Thoughts about the M9

[jonoslack]

I guess that with the S, and the stuff on the R10 we're all assuming that the M9 will be 2011 or 2012.

I got to thinking about this. Clearly it's in Leica's interest to pre-announce the S1 and the R10, because they are not currently producing anything to compete with it.

The M9 is a different story though. If they were to say "there will be an M9 in March 2010" for instance, there would be an immediate impact on M8 sales. So they wouldn't say that would they?

On the other hand, a full frame M9 with decent high ISO is the one thing that really would have guaranteed sales, so you might imagine they would be steaming ahead with it.

I know nothing, but one is entitled to dream

 

[Lars]

I know nothing, but one is entitled to dream

Sensor choices are certainly better now than when M8 was designed.

How much of a challenge is it to get a FF sensor to work with wideangle RF lenses?

 

[scott kirkpatrick]

I agree with you, Jono, that the pre-announcement pressures are for the reflex products and against the rangefinder product. I also think the schedules at Leica are gated by scarce skills, like the folks to make all the firmware work inside these cameras. (And testers, who appear to be scarcest of all.) So I can't imagine real work on anything but the S2 taking place until it ships in, let's say late September 2009, and maybe even for the first few months after release. During 4Q 2009, Leica will be able to tell whether the S1 is selling incredibly well, or just fighting for space on the MF-digital shelves. So sometime in early 2010 a decision about whether to go with equal energy to the R10 or bring out an M9 first will get made. They certainly know how to get an M9 right by now, so the actual execution could be relatively rapid.

Announce in 2010 Photokina? Prototypes in 2010 PK?

scott

 

[jonoslack]

Announce in 2010 Photokina? Prototypes in 2010 PK?

scott

Hi Scott - well, that's the conventional wisdom . . . I was just thinking they might be in more of a hurry than that.

 

[Mango]

A 28mm f1.4 Summilux Asph is (or should be) in the works perhaps? It's the only wide angle focal length with a missing Summilux in the lineup.

 

[johnastovall]

The Full Format will come in good time. Let's wait for Leica the get the S2 right. I thing the fortunes of the M9 ride on the success of the S2.

 

[scott kirkpatrick]

Hi Scott - well, that's the conventional wisdom . . . I was just thinking they might be in more of a hurry than that.

I'm just trying to list what they need to do to bring out a retro-style M9, and my first guess is that firmware is the critical limit. In the past 12 months, working alonside the S2 developers, the M-team has managed to add code for the delayed shutter rewind and auto-ISO, plus update some of the tables (not all of them, unfortunately) for four new lenses, including developing vignetting corrections for each. The M9 I would like to see would give us FF DNG files with 16-bit resolution, which requires about 8 times the processing power and 4 times the I/O bandwidth. If the processing components are evolved from the Blackfin and other DSP-like chips in the M8, that might be a task that can be contained to a very small team. But if in addition, the new ASICs that Leica has developed (with Fujitsu?) are to be used, that is a complete rewrite of the firmware, and will take a bigger team and many more months. Of course, the structure might be much improved, and the M8's problems of halting when the different computing pieces fall out of synch might go away, but ...

The other critical resource is the design skills to make the body a big bigger to accomodate a FF sensor, move the battery box a bit, choose a stronger battery (??), and those people are certainly tied up adding just one more button to the S2 and making sure that it comes back from the factory looking and feeling the way it should, until Sept 2009.

So things like an M8.4 with 16bit, 1.3X resolution and the same processing components and body could sneak out, but I'm shooting with what I have in the meantime, not waiting for the M9.

scott

 

[jonoslack]

:ROTFL:
I stand rejected

just a theory.

 

[fotografz]

I personally don't care about FF as the M8 shoots perfectly good images as it is (with some well documented caveats), but FF would be nice on condition that they get the useable ISO up a notch to make the available light use more diverse. Today's technology should make that doable.

I'd actually prefer a higher ISO ability to FF given the choice. Available light is where its at for rangefinder work IMO. The M camera has never been all that comfortable coexisting with strobists principles.

In the meantime, investments in optics are never a bad idea ... especially when it's Leica glass

 

[Bob]

Now if they just had an ir filter coating technology that eliminated those secondary reflections, I would very happy.
-bob

 

[ecliffordsmith]

Hi All,

Jono, I can see your point regarding killing off sales etc. If I were considering a second body for example that sort of information would make we wait!

I have no personal desire for FF other than the given wisdom appears to be that a bigger sensor can have more and bigger pixels and thus the high ISO performance can be better. I have never been an M film user though and a lot of those people that would like the FF sensor appear to be or have been. It seems that some of the newer crop cameras have greatly improved high ISO performance though so even without a FF sensor it does not seem unreasonable to me to expect a full two stop gain in that regard.

Still the way I look at it is that if a FF sensor comes along I buy a 50 Summilux. If it does not I already have my 35 Summicron.

The removal for the need of the filters would be a big bonus for me for the reasons Bob stated.

 

[scott kirkpatrick]

Here's a question -- Which would you rather see, an M8.4 in 4Q2009 which offers the same imager size (probably the same chip, but wreathed in marketing copy to convince us it has new powers), full 16bit files, produced at half the shooting rate, and some new firmware feature such as optional enhanced high ISO correction in the raw files, or...

An M9 in 1Q2011, with S2 Maestro technology and a full-frame 24x36mm sensor, choice of lossless-compressed (roughly 10 bit per pixel) or uncompressed images. manual focus still, but in-body image stabilization and dust-shaker and the good taste to not offer live view, since the imaging chips still comes from Kodak, or ...

If you are the two-handed sort who loves these discussions, you may answer twice, once for your own shooting needs, and once assuming that you are a Leica investor. The dates are ship dates, not announce dates.

scott

 

[Lars]

Scott,
I'm not a Leica shooter, but I think that full-frame would be a must for me to consider a digital M. BTW 10 linear bits per pixel is not nearly enough, that would be a disaster. So I think your second option is not well proposed. Leica has to prioritize (all aspects of) image quality and usability to defend its name and price point. Nothing wrong with live view BTW - it would be a great addition to a rangefinder, which is at a disadvantage versus an SLR in some situations (filtering, macro, focal lengths not covered by viewfinder). Saying that's not good taste is just snobbery.

 

[gero]

Scott,
I think that full-frame would be a must . BTW 10 linear bits per pixel is not nearly enough, that would be a disaster. So I think your second option is not well proposed. Leica has to prioritize (all aspects of) image quality and usability to defend its name and price point. Nothing wrong with live view BTW - it would be a great addition to a rangefinder.

Lars, I agree with these lines, on the other hand I dremt that a friend had an "I" Leica rangefinder that was like the M but with the S sensor. It was so nice.

 

[scott kirkpatrick]

Scott,
.... BTW 10 linear bits per pixel is not nearly enough, that would be a disaster. So I think your second option is not well proposed. ....Nothing wrong with live view BTW - it would be a great addition to a rangefinder, which is at a disadvantage versus an SLR in some situations (filtering, macro, focal lengths not covered by viewfinder). Saying that's not good taste is just snobbery.

Lossless compression as Nikon does it I believe gives you an average filesize of about 10 bits instead of the original 14 bits, but I haven't used them. I wasn't assuming a lossy compression like the M8 has now. And looking at Olympus' LiveView cameras, I believe they have left about 1 stop of dynamic range behind by using a significant fraction of the pixel real estate for the video circuitry involved in live view output. So, to my taste, I would rather have that stop than the live view. Obviously YMMV.

scott

 

[Gary P]

I agree with fotografz, to me better low light high iso performance would be preferred to FF.

Since we're dreaming, would anyone like a manual shutter? I mean bring back the cocking arm and eliminate the motor and it's ZZZZZZinck.

Gary P

 

[Woody Campbell]

There was a long thread on the Lieca forum that surveyed members views on desirable M9 features. Here are my thoughts.

1. Continue to use a CCD chip. This is a key aspect of the M8 look and I would not want it to change. Implications: probably no live view and not a huge improvement in low light performance unless Lieca adopts a binning approach like the new Phase backs (which would be a terrific solution for the M9).

2. No AA filter.

3. I've found the M8's IR sensitivity to be useful - it effectively lets the user decide on the camera's spectral response through the use of filters. I like this approach but realize that I'm probably in a minority.

4. Environmental sealing.

5. This is important - an electronic rangefinder with focus confirmation. This could help resolve low light focusing difficulty with very fast lenses. It would also permit firmware to compensate for focus shift and could permit users to fine tune focus calibration on a lens by lens basis.

6. Larger viewfinder with more eye relief. Think the current CZ film RF camera.

7. The rest of usual stuff (sensor self-cleaning, great big lcd and useful info displayed on the top plate like the s2, etc.).

4,5 and 6 probably mean that the appearance of the camera will be a break from the M tradition but that's ok with me.

 

[nei1]

If leica had any balls they"d make a black and white only version of the m8.Might just buy one

 

[johnastovall]

If leica had any balls they"d make a black and white only version of the m8.Might just buy one

Hear! Hear! I'll be in line with you. :thumbs:

 

[Lars]

Lossless compression as Nikon does it I believe gives you an average filesize of about 10 bits instead of the original 14 bits, but I haven't used them. I wasn't assuming a lossy compression like the M8 has now. And looking at Olympus' LiveView cameras, I believe they have left about 1 stop of dynamic range behind by using a significant fraction of the pixel real estate for the video circuitry involved in live view output. So, to my taste, I would rather have that stop than the live view. Obviously YMMV.

scott

Oh OK so when you mentioned 10 bits you refer to file size after compression, not actual stored bits? 14 bits would be good, beyond that any added precision has marginal value, and also has a detrimental impact on the efficiency of lossless compression.