topic/920070/1093#12745035

He was talking about "veiling flare" with the ZM18 in an earlier post, which really surprised me, since that lens is very resistant to flare with the M9, far more so than the 28 cron.

Here is Edward trying to get the colorshift correct with the ZM 35/2 on the 240:
http://www.fredmiranda.com/forum/topic/920070/1092#12741933

This is a lens which needs no profile at all on the M9.

Thank you. Now I understand.

These effects are not related to differences in CCD vs CMOS technologies. Instead they are a function of the micro-lense differences and other optical properties of the sensor assembly and, or minute differences in how reflected light behaves inside the camera. A small geometric change could have a great effect (as is always the case with flare).
 
When my M9 was working I did like the look. It produced images that looked much like the Hasselblad. The Hasselblad has a considerably broader to al range and color depth. The color depth was second to none which comes from a true 16 bit capture. Honestly I don't know why manufactures of 35 format digicams don't capture in 16 bit. The difference is like watered down paint vs concentrated pigments. It's more than just saturation, there's something that's hard to put into words. And yes, there is a different look to a CCD capture vs a CMOS.

Until recently the typical ADC bit depth, 12, for "35 format digicams" adequately represented all the analog information in the signals. The information content is limited by the sensor's signal to noise ratio and by how carefully the photographer chooses shutter time and aperture (exposure). The signals' information from very best possible exposures was not compromised by 12 bit digitization. In fact, clever, lossy compression can selectively suppress (filter) photon noise components in the signals so even 12 bits are not needed. Leica and several other brands offer(ed) lossy compression for this reason.

With continual improvements in sensor signal to noise ratios we now find a 14 bit ADC will make a difference for optimum exposures.

Sensors with larger full-well capacities record more signal (even though they may have higher read noise levels). In this case even 16 ADCs would be useful.

In one case the noise is decreased while in the other the signal is increased. But in both the increased information content (higher SNR) justifies ADCs with more bit depth.
 
Until recently the typical ADC bit depth, 12, for "35 format digicams" adequately represented all the analog information in the signals. The information content is limited by the sensor's signal to noise ratio and by how carefully the photographer chooses shutter time and aperture (exposure). The signals' information from very best possible exposures was not compromised by 12 bit digitization. In fact, clever, lossy compression can selectively suppress (filter) photon noise components in the signals so even 12 bits are not needed. Leica and several other brands offer(ed) lossy compression for this reason.

With continual improvements in sensor signal to noise ratios we now find a 14 bit ADC will make a difference for optimum exposures.

Sensors with larger full-well capacities record more signal (even though they may have higher read noise levels). In this case even 16 ADCs would be useful.

In one case the noise is decreased while in the other the signal is increased. But in both the increased information content (higher SNR) justifies ADCs with more bit depth.

Thanks for the information.

There's a visual difference on a wide gamut monitor to my eye.

Most of my work is published on the web or in print. Unfortunately the average monitor won't reproduce the additional information plus most monitors are poorly calibrated. In print the cmyk process is the limiting factor. Just estimating I'd say 50-60% of the information is lost. In reality almost any digital point and shoot would exceed the capability of most monitors and especially the cmyk printing process. Still a 16 bit file can be beautiful. It reminds me of the difference of looking at a 35mm slide vs a 4x5 transparency. It's not quite that dramatic but there is a difference.
 
I guess I'm different than many of the leica users. I don't care how sexy, cute, beautiful or how cool I look with it.

I could give a S what I look like with an M9. I seriously doubt many do.

In my experience people at large do love the way an M9 looks. That's a huge plus, because they mind less when you shoot. I look pretty goofy with it or any camera, and i carry them in spite of that, not because i want be to cool.

If I wanted to look cool, I would not carry any camera.

I've related in various forums to many digital Leica users and have yet to get the idea any bought one to pose with. The idea of Leicas bought as jewelry is pure fantasy, notwithstanding they are regarded by some as beautiful objects. Perhaps it's difficult for some to separate these two concepts.

As to the lack of dynamic range in the M9, it's a paper tiger to me. The Raws are good and can be pushed hard with little new noise, unlike the Sony Raws.

DXO has very low ratings for the M9 sensor, but I'll trust my lying eyes over their lensless measurements of irrelevant criteria.

Pretty M9.

Actually, my hiking and skiing camera of choice is my Fuji GA645i which, interestingly enough, weighs within a gram or two of a fully equipped M9 with the Summicron 35.

Of course, I have to pack extra rolls of film...but you have to pack extra batteries. Like XRay said, everyone develops their own preferences.

A sample of my most recent trip this fall. Not enough snow to go skiing yet. :(


Wonderful camera. I have a pretty clean GS645, and should use it more.


05730008-1-3 by unoh7, on Flickr

Indefinitely? At least until the sensor stock is depleted? Is this sensor still being manufactured?

It will be hard to implement their solution without producing more sensors. Do you read the bulletin at the start of this thread?
 
It will be hard to implement their solution without producing more sensors. Do you read the bulletin at the start of this thread?

Yes, I read it. And I just reread it again, as well as the same thing at the corporate site. I see no where any discussion of indefinite production of the sensors. I must have missed that?
 
Yes, I read it. And I just reread it again, as well as the same thing at the corporate site. I see no where any discussion of indefinite production of the sensors. I must have missed that?

20 bucks says there is another production run. :)

You in?
 
If it is "indefinite" you win. Otherwise, you lose. You in?
OMG seriously?

I said they said they would stand behind the sensor indefinitely. They don't need to make it indefinitely to do that, but I do think they need to make another run to address the current problem, which they imply they will, when they say they are working on the problem.

If they get that fix right, they only really need to make one more run, but of course they might get that wrong too, so who knows how many?

But, I bet you 20 bucks they will make at least one more run within 3 years.

You imply they won't make any more at all, so that should be a no brainer, easy money proposition.

We on?
 
Is it true about 40,000 M9s were produced?

I would say there will be more than one more production run of the current sensor assembly.

When the deamination problem is solved, there would have to be more runs to produce those assemblies.

And I'm only assuming about 15,000 cameras will eventually end up with the final sensor assemblies installed.
 
Hi,

I think there's a lot of owners of Leica CL's and M5's reading this thread and wondering about some of the things that have been said and wishing and sighing...

Regards, David
 
Back
Top Bottom