Godfrey
somewhat colored
I have both an M9 and GXR with A12 Camera Mount.
I like the GXR quite a lot, but I think it would be something of a mistake to update the camera unit to a larger format without first addressing some of the issues with the body. For me, the body needs more responsiveness: it is both slow and a bit inconsistent on the shutter lag. The write speed is modest, and it doesn't buffer writes to media so it blocks for a third to half a second. I'd like to see updates on both those fronts as well as a newer, higher resolution EVF before I went to the expense of a FF sensor camera unit.
I expect it would be difficult for Ricoh to produce a FF sensor camera unit with M-bayonet customizations, that performs as well or better than the current A12 Camera Mount, that costs much under $2000 MSRP. I would expect a price in that ballpark at least. They'll need a new shutter and I have to wonder what sensor they'd use. Optimizing a FF sensor for RF lenses is a non-trivial effort—look what Leica is doing with CMOSIS, what they did with Kodak in the past. It's a much more difficult sensor optimization than optimization for APS-C or SLR lenses due to the short mount register and the size of the format relative to the lens mount and distance to primary nodal point with a large number of lens designs.
I think Ricoh would be better off updating the A12 Camera Unit to the latest 16Mpixel Sony APS-C sensor, and optimizing that. They've already got a good shutter for that, the micro-lens optimization would be much easier, and the net gain of 1.5-2.5 stops of sensitivity would be welcome. That could potentially keep the costs down in the same ballpark as the current A12 Camera Mount (which, frankly, continues to work very well compared with even the latest cameras out ... getting better than that is essential if any update at all will be profitably marketable).
G
I like the GXR quite a lot, but I think it would be something of a mistake to update the camera unit to a larger format without first addressing some of the issues with the body. For me, the body needs more responsiveness: it is both slow and a bit inconsistent on the shutter lag. The write speed is modest, and it doesn't buffer writes to media so it blocks for a third to half a second. I'd like to see updates on both those fronts as well as a newer, higher resolution EVF before I went to the expense of a FF sensor camera unit.
I expect it would be difficult for Ricoh to produce a FF sensor camera unit with M-bayonet customizations, that performs as well or better than the current A12 Camera Mount, that costs much under $2000 MSRP. I would expect a price in that ballpark at least. They'll need a new shutter and I have to wonder what sensor they'd use. Optimizing a FF sensor for RF lenses is a non-trivial effort—look what Leica is doing with CMOSIS, what they did with Kodak in the past. It's a much more difficult sensor optimization than optimization for APS-C or SLR lenses due to the short mount register and the size of the format relative to the lens mount and distance to primary nodal point with a large number of lens designs.
I think Ricoh would be better off updating the A12 Camera Unit to the latest 16Mpixel Sony APS-C sensor, and optimizing that. They've already got a good shutter for that, the micro-lens optimization would be much easier, and the net gain of 1.5-2.5 stops of sensitivity would be welcome. That could potentially keep the costs down in the same ballpark as the current A12 Camera Mount (which, frankly, continues to work very well compared with even the latest cameras out ... getting better than that is essential if any update at all will be profitably marketable).
G
Godfrey
somewhat colored
Optimization of a sensor for a) a new interchangeable line of lenses dedicated to the camera you're going to put it in, or b) for a new non-interchangeable lens camera is FAR less difficult and less expensive than optimizing a sensor for a wide range of lenses never designed to work with it in the first place.
That's a fact of life in today's technology.
bwcolor
Veteran
I'm guessing that all of us want Ricoh to make money. Also, we would all like them to do a first rate job and rival the Leica 'M' files and ergonomics.
All of this takes money and a lot of development time/cost. So, I don't expect nor want a product on the cheap. I'm betting to meet the above, we are talking around $3K for a complete camera... body and module, or camera.
All of this takes money and a lot of development time/cost. So, I don't expect nor want a product on the cheap. I'm betting to meet the above, we are talking around $3K for a complete camera... body and module, or camera.
rbelyell
Well-known
pardon my ignorance, but how can one 'optimize a sensor' for 'lenses never designed to work with it'? that seems 'definitionally preclusive' to me, ie, cant happen as a matter of logic. and it wasnt really what i was talking about...the canon 5d i used for years, with only one lens 'designed' to use with it, worked amazingly with a whole bunch of lenses never designed to be used with it. we're talking about a FF mirrorless, doesnt matter if its sony mount, fuji mount, whatever, we on this forum will most definitely be using it with 'non optimized' lenses. extra 'technology' cost is not apparent to me.
rxmd
May contain traces of nut
Actually it seems to sum up pretty much what you were talking about. Basically you were saying that just because Fuji managed to produce a small-sensor, not-for-generic-lenses camera for a comparatively bigger market for $1500, a generic M mount full frame module should be cheap.
What Godfrey meant was that it's a lot harder (read: more expensive) to build a camera that works reasonably well with all sorts of known and unknown lenses that people might stick in front of it, than it is to design a system where you build both the sensor and the lenses and can fine-tune them to each other. A generic M module (or whatever) is a case of the former, the cameras you mentioned are a case of the latter. In other words, if you want generic, prepare to pay more.
The other reason why the former is more expensive is that there are a lot fewer buyers. The market for generic M modules is marginally small, hence less economies of scale and higher prices.
In other words, this kind of comparison doesn't work, it only leads to wishful thinking.
Matus
Well-known
Something around €2000 would be doable, under that it would be OK - IF the price of the 'new' Ricoh GXR II body would remain where it is now, or for €250 more would include (as large as possible) EVF on board.
Godfrey
somewhat colored
The whole point of a digital camera with an M-bayonet lens mount is to allow the use of digital capture with existing M-bayonet (and M39) mount lenses. The *vast* majority of M-bayonet lenses were designed long before digital capture even existed: they were designed for film cameras of a particular type. So designing a sensor for a Leica M digital camera or for a Ricoh GXR M-bayonet lens mount camera unit is indeed designing an optimization for lenses that were never designed to be used with a digital sensor.
I don't know that there is something "definitionally preclusive" or whatever about that statement.
The question that might shed light on the problem for you is "What's different between RF camera lenses and SLR camera lenses that makes it harder to optimize a sensor for the RF camera lenses?" Or, from the context of your post, "Why do my SLR lenses work so well on the digital bodies when they weren't designed for digital sensors either?"
RF cameras do not have the big mirror box with a swinging mirror that has to clear the back of the lens. Because of the mirror box and mirror, lenses for SLR cameras sit further from the image plane and have been designed with optical tricks that have the side benefit of allowing the light rays to strike the image plane at a more orthogonal incident angle, which is what digital sensors work best with.
RF camera lenses sit closer to the imaging surface because of the lack of mirror box and were designed to be more compact since they didn't have to worry about clearing a swinging mirror. Generally speaking, as the focal length get shorter than the format diagonal, the lens insets further and further into the body and the nodal point gets closer and closer to the image plane. This causes the ray trace to move well off the orthogonal across the image format.
This is no problem with film but causes issues due to the geometry of the components on a digital sensor. That's why to handle RF lenses, Kodak designed a sensor with selectively offset microlenses for the CCD and the new CMOSIS sensor uses a new design for microlenses and CMOS circuitry that helps re-target the ray trace appropriately.
There's a lot of development cost in that design necessity, as well as a good deal more cost in manufacture due to production variances - many more sensors are scrapped as not meeting the desired quality.
rhl-oregon
Cameras Guitars Wonders
How about $1250 (based, roughly, on GXR M-module $649 x 2). Focus on the FF sensor for now, not on other extras (some of us are happy to leave brightline VFs to film cameras and work with Ricoh's mix of focus-peaking EVF + sharp display.
rbelyell
Well-known
thanks godfrey, that makes sense. i think we were just talking past each other. i still dont think itd cost sony very much to trade its $1000+ zeiss lensd FF for a similarly priced lensless FF...we'll see it soon, i,m certain of that.
bwcolor
Veteran
I thought that the yield on a new FF sensor was substantial lower than an APS-c sensor. I think it unlikely that this ratio would be sufficient to pay for the larger sensor/electronics and R&D. There are now at least two (Japanese) and perhaps one (French) manufacturer of high resolution/fast EVFs, so that part should be pretty easy, but many want EVF in the body. That will both make such a camera more appealing to many and more expensive. I won't spend good money on a FF camera without a really stellar EVF, or OVF.
Godfrey
somewhat colored
I suspect that if/when the Sony RX1 is morphed into an interchangeable lens camera, the cost of the body will be $2500 or so ... And it will be designed to work best with a new range of interchangeable lenses that are optimized for the sensor design, if they want to keep the camera as compact as possible. Fitting M-bayonet RF lenses to it will likely produce the same issues that fitting M-bayonet lenses do on current Fuji Xpro1, micro-FourThirds and NEX bodies ... but moreso as the format is much larger.
The only cameras with sensors optimized for RF lenses are the GXR, M8, MM, M9, and upcoming new M at present. They do the lenses justice. Upgrading the GXR with a FF sensor camera unit would be great, but I expect it will not be inexpensive.
I will be delighted to be found wrong if and when they do it!
bwcolor
Veteran
Sony won't be a player in the Leica 'M' alternative for my Biogons. Zeiss can provide them with three primes and their FF 'A' lens adapter can get them the rest of the way.
LeicaTom
Watch that step!
I'd pay $3,500 for a perfect operational M module for a Leica M6, something that's 18MP plus + etc.
Leica itself should have built a M digital module way before they even made the M8, it's a shame that they don't get engineers to build it to accept M6 & M7/MP bodies, they would have thousands and thousands of NEW customers and more people dependent on company for repairs/support.
I don't think they will ever get it right.....that 240 is just a overpriced trinket, who the hell buys a Leica to take videos??? If they would just look at what the public really wants and also make their products more affordable, the company would be even more successful......
$3,001 to $3,500 would be acceptable if they could build a reliable, dependable user friendly, company repair supported digital module, but it'll NEVER happen, the company is run by a bunch of greedy immoral capitalists, who don't give a damm about what real photographers really want, with Leica it's just a money thing, their concern over what Photographers think ended in the late 1960's, since then it's all been about what they can do to save their own ass from bankruptcy , the Japanese should have bought them out years ago, we'd have more affordable equipment and HIGHER quality gear if the Japanese owned Leica.
Tom
Leica itself should have built a M digital module way before they even made the M8, it's a shame that they don't get engineers to build it to accept M6 & M7/MP bodies, they would have thousands and thousands of NEW customers and more people dependent on company for repairs/support.
I don't think they will ever get it right.....that 240 is just a overpriced trinket, who the hell buys a Leica to take videos??? If they would just look at what the public really wants and also make their products more affordable, the company would be even more successful......
$3,001 to $3,500 would be acceptable if they could build a reliable, dependable user friendly, company repair supported digital module, but it'll NEVER happen, the company is run by a bunch of greedy immoral capitalists, who don't give a damm about what real photographers really want, with Leica it's just a money thing, their concern over what Photographers think ended in the late 1960's, since then it's all been about what they can do to save their own ass from bankruptcy , the Japanese should have bought them out years ago, we'd have more affordable equipment and HIGHER quality gear if the Japanese owned Leica.
Tom
mfogiel
Veteran
I am not interested, however, if a decent digital sensor a' la Leica S2 came out inside an M body, I would pay the equivalent I spend in film, chemicals and scanning time value multiplied by the duration of the warranty. As an example, last year I shot about 150 rolls, so that would be 600 EUR for film, another 100 for the chemicals, and a very conservative estimate of 1500 EUR for the scanning hassle, that makes a total of 2.200 EUR per year. In case of a 2 year warranty the price turns out at 4.400 EUR, actually quite close to what the current M9 goes for.
pvdhaar
Peter
I'm perfectly fine with APS size mirrorless. Going full frame would mean that the lenses scale up to FF size and weight as well, defeating the whole purpose of mirrorless as far as I'm concerned, which is compactness without sacrificing the possibilities that a DSLR offers.
I can't see myself paying more than $2000 for a body anymore.
GaryLH
Veteran
I would love to have a Ricoh gxr ff m module... But I don't think this will ever happen. Outside of that, I am already happy w/ what I have in apsc mirrorless. But I agree w/ others, everything scales up in size and weight for ff and depends on who implemented it. If it is a Nex, not sure if I would be interested due to their UI and native Sony lenses tend not to be as good as third party offerings.
I also feel these days nothing over 2k for a body.
Gary
I also feel these days nothing over 2k for a body.
Gary
GaryLH
Veteran
Yield tends to improve over time due to better technology being available and better understanding of what an be done in the manufacturing process to fix the causes.
An example is the Nikon d700 sensor vs the d800 and then d600. I currently am presuming ff sensor yields have improved a lot to be able to offer a ff dslr at the price of the d600.
Gary
Matus
Well-known
Garry - the quality of the processes to produce FX digital sensors surely improved over past few years - no question about that. But to make such a sensor really suitable for rangefinder lenses it needs 'special treatment' AND the production volume is small, much smaller than for APS-C mirror less or FX DSLR.
Even the latest M240 may give you problems with some wide lenses and Leica designs their wide lenses with some retrofocus to improve the situation AND it needs lens profiles.
But yeas - it would be something if Ricoh would bring FX M-mount modules ... with a new body ... with built-in EVF ... ... probably not going to happen.
GaryLH
Veteran
I would tend to agree, sadly, that Ricoh may never do it. I think they have lost a lot of focus after acquiring Pentax. Lets c how well they refocus now that the grd is done.
Everything u mentioned about ff sensors I agree w/... Which is one of the main reasons I am not really interested in ff any longer plus over time I have come to realize, for the type of pictures I take, ff is not needed.
Apsc is going to be the sweet spot, ff is is really going to be catering to a smaller segment of the market until the cameras hit a sweet spot in pricing. Given how well Fuji did w/ their xp1 at 1700, I would expect around the 1500 price range is where the sweet spot will be. A used d600 is selling for around 1600, so that sweet spot is not too far away IMHO. The big issue that would still remain is the need to upgrade to lenses that support ff image. If u have a big investment in dx (Nikon speak for apsc image circle lenses) then it is going to be a sticky point for sure, which is why something like the Ricoh ff m module would be a great nice to have.. A Nikon or canon ff csc (for those who want non dslr solution) for those w/ a lot of legacy fx lenses would also be up there if either of these companies would just get their act together.
Gary
Everything u mentioned about ff sensors I agree w/... Which is one of the main reasons I am not really interested in ff any longer plus over time I have come to realize, for the type of pictures I take, ff is not needed.
Apsc is going to be the sweet spot, ff is is really going to be catering to a smaller segment of the market until the cameras hit a sweet spot in pricing. Given how well Fuji did w/ their xp1 at 1700, I would expect around the 1500 price range is where the sweet spot will be. A used d600 is selling for around 1600, so that sweet spot is not too far away IMHO. The big issue that would still remain is the need to upgrade to lenses that support ff image. If u have a big investment in dx (Nikon speak for apsc image circle lenses) then it is going to be a sticky point for sure, which is why something like the Ricoh ff m module would be a great nice to have.. A Nikon or canon ff csc (for those who want non dslr solution) for those w/ a lot of legacy fx lenses would also be up there if either of these companies would just get their act together.
Gary
Last edited: