Jump to content

Any truth..


Recommended Posts

Advertisement (gone after registration)

Baloney is an accurate description. 'Full frame' sensors are by now at least moderately old hat, and can in theory be had from many sources -- including Kodak. If Kodak can make a S2 sensor, why can't they make a smaller M sensor?

 

The problem is the very small back focus distance of the M, 27.8mm as compared to around 40mm for SLR cameras. This means that rays coming from the lens hit the outer areas of the sensor at a very oblique angle. Film can handle this, but a digital sensor can't. Hence the offset microlenses over the pixel wells of the M sensor (other cameras also have them, but the M8 needs them more). Increasing the size of the sensor from 18x27 to 24x36mm will also increase the obliqueness of the rays, too much for the technology used in the M8. I thought this was pretty well known by now, at least around this forum.

 

Other technologies are being developed to deal with this vignetting, including Fresnel plates. One problem is that an absorption-type IR blocking filter over the sensor has to sit under the microlenses or Fresnel plate, directly on top of the pixel 'wells', because rays passing very obliquely through such a filter will suffer absorption not only of IR but also of much red, leading to a colour shift. I doubt that it is even theoretically possible to develop a sensor which is inherently immune to IR. So if there is a full frame M, then I suspect that it may completely lack an over-sensor IR filter and rely exclusively on filters on the lenses. The internal filter is in any case superfluous on the M8, as the lens filters block IR completely all by themselves.

 

Bu these troublesome facts do not trouble the professional rumour-mongers of course.

 

The old man from the Age of Glass Plate Negs

 

What? Send a man to the Moon. That's utterly impossible!!!:rolleyes:

Link to post
Share on other sites

x
  • Replies 43
  • Created
  • Last Reply
The 5DII sensor is fantastic, as are the Nikon D3X, D3 and D700 sensors. See dxomark.com for ratings based on color depth, dynamic range and low-light ISO. Having a great sensor is critically important at Leica's price level.

Yes, but on a rangefinder it would be worthless.

Link to post
Share on other sites

Yes, but on a rangefinder it would be worthless.

 

Sure, any sensor, including Kodak's, would be worthless if not made specifically for the rangefinder's internals, as explained by Lars. I suspect that Canon and Nikon could adapt their existing sensor technology to make a fantastic rangefinder sensor. As I wrote above, if Kodak can custom-build a sensor for Leica, so can another manufacturer, including Canon.

Link to post
Share on other sites

The patent is held by Kodak and Leica jointly....

 

So no one can make a sensor for any rangefinder other than Kodak + Leica? :confused: If that's true, then I hope Kodak + Leica can match Canon and Nikon for color depth, dynamic range and low-light ISO in the M9.

Link to post
Share on other sites

Advertisement (gone after registration)

Hmm - I had a 5DmkII for a month and was underwhelmed by the color rendition and dynamic range compared to the M8 (noise I will certainly concede). The Kodak CCD has much richer color capture. Bit depth is like megapixels - more is good only so long as everything else in the image chain - Bayer filtration, lenses, etc - is up to the challenge.

 

I tend to think of a sensor as the silicon - the other stuff on top (IR filtering, AA filtering, microlenses, etc.) being more or less customer options. In that regard I could see a technical possibility fo putting Canon (or Sony) silicon under 'stuff' compatable with M lenses. Whether the companies on either side are willing to do so is a different question. Canon sensors are not on the open market as a product.

 

For the hypothetical, EVF "solution" for R lenses, I can see where a 5Dii-like sensor with live-view capability is going to be a must. Of course, Panasonic is already doing live-view sensors that work with R lenses (G1, GH1) - scaling those up is probably more likely than setting up a "strategic partnership" with yet another competitor (viz. PhaseOne).

Link to post
Share on other sites

I hope Kodak + Leica can match Canon and Nikon for color depth, dynamic range and low-light ISO in the M9.

 

But I like the color depth and dynamic range of the M8. I don't really understand what the DXOmark numbers mean (and I was a math major but that was a century ago) but, without doubt and disrespect, I rather prefer the images made with an M8 over the Nikons and Canons.

 

As for the poor ISO performance, I took this pic in my room in the middle of the night, lit only with the light from the TV a good 12 feet from the head of the bed with a 50 Lux, hand-held at ISO1250 -- direct from .dng to .jpg, completely un-enhanced. Frankly, the 'graininess' that some disdain, I personally find very acceptable. :)

 

L1010472.jpg

Link to post
Share on other sites

So no one can make a sensor for any rangefinder other than Kodak + Leica? :confused: If that's true, then I hope Kodak + Leica can match Canon and Nikon for color depth, dynamic range and low-light ISO in the M9.

I hope not. except low light (and that is software, not the sensor) I greatly prefer the Leica offering.

Link to post
Share on other sites

Back to the OT a FF sensor of the 5DII type would make more sense for the "R10 like thing, which isn't an R10 but will make you happy", not for the M9 as it is not in the same league (except for noise which isn't the prime factor).

Link to post
Share on other sites

JT- It seems it's your knickers in the twist. You posted a reasonable thread worth considering and he simple recalled the current knowledge reviewed here many times, right or wrong. I think you were much more confrontational.

 

Thanks to both of you.

 

Keep it up Lars, Man from the land of common sense.

 

QUOTE=John Thawley;954458]Jesus Lars. Doesn't take much to twist your knickers, does it?

 

It was a question. "Any truth?"

 

Thanks for your centuries of knowledge, but I don't think biting peoples heads off is a reasonable response.

 

Let's not forget how financially fragile Leica is. One of these Japanese mega-conglomerates could easily swallow them up by digging out a little couch change... so, I'd suggest to you Old man of glass... or whatever, ANYTHING is plausible if not possible.

 

Keep an eye on that blood pressure. LOL

Link to post
Share on other sites

JT- It seems it's your knickers in the twist. You posted a reasonable thread worth considering and he simple recalled the current knowledge reviewed here many times, right or wrong. I think you were much more confrontational.

 

Thanks to both of you.

 

Keep it up Lars, Man from the land of common sense.

 

Sorry Carcam... you're off the mark. Reasonable people don't shoot the messenger.

 

Regardless, I'll bow out and leave it to you experts. LOL I've got pictures to make.

 

JT

Link to post
Share on other sites

The problem is the very small back focus distance of the M, 27.8mm as compared to around 40mm for SLR cameras. This means that rays coming from the lens hit the outer areas of the sensor at a very oblique angle. Film can handle this, but a digital sensor can't. Hence the offset microlenses over the pixel wells of the M sensor (other cameras also have them, but the M8 needs them more). Increasing the size of the sensor from 18x27 to 24x36mm will also increase the obliqueness of the rays, too much for the technology used in the M8. I thought this was pretty well known by now, at least around this forum.

 

Someone just has to take a Nikon mount CV 15mm lens and put it on a 5DII via an adapter - use live view mode and put the mirror up first. This will tell you how poorly it will do with a lens that has quite an oblique angle. I bet the existing Canon microlenses are something special and it will work pretty well.

Link to post
Share on other sites

Someone just has to take a Nikon mount CV 15mm lens and put it on a 5DII via an adapter - use live view mode and put the mirror up first. This will tell you how poorly it will do with a lens that has quite an oblique angle. I bet the existing Canon microlenses are something special and it will work pretty well.

 

Um, no...

 

If I understand correctly, the 5d and other Canon dSLRs have a much greater flange to sensor distance than the M8.

 

So there is a built-on angle problem with the existing M8; a full-frame sensor just makes that issue worse. Having said that I'm sure Canon could solve the problem if they wanted to.

 

But personally, I'd prefer if Leica stick with Kodak, though I have no knowledge of what any sensor company has up their sleeve right now. In general, though, I've preferred the look of the CCDs in my M8 and DMR to anything I've ever used or seen from a Canon or Nikon.

 

As for the DxO measurements, well, it might be true the latest sensors have more latitude (DR) in terms of the noise floor and the ability to record signals. In other words, I can pull a bit more out of the shadows on my D3 than my M8 (which can pull a bit more the DMR) at a given ISO (and the Nikon bests them all for how high that ISO goes without breaking up).

 

But for whatever reason, the resulting colour definition / scope / whatever is simply better on the M8 and best on the DMR. There really is comparison in my view.

Link to post
Share on other sites

Um, no...

 

If I understand correctly, the 5d and other Canon dSLRs have a much greater flange to sensor distance than the M8.

 

 

You and I both understand that. But that CV lens was made in a Nikon mount that pushes the lens way back into the camera when the mirror is up. (Just like the old Nikkor fisheyes and the old 21 f4) It is a non retrofocus design that has a more oblique light angle to the sensor than the current Leica wide angle lenses have. (Whether it is in a Nikon mount or in an M mount - it's the same.) So the flange distance is compensated for in the CV Nikon mount lens. (This is not like trying to put a Leica M lens onto a Nikon or Canon.) And if one either used it on a full frame Nikon in live view mode or via an adapter on the Canon also in live view mode, we'd be able to know if the oblique ray problem is really as severe as everyone thinks it is. (The Canon lets you turn on live view mode without a lens mounted, so then the mirror will go up and you can install the lens. I'm not sure if the Nikon works the same way.)

 

I'd think there must be someone out there who has this lens and camera combo and can test it. Theory is one thing, practice is another. I have a suspicion that the Canon 5DII microlenses may be better than we think. In which case, a full frame digital M that uses the Canon sensor might work well with Leica wide angle lenses which are of a retrofocus design now and fall somewhere between SLR lenses and the symmetrical wide angle designs when it comes to the angle of the light as it hits the sensor. (Too bad I long ago sold my old Nikkor 21 f4 or I'd be able to test it with that extremely recessed lens.)

Link to post
Share on other sites

What? Send a man to the Moon. That's utterly impossible!!!:rolleyes:

That was impossible with the technology of the 1930's (and most people do indeed base the judgment of what is possible on what was possible a generation ago). But it was never scientifically impossible. It did not violate any 'natural laws', as Tsiolkovsky and others, like Verne, had pointed out long before. And with the technology of the 1960's it became possible.

 

I got a chuckle out of your post because that was just what my mother said before 1957: "An artificial satellite? But it would just fall down because there's nothing to support it!" -- But she had missed out on the scientific part. Still, 'natural laws' do prohibit some things. Which is fortunate, for otherwise, the universe would not have existed. Technologies do change, but not the cosmos.

 

The same old man

Link to post
Share on other sites

Not a big loss as far as I'm concerned, never found C1 appealing.

 

Yeah, but Leica was emphasizing the fact that the collaboration with Phase One will guarantee the professional service needed in order to interest pro photogs into the S2. The development might be still on track, but it seems the after sale service just went sour.

Link to post
Share on other sites

Yeah, but Leica was emphasizing the fact that the collaboration with Phase One will guarantee the professional service needed in order to interest pro photogs into the S2. The development might be still on track, but it seems the after sale service just went sour.

 

Yes, I think you are absolutely right here. I don't think anyone at Leica has to grieve too much over the "loss of C1" since there a other converters available. But they can't throw the S2 on the market when they don't offer a professional service network at the same time. CS at Solms is hopelessly inadequate for amateurs, and professionals would turn their backs at once...

Link to post
Share on other sites

Yeah, but Leica was emphasizing the fact that the collaboration with Phase One will guarantee the professional service needed in order to interest pro photogs into the S2. The development might be still on track, but it seems the after sale service just went sour.

Point well taken. I had not consider the impact on the S2.

 

Cheers

Link to post
Share on other sites

Archived

This topic is now archived and is closed to further replies.

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...