M9 Sensor Qualities

[j j]

... the sensor is the sole source and interpreter of that data ...

Did you mean to say that? It's clearly not the case.

I have no particular interest in or desire for Leica cameras. There are far more suitable options for what I do available for far less cash in film and digital (rangefinder and SLR). My digital choices more or less follow those that you champion in this thread (current SLR with long lenses and mirrorless with short lenses).

But why you seem to think that a Leica user is falling for the hype any more than someone who believes Sony hype, Nikon hype or DxO hype is beyond me. There really is very little difference between the sensors you laud and the ones you lambast. Next year there will be another sensor that will render all others obsolete yet the pictures will look much the same.

I would love to see an example photo that could only have been captured with a Sony sensor or Nikon camera. Maybe you can post one?

 

[hepcat]

Exactly. Garbage in, garbage out. The old software mantra.

All software starts with input data and the sensor is the sole source and interpreter of that data. It captures and converts. The more it captures in an utterly quantifiable manner, the more "magic" you can work with.

Since sensors and in-camera processors, starting with the interpolation, are whereat all begins, the science of photography (as opposed to the art) cannot occur without this data.

This entire idea has gotten off track for you. You're treating the sensor as though it were the maker of images, and it just isn't so.

The sensor is NOT the "sole source of input data." The source of the input data depends on the photographer, how he selects his scene, and how he chooses to use the camera. The sensor merely records his choices. The software crunches the data, which is then turned into information. That information can be edited repeatedly again by the photographer in other software, and yet again with display or printing choices.

The data is gathered from the world and the process begins NOT with data processing, but with data selection. You can have the cleanest data in the world, but if it isn't the right data, it's useless.

The software is not the information. It all starts with the idea about how to capture the data... and if the idea is flawed, then the output is flawed. THAT is the source of GIGO... not what the software/hardware does with it. The data is merely data, some perhaps cleaner than others, but the same output can be had if it's vetted properly.

You're putting more value on how the software crunches the data than the process of selecting the data to gather.

Photography is about seeing an image and interpreting it for others to see. The process itself is unimportant as long as the final image looks like what the photographer imagined he/she wanted it to look like. The process itself (and therefore the camera/sensor etc.) is a very small part of that process. You're ascribing far too much value to the engineering of one piece of the process.

 

[Richard G]

I was at work today. Very busy.

 

[Lss]

Leica's sensors capture fewer photons than the market average for equivalent sensor sizes, especially per price unit (there are 2 deltas). The M9, and also the M240, under-perform.

You may have missed the fact that a great number of rangefinder users would happily see for these precious sensors to be put into an M mount rangefinder camera by any of the lower cost manufacturers. But no one delivers. Meanwhile the Leica cameras actually perform very well, and they can be bought from a store (for what you feel is too much money per perf unit). That's the bigger delta.

If you are not in the market for a rangefinder camera, none of that makes any practical difference.

 

[leafster]

After reading 6 pages of stuffs, is like reading a report without any purpose stated. I respect the OP inputs and find some of it interesting. I believe some of the readers do as well.

But I suppose not all people understand what OP is trying to achieve for this discussion. Whether OP wants a discussion, or comment, I really do not know. To have a meaningful discussion, I think OP should state his purpose for starting this thread. Because if we know what is his intent, perhaps we can better understand the perspective he is coming from.

Though at the end of the day, everyone will have different takeaway and move on. No reason to force throat.

 

[goamules]

If the scientific analysis of sensors mattered to most people you wouldn't have:

1. People shooting old, older, oldest mediums for art, work, fun. I shoot wetplate collodion, my Gosh, that's an obsolete sensor. Weplates have been featured in National Geographic the past few years.

2. People discussing other camera parameters. Focus peaking is one that only came about recently, but technocrats insist any mirrorless camera without it are not worth owning. Camera bodies, weight, rangefinder base distance, viewfinder brightness, shutter accuracy, shutter noise, lens quality (oh yes, 90% of our discussions)....et al matter to photographers. Tri-X would have died decades ago if it were just about sensor parameters.

Evidently, sensor parameters don't matter to most people. And people still buy Leicas. The OP just can't stand it, he wants, no insists everyone share his bitter disappointment. Sorry, all your clever arguments and big words aren't convincing anyone. People are just enjoying arguing with you. We're not changing you mind, you're not changing our minds, why don't you give it a rest now?

 

[hepcat]

Frankly, if we need to get our panties in a bunch over something Leica doesn't do, it's that they've apparently stopped development of firmware for the M9 series with a number of issues still unresolved. THAT is reason to be upset.

 

[Aristophanes]

This entire idea has gotten off track for you. You're treating the sensor as though it were the maker of images, and it just isn't so.

The sensor is NOT the "sole source of input data." The source of the input data depends on the photographer, how he selects his scene, and how he chooses to use the camera. The sensor merely records his choices. The software crunches the data, which is then turned into information. That information can be edited repeatedly again by the photographer in other software, and yet again with display or printing choices.

The data is gathered from the world and the process begins NOT with data processing, but with data selection. You can have the cleanest data in the world, but if it isn't the right data, it's useless.

You are reading too much into it, muddying the objective with the content.

Leica's sensor choices simply gather much less data (of any type from any source from any person triggering the shutter) than others, and are especially lousy at price per data point.

The tech limits the use of the tool.

And you bring up editing. The more data the easier the editing and the more fidelity preserved with each edit. Less data = poorer outcome.

Shoot wet plate for National Geographic and that's great. I actually watched NG do a massive large format shoot at Tikal almost a decade ago. Very impressive. I spent some time talking to some of the team. You know what a major concern was? Scanning. They had this massive movie set lighting system to illuminate the main plaza and they were concerned they could not get it all into the publication because of scanner d-max. Just like luxury market denominator is now the limiting factor in the quality one can get from a Leica sensor, their output was limited by a digital scanner.

 

[uhoh7]

Today, in the real world, what matters most apparently, is not the configuration inside the sensor itself. It's what you put on top of the sensor. Sony may have a wonderful sensor, but they stacked 1.75 mm of crap on top of it.

To make up for all this problem, they have a very strong processor designed to recognize the native lenses and try to make up for it.

The "Leica" Mistique is a funny thing: some of us see it as a reason to consider a camera, and myself, I considered the camera in spite of the name.

It's frustrating, because both the Leica and the Sony could very easily be better cameras with some simple changes.

 

[semilog]

This entire idea has gotten off track for you. You're treating the sensor as though it were the maker of images, and it just isn't so.

No, you're completely misinterpreting what he's written. And you are doing it over and over again.

 

[Corran]

I would love to see an example photo that could only have been captured with a Sony sensor or Nikon camera. Maybe you can post one?

I've already asked for this. I'm still waiting! 😀

That's the crux of the problem. Some folks say "Well, you can't make some photos with just a Leica M9 sensor." What hogwash. Let's see 'em.

 

[hepcat]

No, you're completely misinterpreting what he's written. And you are doing it over and over again.

I think I have a pretty solid grasp of what he's saying.

The O.P. is playing Chicken Little and is yelling "The sky is falling!" And I'm calling B.S. The sky is NOT falling, and you're right; I've done it over and over and over again.

He's saying we should all be angry, upset, or something because since Leica is charging premium prices for their equipment, they are somehow cheating or defrauding users because the equipment is using "inferior" sensors rather than the most current high-end sensor available today. We're obtaining inferior files with inadequate information and that there's just SO much more we're missing out on.

The sensors used in the M9 and M240 are relatively low-production chips that use micro-lenses that are designed specifically to accommodate the angle needed for the sensor-to-flange distance. No other camera produced needs that. When the M9 was designed, the full-frame Kodak CCD WAS state-of-the-art. The M240 CMOSIS sensor was also designed with that same micro-lens design specifically for the M240. Both cameras perform admirably in the real world, and the test-chart differences that are apparently SO big to him, are just not an issue in using the cameras. The answer is really quite simple: if you don't see Leica as a good value, don't buy Leica.

The sky is NOT falling. This is NOT a big deal. You can't tell what high-end camera took any specific image unless you look at the EXIF data. If the end product was a file and signal-to-noise was the end result this would be important. The end product here though is a photograph. There are many variables in the production of that photograph, of which the sensor output is merely one. The Leica sensors work fine. Give it a rest already.

 

[anerjee]

I recently attended a talk by the Dr Karbe. He mentioned that the thickness of the glass layer on top of the sensor was the most important characteristic determining the resolution of Leica lens on digital cameras.

By his admission, they went overboard with the M8 (too thin), and corrected in M9. The M240 is thinner than the M9 but thicker than the M8.

M cameras will never have high-iso performance of same generation Nikon cameras, as they have different design constraints, backward compatibility with a generation of Leica lens.

However, these constraints allow for small and high performance lenses to be designed. Looking at raw sensor characteristics is moot.

 

[mfunnell]

The sensors used in the M9 and M240 are relatively low-production chips that use micro-lenses that are designed specifically to accommodate the angle needed for the sensor-to-flange distance. No other camera produced needs that.

This is quite possibly an important point - I have no idea what the engineering constraints imposed by this are, nor do I have any idea how this impacts commercial arrangements for sourcing relatively low volumes of specialty chips, especially how the costs of tooling for their production are spread over that relatively low volume.

M cameras will never have high-iso performance of same generation Nikon cameras, as they have different design constraints, backward compatibility with a generation of Leica lens.

"Never" seems an awfully big word here. It may be that some future developments make this type of production routine or at least relatively unproblematic. I'm pretty sure, though, that isn't the case right now.

I suspect (though I guess don't know for sure) that critcising the performance of full-frame digital sensors suitable for M-type digital RF cameras against some supposed ideal is akin to criticising a dancing bear for poor performance of the very hardest ballet steps, rather than praising the poor critter for being able to dance pretty well overall.

...Mike

 

[Aristophanes]

I think I have a pretty solid grasp of what he's saying.

The O.P. is playing Chicken Little and is yelling "The sky is falling!" And I'm calling B.S. The sky is NOT falling, and you're right; I've done it over and over and over again.

He's saying we should all be angry, upset, or something because since Leica is charging premium prices for their equipment, they are somehow cheating or defrauding users because the equipment is using "inferior" sensors rather than the most current high-end sensor available today. We're obtaining inferior files with inadequate information and that there's just SO much more we're missing out on.

The sensors used in the M9 and M240 are relatively low-production chips that use micro-lenses that are designed specifically to accommodate the angle needed for the sensor-to-flange distance. No other camera produced needs that. When the M9 was designed, the full-frame Kodak CCD WAS state-of-the-art. The M240 CMOSIS sensor was also designed with that same micro-lens design specifically for the M240. Both cameras perform admirably in the real world, and the test-chart differences that are apparently SO big to him, just not a big issue to most of us who use them. If you don't think Leicais a good value, don't buy Leica.

The sky is NOT falling. This is NOT a big deal. The Leica sensors work fine. Give it a rest already.

The Kodak CCD sensor was NOT state of the art.:

http://www.dxomark.com/Reviews/DxOMark-review-for-the-Leica-M9

"In comparison with the sensors used by other full-frame main manufacturers, the pixel quality of the Leica M9 sensor remains low."


"The results of the Leica M9 are very close to the measurements for the Canon EOS 5D, launched four years ago. The Leica M9 provides good image quality for low ISO, but its results for high ISO are weak, with dynamic range decreasing very fast. So its Lowlight ISO score (shown below) is a little disappointing, especially for this type of camera. The Leica M9 achieves the lowest score among measured full-frame sensors."

That was March, 2010.

Canon had gapless microlenses in the 50D in 2008. Microlenses are nothing special to Leica and the M-mount has quite a bit more real estate to work with compared to Fuji and m43. One problem for Leica appears to be their sensors have greater stack height, in part because they are CCD. Definitely NOT state of the art at that time.

Fast forward 4 years and we see that Leica continues to advertise a luxury brand with sensors that under-perform compared to consumer-grade sensors. My observation is simply that this is part of Leica's business model; under-deliver on sensors and continue to charge a premium. It is the impression of having bought superior quality that counts, not the measure of quality. Form over function often drives luxury brands and marketing.

This does not appear to be a choice along the lines of adhering the to manual focus ethos. The test chart differences are technically important but also economically important because they who that despite a premium price tag, a Leica cannot capture a shot any better than a consumer grade sensor camera can. How Leica overcomes that difference is through marketing the past.

The old Leica advantage was technical AND economic. Your Tri-X and Velvia would perform better in your Leica because it had an ultra-flat film plane, because the optics were so clear and high rez, because the RF alignment was so scrupulous, because you could both focus and shoot swiftly, because RF design allowed for larger apertures. Now most of those advantages have washed away because they are filtered through sub-par sensors. Maybe they are "good enough" for some here, but in terms of Leica's history, this is not up to the pedigree nor reputation. People's emotional responses are quite strange about this topic.

 

[PKR]

It's been some time since I left a post on this forum. But, the "sensor wars" forced my hand. I take pictures for fun and profit. I've used everything from a Nikon Cool Pix 6000 to a Phase One P25.

Sensors are an electronic device. They have photo sites that react to being banged by a photon. As a pal of mine often says .. ITLOPS (it's the laws of physics stupid). He designs IC devices.

I find that I can live with 12-16 MP for most all I do. I think that Semilog (as usual) is correct in his description of sensor behavior (he has to perform sensor tests as part of his day job) - in short, it's empirical knowledge. I also agree that SONY has lead the way for the past few years. I was told that they were looking for some older foundry gear that was designed before the current standard (150mm ?). This will allow production of nonstitched sensors in large sizes - 35mm frame and likely bigger.

Leica cameras, in my opinion, have become an accessory for the rich. Fashion before S/N ratio I think. I don't know any pros in my circle of friends who use Leica digital cameras. Many can easily afford them. Things just change..

best to a few who might remember me. pkr

 

[goamules]

OP... People are just enjoying arguing with you. We're not changing your mind, you're not changing our minds, why don't you give it a rest now?

<self quote> I guess not. </self quote> I'll check back in in 24 hours to see if the OP is still obsessing on his sensor diatribe.

 

[uhoh7]

Leica cameras, in my opinion, have become an accessory for the rich. Fashion before S/N ratio I think. I don't know any pros in my circle of friends who use Leica digital cameras. Many can easily afford them. Things just change..

best to a few who might remember me. pkr

opposite of my experience.

Would you consider a decent motorbike, an accessory for the rich? Leicas are considerably cheaper than a good new or used motorbike.

Alot of my friends, who are not rich, have bicycles which cost more than a M240.

All that said, for me M240 is too high. But M9 is at decent price now.

Price for what? For a fairly small camera with a fantastic lens set.

The body and lenses are not separate in the real world--it's the package which still dominates for sheer image quality from 18 to 135mm.

As to pros> more have a leica today than since the early sixties.

Leica in fact is thriving, doing equally well with the rich as the obsessed.

 

[rivercityrocker]

Yes, but the fact that there are lights on the subject changes everything. Concerts are also by far not the most demanding work for a camera's ISO performance.

A while back I was in Hong Kong doing a set on the annual June 4th commemorative march. The streetlights were dimmed, and an entire street next to the Pearl River was illuminated by candlelight . That's what I mean by dark.

And however much I applaud your technique of getting people when they are between motions, a better camera would allow you to simply shoot whenever you want. Back in the 1d mk3 and 7d days I also had to do this, when I use the NEX-7 I still have to since I'm mostly at 1/30-1/60. But on an 1DX I just fire away at 1/250, and performance is admirable enough at 25,600 that I couldn't care less. That's good high ISO performance.



Really. You feel the need to actually rant like this. Well, sorry to tell you, but I am pretty good, concert work or not concert work. I shot film for several years before going to a Canon system, which I sold between 2010 and 2012. Yes, I may have not shot as many concerts as you, but I used to shoot Canon/Leica, now I shoot Sony/Medium Format, the local paper uses Nikon so I work with Nikon files all the time. I've been using Photoshop since CS2, and I can tell you, if you've carefully worked on 30,000 files from different cameras, you can tell how good a sensor is by looking at a couple of files.

Please give me one single example of you pushing an M9 file to above 6,400. That's four stops beyond ISO 400. I simply don't believe that it can be done with any respectable quality of output. And F1.4 is not fast for a 50mm prime. If you are indeed working with the M9, the super-fast options are both cheaper and more plentiful than Canikon - I'd set the bar at 1.2 for a standard lens, at the very least.

Oh, and I get plenty of contract work from bands in Asia. I don't think most "bar bands" perform in soccer stadiums....



No, I'm not doubting the M9 as a concert tool, I am doubt it's high ISO performance. I don't think that the concert pictures you made are bad - far from it - but in 2014 there are a lot of ways to get better output.

Yeah, in 2014 there ARE lots of ways to get better output. I have a D700, Df, and D5300 presently. I've owned every Nikon DSLR since the D1 with the exception of the D4 (I've written TWENTY books on Nikon cameras). I know my D700 and Df can get great shots in low-light. Yes, they are much better at high ISO settings, but I still prefer to use the M9 (and M8) because I prefer the interface. Shooting DSLR's bores me to tears. The M9 challenges me. Manual focusing in the dark using a camera that sucks in low light makes me work harder. I could have kept my D3s so I could fire off 9fps and get lots of great images, but that's not how I work.

You can take the easy route. That's your prerogative. But I will put you to the challenge that I could take a Nikon D1 and still come out with better concert images than you can with a D4s or 1DX or whatever the latest low-light super-camera is out there.

 

[Ranchu]

I think ccd's do look better than cmos. Colors would probably be better with only one green instead of two, though. DxO's raw converter was a terrible looking thing, imo.