« Newthink! | Main | Deals o' the Day »

Wednesday, 11 November 2009


Feed You can follow this conversation by subscribing to the comment feed for this post.

Do you think this idea might have been 'inspired' by the modular RED video camera system?

The last line of this post/conversation, by Ctein, effectively sums up much response to the GXR: hard to get a handle on, but an interesting idea.

The puzzle behind the product may be resolved if the launch, so far, is incomplete. Yet-to-come modules may clarify the larger strategy quite soon. (Or not.)

Is is a potential clue—or perhaps just a red herring—that other Ricoh Company's core compentencies include printer and thermal technologies?

Mike, Ctein, since I don't believe it was mentioned in the discussion, except as a hint in Mike's response to the monochrome sensor (and BTW, I would also buy one) concept, there's a scenario that has been repeatedly described in various discussions here and elsewhere: A module with a FF sensor and adapters allowing use of many types of lenses, especially rangefinder. Based on comments I have seen, such a module could be very popular. So imagine, to take just one example, a 25mp monochrome Kodak sensor in a module with available adapters to fit Leica and other mounts. Or the same with the Sony CMOS, as used by Nikon and Sony. I would think other back units would be appropriate under such circumstances. Here's one tantalizing link (check "Future system development" at the bottom).


Mike wrote, "As a camera reviewer, I like it—it's new, it's fresh, it has possibilities. As a camera user-and-consumer, it looks like a big headache."

Yep... when I first saw the camera I thought of the long-gone 35mm Rolleiflex SL2000/SL3003 system... loved the "idea and look" but quickly returned the loaner to the distributor!

I'd like to see: a two-lens stereoscopic module. Or how about a swivel-lens panoramic module, i.e. a digital Horizont? Both these ideas are pretty far out, but then, so is the GXR concept.

How about a module with a filter-wheel (like the Minolta fisheye) for infrared photography?

There could by modules designed for astronomical photography, incorporating Peltier cooling of the sensor to reduce noise during long exposures.

A module could have a fiber optic endoscopic lens for photographing hard-to-reach places.

A slide-copier module.

What else can you think of?

Many people seem to want B&W only sensors that do away with the Bayer array. In Michael Reichmann's review of the Phase One Achromatic Plus, he found no significant resolution advantage to a pure B&W array. A second opinion by Mark Dubovoy did find a noticeable difference. Still, with the wealth of pixels we have available, how many of us really need the extra resolution that special purpose sensors B&W sensors offer?

Likewise, doing away with the anti-aliasing. Worth the difference?

I can see the attraction of no IR filter and have thought of converting a DSLR to IR. Given that I would seldom use it, I have a (minor) business idea. If you own a rental shop, rent out IR cameras! While there may be few people who want to own one full time, I'll bet lots of people would like to take one for a spin. The closest my local shop offers is a Leica M8.2.

A commenter to the previous TOP post about the GXR wished for swappable sensors to get different looks. Ignoring for now the volume economies mentioned in the current blog post, how much of the look depends on the sensor electronics vs. the software processing of the sensor output?

It seems to me that a system like this will succeed or fail based on things it can do that the other systems can't because they've chosen their sensors for flexibility rather than single purposes. Really good wide angles below 14mm, with perfect rectilinear correction by sensor shape. Small and cheap supertelephotos on tiny sensors that work just fine in good light, for wildlife photographers - imagine a quality 400-800mm zoom that fits in a fanny pack.

I doubt Ricoh's engineering, or their market placement, is really up to that, though.

Just a heads-up: The prime-module is a 33mm f2.5, it's 50mm equivalent. Which makes it a lot more useful for most purposes (and less for macro).

With that, these are effectively two compact cameras for two types of shooters: Those who prefer zooms, and those who prefer primes.

De modular-gizmo just saves them production costs on the screens and buttons.

Once they split up he lens/sensor module (so the camera has three parts), this becomes sensible. Screen too low-res? Upgrade it. Want B&W? swap the sensor. Need a wide-angle? swap the lens. (Sounds familiar? think film).

Until then, this doesn't offer much over fixed-lens cameras, since I doubt those modules will be a lot cheaper than a complete P&S with similar specs. And how often do you want to upgrade the buttons on your camera?


I think you made some error in your post, the lens of the aps-c module is a 33mm, 50mm-e. So this is right now the most compact system with a 50mm-e (e.g. standard) lens with aps-c size sensor, could be appealing to somme poeple...

Considering the system, I think that it's the only one that allows to switch lenses AND sensors.
Then you could use it as an apsc prime lens (e.g. a DMD) with a compact zoom/sensor as a second body. That way you re sure that you won't miss a decisive moment !

But what REALLY bothers me with this system is the body.
The body is the cd player and the modules are the Cds, you must pay for it and it gives you nothing in itself (well buttons and a screen, what what can you do with that ?).
Then comes my main complaint, the PRICE. The body costs you more than a G11 and has no lens and no sensor. Sorry Ricoh but I can't follow you here.

For me this a very big issue, and a strategic mistake by Ricoh. Consider a cheap body (200$, or even 150$), people could buy it and then get (stuck ?) in the system. They'll buy some body at start for sure. But if the want some new compact, the could buy just the module which would be less expensive than a "full" camera. Same if they'd need some macro or portait lens, and so on...

For that's the way the system could work.

P.S. I'm French and this is my first post, so sorry for my English

We are on the threshold of a very tight couple not just between the lens and sensor, but including the processor and software as well. Viewed as a complete, customisable, system the potential advantages are huge.

We have seen with the wide angle distortion from the LX-3, and now the Canon S90, that we can now use lenses with huge distortions and fix the image in software.

Lenses are expensive to make and assemble, software costs nothing to reproduce once it has been created. I suspect that the future holds lenses that are physically smaller, with fewer elements, and clever software and processors tightly coupled to the lens.

At present, as far as I know, it is possible to remove distortion from lenses of the same design as long as they are all completely consistent. A likely future is that lenses are allowed to vary individually, and each individual module calibrates itself automatically by being shown a few charts with lines and colours.

A good example of self calibration is the way that ink jet printers with nozzles in their cartridges now re-align themselves when new cartridges are installed.

It won't just be distortion that is fixed, but colour and fringing and bokeh and all sorts of things that I don't know about ..... The purists will hate it !

If there is brilliant strategic thinking behind this product, then it must be really brilliant since none of us here seem to even come close to getting it.

If Ricoh intends this as an open system where others offer lens-sensor modules, then Ricoh would be reduced to manufacturing a few "shells" (can't call them "bodies" any more) and the others will be making the money selling the lens-sensor modules. The lens-sensor modules must contain most of the electronics too since I imagine these are sensor-specific, so the shell has only a few control dials and buttons, the UI and the rear display.

I'd think Ricoh would do better economically offering a direct competitor to the GF1, even if it's just another "me too."

Nice discussion. :-) Yes, I agree with almost all you two say. Particularly about the marriage of sensors to lenses.

But there's a couple of nitpicks for extraneous issues. Ctein, how video doesn't add to the cost? If nothing else, they have to use stronger/faster CPUs to process the video. I may be wrong, but I don't particularly see manufacturers absorbing that cost. Of course, there is no hard data. There is no D90 or 5D MkII without video to compare the price with the video-capable models.

Mike, who's hoarding Wankel rights? NSU licenced the rights to everybody who wanted it. Yes, Mazda is dominant with Wankels right now in the car industry, but I'd say it's more due to the lack of interest than to Mazda hogging the market. There are other uses, like in aircrafts and small go-karts where Wankel engines apparently do quite well. Besides, RX-7 and RX-8 look like nice cars.

A clever solution to a non existent problem? I don't know. I suppose that it is at least encouraging that Ricoh has moved into the APS-C market and there is the possibility of a fixed prime lens APS-C DMD from them in future.


the possibilities are wonderful, aren't they? but is ricoh too small too succeed? just like wall street firms were too big to fail? i hope the gxr takes off, but the price of getting one is going to hurt.

btw mike, that's 12mp cmos sensor and 50mm-e lens, not 75mm-e

Now if they had interchangeable sensor modules which would accept inter changeable lenses, then that could be interesting but why would I spend all that money on a GXR that more or less duplicates the Caplio GX100 I’m perfectly happy with? I fear it’s a dead end and that Ricoh could waste a lot of capital which would be better invested in more conventional cameras..


for the record, it's not a 75mm-e lens, it's a 50mm-e lens with the APS-C sensor. This makes it a bit more useful, but the concept is still odd.


I think the way to go for Ricoh is making an interchangeable lens Leica X1.

With lens modules you could and should do collapsible lenses for big sensors.

Making big fixed lenses gives you no advantages.

You write “75mm-e prime”, but it is 50mm-e (it has 33mm written on it).

The more I think about the GXR, the more I like the thought actually. And the more I think "it's about the uniform interface". Would I buy a camera where I can be sure that regardless of wether I go for an APS-C prime module or a fingernail-chip with a large-range zoom I will allways have the exact same handling? Sure as hell.

The current problem for me is that different cameras - even when they come from the same producer - have different handlings. I am eyeing a GF1 as a secondary body to my GH1 - and hate the fact that the button layout is not identical and that there are small, but for me significant differences in handling. And it's worse with my compact, as that is a Ricoh GRD II, which actually handles much better than the GH1 in some aspects. A GXR-like system would be the solution.

Another thing you should keep in mind is the fact that this is a Ricoh from the top-of-the-line range. What makes them special? The very high configureability and the (up to now at least) very good update support for firmware. This is very different from many other products in the compact range. So having a uniform body for different modules with the high configureability of Ricoh top-of-the-line models, combined with the expected constant updates to firmware, this sounds like a very good investment in the long run (well, of course only if Ricoh is there for the long run ;) ).

Of course it all can fall down flat on it's face if there are not enough different modules in the end. But maybe it's wrong to look at it as a classic systems camera, because it is very different from that allready - it's more an upgradeable camera. You buy into it like you now buy into the GRD or GX "system", but unlike there, you don't have to buy a new body every now and then to get new features, you can customize your body and even upgrade later if some better matching module comes along.

I think it's exciting. If only for bein different than the rest of the market. Ricoh at least showed with the GRD that it can be different and useful, so I for one welcome the GXR on the market.

Dear Ctein and Michael.
There are some stuff that must be taken into account.
I was the one talking about the Wankel engine. In reality, the Wankel and Diesel engines are similar concepts applied to the competition with the Otto engine, which is the main engine used in US, I´m afraid.

There are four internal combustion engines nowadays in production with significant numbers: Wankel [Rotary Mazdas], Attkinson [Hybrid Toyotas], Otto [regular petrol engines] and Diesel [there are no significant turbine engines out there, such as one Chevy had or the Lotus 56 had]. Each of them represents a different approach to the very same concept: how to take energy out of burning fuel within a chamber.

The winner of all this is not the best engine, but the one which has been further developed due to competition. In principle, the most efficient of them are the Wankel and Diesel engines. But, until recently, both of them were hampered by not enough development, as they were good for the market they were used at.

However, recently [20 years or so], Fiat decided to push forward the Diesel engine and introduced mass market direct injection into Diesel engines. From that moment on, this engine type started to develop further on, when germans realized Italian machinery was going ahead. Diesel engines started to go on [helped by their lower consumption and the feeling of better torque or midrange response], and nowadays it is very easy to see small diesel engines of 1.6l reaching 150cv, or 2.0l engines to 180cv, putting them very close to Otto cycles.

Were Fiat be alone in this development, and had not the German and other Italians [Perkins-DetroitDiesel, owned by VM Motori] got into the race, modern diesels would have stalled. Competition has made the Diesel engine go further the last 20 years than previous 80 years.

Contrary to this example, the Wankel engine had a good start. Two very opposite companies and very powerful industrial conglomerates [Daimler-Benz and Mazda-Toyo Cork Kogyo Co] decided at the same time to start developing the Wankel engine, with the W111 and the Cosmo. Promising as it was, only Mazda went on. Mazda has gone lightyears ahed from the nowadays clumsy engine of the RX7 into the Renesis RX8, but it does have no competition to leapfrog but themselves.

Had there been other companies to further improve the Wankel engine, we might all be driving mad revving cars nowadays.

Not always does the best win this market competition.

"Newthink" indeed!

Instead of lenses as singular objects of admiration, maybe we should be thinking about lenses in combination with sensors?

This seems as if it could be the first compact camera that could truly compete with DSLRs, if the concept ever reaches fruition. You could have all these specialized modules for every photographic application you could imagine like a macro module or a perspective control module. How about the "wedding module" with a square sensor and appropriate lens, or the "sports module" with a monster telephoto and APS-C sensor, or a "landscape module" with a FF sensor. My imagination fails me.

Mike and Ctein,
Thank you for a very informative discussion. Well, here is my two cents worth. I don't think that Ricoh went far enough. If they separated the back, the sensor module and the lens (with a standard mount regardless of sensor size) then you could allow people to truly customise and upgrade the camera system as they like. It would be easy enough to build in a blind/shutter mechanism that closed over the sensor part when the lens was removed, and they already have the slide-in sensor part solved. Stick in a Foveon or a 1/1.7in sensor or an APSC, 4/3rds, CMOS, CCD, infrared sensitive sensor, mix the lenses, even though some may not give the coverage on a particular sensor, that's the photographer's choice, change the back when a bigger, faster processor/hdr processor back becomes available but keep all the other bits. Why not? I think the key here is interchangeability/customisability (don't think that's a real word), not compact size per se. Look how people mix and match large format components. They (Ricoh) could have a back with built-in evf, or not, even a stereo back that held two lens & sensor packs, technical backs, backs with special mount/grip functions, beefy battery packs, backs with GPS plug-ins, it's endless. That would be a new and unique system and one that I would seriously consider investing in. Think of all the special purpose backs, lenses and sensors that could be developed, especially if all the lens/sensor/back interfaces were standardised. Think of the unique and wonderful development that would then occur in the photographic world. Well, if they had just put a 4/3rds lens mount on the thing then they would already be, halfway there.
Best regards
Steve Bruhn

Seems to me you have missed another possible "strategy", a small company desperately looking for a niche big enough for them and too small to attract competition from the big boys.

I wish them well, but can't imagine buying such a thing.


This is about lock in, if most users only but two lenses and one body then the balance of investment is in glass. However, I suspect this is the other way round and most only buy a body and one lens. By modularizing the components of a camera the upgrade investment is probably lower for the customer and yet the expected revenue for the manufacturer could be higher because the customer is locked into their model. This is a bold move but may be based on a flawed assumption about how a camera is used and what photographers consider a good investment as illustrated by your comment toss the lens and sensor.

It could be described as a budget Red, but unlike the Red there is no real value offered.

Essentially there are three elements to a camera, a lens, the sensor and supporting electronics and the user interface – the body. What they have done is to reduce the three to two and consequently limiting the photographers choice. The real innovative approach would be to create a camera that could accept any lens from all the leading makers and also support exchangeable sensors. Then we would be back to the flexibility of film and able to choose the lens and stock to suit the subject.

Also your point about the Favron and the Wankel engine is very valid and supported by history, for example Dupoint's revenues from Nylon grew more than ten times once they had allowed other companies to license the technology.

Now we will have to see what the market makes of this camera?

I like to look to a human explanation for weird things like the Ricoh GXR "system." Imagine that Ricoh has two development teams, each fiercely dedicated to their camera concepts, each believing that only they can return to the glory days of the compact GR film camera, which Moriyama Daidoh made into a cult object. One has come to believe in the DMD with large sensor, the other to believe that the usability of the GX and GR products is the key. Together they convince upper management that by using a common frame, with the rather successful controls that have been developed, the cost of letting all these flowers bloom together will be minimized. Looking at this as engineer-driven with a strong usability bias, I can see it as a lower cost platform for introducing multiple approaches. Not exactly conventional marketing, but this is Ricoh.


I tend to divide my photographic activities into two parts, each requiring a different mind set and often different equipment.
In the first, I have a photo project in mind and will choose the appropriate equipment , or the nearest suitable, from what I have. Hopefully that choice is right for that project and from a technical point of view, appropriate. It could involve a large camera and lens and even a tripod. Weight and bulk may not be an issue.
In the other mode I am intending to take photographs but at that stage have no idea what I will be photographing and in fact may choose to photograph in many varied genres and situations. Typically this is what happens when I go on holiday.
In that situation I find the better small sensor zoom digital ideal for most of what I want to do except for low light work when the APS size sensor would give me the result I want at high ISO's. But I don't want to take two camera outfits with me so elect for one or the other and usually at some stage regret the choice. One is not good enough in low light and the other gets heavy on the shoulder by the end of the day.
The GXR set up would score for me purely on the basis of compact size, light weight, and being able to replace taking two outfits with one adaptable one. It's a little expensive at the moment but prices will probably drop. I would prefer a wider angle lens than the 50mm with the APS sensor, maybe a pancake type - no doubt that will also come. But I can see this sort of lens/sensor combo becoming personally very attractive when we are a year or two further down the line of its development. For me, most of this year's new offerings are about smaller size and portability - a welcome trend away from concentrating development on the larger DSLR.

Mike, ctein, your conversation reminded me of this :

Very entertaining, and the fun is seeing how relevant Ricoh can make this solution to make it economically viable, and to thrive... they're not a subsidised company ;)



I think the GXR is a concept that can appeal to particular photographic needs. Personally, I see two targets with the current strategy (not including exotic modules with other mounts, etc.)

First, it could be a good solution for those looking for a small-sensor interchangeable lens camera. There is a rumor of an ultra-wide module coming up. Besides being the first small-sensor ultrawide, it could also make for a nice combination with the 24-74mm-e module. For small sensors, the cost of duplicating the sensor may not be very significant.

The second target I see are those photographers who want a small carry-around camera, but do not want to give up on low light performance or shallow DoF possibilities entirely. The small-sensor modules allow for tiny zoom lenses (much smaller than m4/3) for general-purpose use, and you can add a single (or perhaps two) large-aperture APS-C modules for low light/shallow DoF use.

In short, I think the system has its merits, but only if you intend to use at least one small-sensor module, and not too many large-sensor modules (due to the sensor cost). Useful additions to the module line-up would be:
- small sensor superzoom [rumored]
- small sensor ultrawide zoom [rumored]
- small sensor tele zoom (small!)
- large sensor 35mm-e (very fast or very small)
- large sensor 85mm-e (f/2 for portraits)
- large sensor M-mount module
I'm curious to see which modules are around the corner...


Hi Mike,

I'm probably person 276 pointing this out, but you mention a 75mm-e lens. That should be 50mm-e. (it is a 33mm lens).
Interesting discussion!

Thanks, Nick

So I'm not the only one unable to figure out whether it makes sense or not.
Being a modulr oblject, I'm thinking what will be kept constant and what will change. Certainly the UI is kept, and it is a good thing but not enough to justify the design. I think what will be kept relatively constant is the size. So you will have a always-compact camera, which can have (hopefully) the highest image quality for that size. You want a normal prime, then you can get it with large sensor. You want a good performance zoom, then instead of getting a large lens you get it with a smaller sensor.
Still, I'm no convinced it makes sense...

Reflecting on this, I see this system working commercially only if a few conditions are fulfilled:

First, the systems is actually deployed as a long-term standard.
Second, it addresses some some massive niches AND also some specific niches populated by photographers with high budgets, allowing a profitability which is not strictly linked to mass sales.
Third, the price of the sensor + lenses gets close enough to that of lenses from competing, massive systems, making the price difference a no-issue.
And fourth, the combination of sensor/lenses is exploited in such a manner that make their image quality outcompete similar lenses in other systems.

Whether all of them will be achieved by the GRX is completely unsure (and possibly improable) in my view. But still, it seems possible.

Gentlemen, there is a subtle but disturbing whiff of 'oldthink' in your otherwise highly informed and stimulating exchange.
For one thing, I think we should salute the first consumer digital camera with interchangeable sensors — at least the first one that I am aware of. That's one seemingly obvious feature that I have been waiting for since, oh, the Sony Mavica. If Ricoh can do it, surely others will catch on?
It's a bit early to tell if and how the matching of sensors and lenses in sealed units will work out, but your basic skepticism at this early stage (some of which I share) somehow reminds me of the comments on the first iPod: who needs another MP3 player, a sealed one, with sealed battery compartment, non-exchangeable disk unit, etc. Well, we all know how that played out and what it grew into, so let's see whether Ricoh can grow, amend and expand a GXR eco-system.
Finally, do you remember Gestetner, Nashua, Rex-Rotary, Monroe, Lanier? Each of them stood for duplicators/copiers/printers. Now gone, swallowed by Ricoh.
Ricoh is not Nikon. Maybe Thom Hogan would italicize that: Ricoh is not Nikon. If I remember the numbers correctly, they depend a lot less on their camera business than even Canon or Sony. Theirs is a niche in the camera market anyway. So, if they want to play in this field, it's not like they're betting the farm. Your risk analysis should take that into account. As the astonishing example of Glock shows, it is wiser, and in the end more profitable, to outreach than to encroach.

Ctein said: "I'm still trying to figure out the rationale for the camera".

I think the rationale is quite simple: Offer an interchangeable-lens compact camera that offers advantages over competitors:

1. a sensor that will never get dust on it, and
2. an even more compact body,

and does so within a design that also incorporates the market's acceptance that improvements in sensor resolution have plateaued but that improvements in image processing "engines" are still being made.

The design also offers "optionability", which is said to be part of the psychological outlook of Generation Y...surely the target demographic.

As if the marriage of sensors and bodies wasn't too much of a disadvantage, Ricoh married sensors to lenses.

However, someone else already suggested that if modules had different mounts (M-mount, K-mount etc) the GXR would garner tremendous interest.

And add to that the fact that it would uncouple bodies, sensors and lenses.

So maybe, as Ricoh sees their married sensors to lenses fail in the market (except as a niche product), they would be able to transform the GXR in success by introducing modules with mounts. And we would have exchangeable and upgradable sensors, while having the same body and not having to toss out lenses. The potential is already built in to the system engineering-wise, the only thing missing is a mount (or maybe it is an already planned module).

Interesting exchange, this. It will indeed be interesting to watch events unfold. personally I find it exciting, if only as an intellectual exercise (people are maddened already to get the correct lens, so forget about providing them with correct lens/sensor combos...)

Going back to the "oldthink" LL column, it was nice to see that you said something true about FF sensors in 2002. (I say that even as the spoiled brat in me clamors for a FF GF1 :))

To me it looks like a solution to a question no body has asked.

At its current pricing, and generally here in Oz we tend to pay double what you folks in the US do, it seems very expensive and I think that is more likely the hurdle it faces rather than the concept.

Nikon was rumoured to be working on an interchangeable sensor camera. Having independent lens, sensor and "base" modules makes more sense.
First, the sensor seems to be a major cost component and the in-camera processing pipeline is a major factor in final image quality.
Second, the grip and control UI does not change that much between generations, and is easily updated via a firmware upgrade to accommodate new sensor features.
Third, having the lens as an independent unit retains the value of prior lens investments. I am looking forward to test mounting Nikkor glass on the GF1. Depending on the camera the 20mm can be a 20mm, a 30mm e, or a 40mm e. I realize there exist technical reasons that may result in lessened image quality. But I use a 400mm so infrequently that a 400mm e would be a fine solution for those few times sucha focal length is required.

What about modules with no lenses just mounts for Canon, Nikon, Sony, Pentax, 4/3, m4/3 etc?

Note that the modules do not have to be limited to sensor-lens sets only. On DC Watch page (http://dc.watch.impress.co.jp/docs/news/20091110_328005.html) you can see they have apparently HDD, projector and printer units as well.

The video hints at additional functionality and interoperability to come for the body/control unit. It may be that the units currently announced are the tip of the iceberg. But yes, the market(ing) may be the Titanic.

This approach still allows for the separation of sensor and lens in future models. The great innovation here is that for the first time in digital (save for the previous Minolta approach), the sensor has been separated from the user interface (or camera, as we used to call it).

IMO this approach will fail because (A) it was born out of greed and (B) it is way too expensive. The greed part refers to the fact that this is an attempt by RICOH to lock consumers into a closed upgrade cycle.

Now consider that Olympus (for example), could come up with a modular micro4/3 camera separating sensor from user interface, and the sensor having a micro4/3 mount. That would kill RICOH instantly just because of the sheer number of lenses that could be mounted (and that many people already own).

Interesting... by the way, I believe the 12-MP APS-C CMOS version module is roughly a 50mm-e (the lens is 33mm), is it not? That would make it very much more to my liking than a 75mm equivalent. As you have said in the past Mike, and I agree, 40mm-e would be better.

Thanks for all the thoughts!

A couple of months ago I would have shrugged the whole GXR thing off as a stillborn idea.
But now, as a happy user of a GRD III, I have gained a lot of respect for Ricoh's engineering sense. The camera's user interface, build quality, excessive customisation possibilities and overall handling make it obvious that the people who concieved it actually care about serious photographers.

The idea of a unified user interface for different camera concepts sounds good to me, especially if the interface is as intuitive as the GRD-s.

My combo of desirable modules would include:
1) small sensor with sensor shake IS + a collapsible 28/2-equivalent (replicating the GRD-III, maybe even f/1.4, as long as they can make it collapse near-flush to the front panel)
2) the already announced 50mm-equivalent macro (wallet groans)
3) small sensor with IS + a fast (~ f/2) normal zoom, entirely dedicated to video
4) APS-C + high quality 35mm-e f/2 collapsible
5) APS-C + 90mm/2 with a matched bayonet-mounted 2x teleconverter

For kicks, they could even build a Micro 4/3 compliant module. The design doesn't require all sensor units to have a fixed lens, does it?

* wipes gearhead drool from beard *

I really hope Ricoh can get enough momentum behind the system to produce the cool stuff. This will never be a mainstream solution but variety of desing is much needed in our world of trusty (read: boring, clunky) SLRs. Go, Ricoh, go!

Ricoh is already a niche player in the digital camera market. I'd hate to see them disappear if this system fails. Some of the cameras they have introduced in the last few years have been very interesting. One other point. They seem to have a really poor distribution network in the US. Perhaps some innovation there would be worthwhile.

It would actually make more sense if they also offered interchangeable bodies. Say, one compact body like this one, one with a good optical rangefinder coupled electrically to the lens, one larger, waterproof with lots of battery power for long nature hikes...

That was a great read! This was one of my favorite posts on TOP in a while. I always enjoy both your (Mike) and Ctein's writing and it was really interesting to see a dialog between the two of you.

I'd be even more interested to read a dialog about image-related issues. I'm puzzled/curious enough by Ricoh offerings though that this made for a great read - thanks.

Interesting thoughts - it seems to me though that the one thing that could make this work, could never be justified economically. Imagine that a 21mm, 35mm, and 50mm B&W only module were available. A dream come true for some people, but how many units of each module would actually sell? Too few I suspect to make production costs realistic - and the more variations supplied, the worse it gets.

To me this camera is neither fish nor fowl. It's a large price premium for what it is in either of the offered configurations, but nor is not truly modular (like Red).

While an interesting idea, I can't identify a single reason why I need (or even want) one.



The level of discussion that this has generated must show the level of interest for 'serious' photographers in re-thinking current systems in the digital age. 'Newthink' is popular.

I though, can't see this working commercially. The most expensive bits of the camera are (i) lens and (ii) sensor (iii) shutter (will stand corrected if this is wrong!) and all these three bits are replaced. One has to buy (almost) a brand new camera if you want to change the lens.

Also, from my perspective, the modules are simply too large. I can't imagine anyone needing a small Ricoh lugging around all that extra - it kind of defies the point. And an odd intital lens combination to launch - what 50mm prime user is going to want a compact type option and what compact user will pay for all this stuff?

That said, I thought that the ISO 1600 samples look great...

The concept of modularity is not exactly new, I think Red has such concept but anyway the big mistake is the coupling of the sensor and the lens. That sounds so inflexible to me that I don't care for this new system. But give me interchangeable lens AND sensor and I would be all over that. That's one thing that we lost when going from film to digital...

'..."that's never going to work."' Thinking outside the box is nice, but if it doesn't make sense, it doesn't make sense.

Before I lose this thought, didn't they say it's about having the optimal lens/sensor configuration for your subject?


I love the conversational style, would you consider this kind of article more regularly?



Thanks for an interesting exposition. I too am struggling with what this format offers as an advantage, given that the lens units are expensive and tied to the sensor - the part you are most likely to want to swap in the face of future developments. However, if you look the other way around this approach may (subject to other lens unit availability) end up offering more opportunities than the entirely fixed lens approach of e.g. the Leica X1 if not the range of promise offered by M4/3.

Other modules than lens+sensor can be attached to the camera: projectors, hard disks, wifi controllers...


On the other hand, mount-equiped modules are possible too... For instance, a micro 4/3 module or a M-APS module. It is no the same to ask Panasonic and Olympus to enter in their system, with nothing in the pockets, than asking that with a working system and lots of potential customers.

I think this is only a starting point. Ricoh is free now to get any direction.

The GXR makes total sense to me. It's a kit that gives me a very compact camera for general use in good outdoor light with ample depth of field to get outdoor scenic shots totally sharp from near to far, with an optional module that isn't too much burden in the camera bag that can be used to get a nice shallow depth of field for that smooth blurred background that we sometimes like. And then when I go indoors into low light, it gives me the ability to shoot with safe shutter speeds without resorting to ugly ISO settings, and if I did need to use a high ISO, the bigger sensor will perform way better than the small sensor in the other module.

All in all, it covers the two most common situations a snapshooter is faced with: good outdoor light and low indoor light - and on top of that, it allows one to achieve extreme depth of field (a struggle with larger SLRs) and a really shallow depth of field too, when needed.

And all of that without ever needing to clean the sensor.

A pretty clever concept, I'd say.

I will be interested if Ricoh has separate lens and sensor modules. That makes a lot more sense to me.

For the future, Ricoh: how separating the sensor module from the lens mount, similar to what RED camera is proposing for their future lineup?


I knew Ctein would get his own post on the new Ricoh, but a dialog is even better, really enjoyed that, and may I propose... you are welcome to do this some time again.

You mentioned my main concern here, that is, will/can Ricoh provide enough modules in the future. And maybe more important, is this modularity intended to be open to other manufacturers or exclusive to Ricoh? If yes, and open, then the prospects are really sweet, it would be a UI-frame with all sort of fixed-prime-, fixed-zoom-, 3rd-party-mounts-, storage-, video-, projectors- and whatever modules. If you don't like to mix sensors, you don't have to. But if you think you need a compact zoom snapper along your prime-workhorse, there you go... I would for sure.

In this sense I don't think the introduced modules are the end, but rather a representative example, a conceptual drawing of the system's nature. Unfortunately many people seem to not get the hint, rather being confused with different sensor sizes!?

This is perfectly logical to me, since very human. But on the other hand I remember that many people (and many of them excellent photographers) -still- rave about the magnificent Leica Digilux 2 for its outstanding lens and classic ergonomics. Then I don't think people would shake their heads if one could suddenly replace that lens and sensor with a larger sensor/prime combo. Or replace the UI-frame with a new one that has better jpeg processing, which was one of main drawbacks of the Digilux at ISO 200 and 400. Heck, people are still paying premium for used Digilux, notwithstanding their limitations, and Thorsten Overgaard still uses it on assignments for high profile clients.

Or see it that way: imagine the 50 Macro was not a module but a large sensor compact like the Leica X1, Sigma DP2 etc... with the seemingly high image quality and excellent ergonomics and build quality that we expect from the Ricoh. Noone would find that odd, rather applaud Ricoh for such a decent product, an evolution of the well received GR and GX cameras. But the optional modularity is perceived with great skepticism. This is somehow odd, still human...

However, I sooo hope that Ricoh succeeds.

As someone who owns a set of Walsh 2 speakers I understand your thoughts regarding how this idea may turn out if more than one company was together on it. All too often companies with a "newthink" idea are so sure that it will be "the" idea that they refuse to let others play in their sandbox and the great idea just dies out. I believe Ricoh has a great idea here but just like the Ricoh Mirai from many years ago (another camera I liked but most others didn't)I don't think it will stick around sadly. Micro 4/3's seems to have the traction to go the distance.

It appears their design is dictated by 2 factors (in order of importance): one is the system has a maximum box size the camera/lens combo must fit into, and second, a perception that the user of zooms and telephotos may be willing to compromise on the additional image quality to be gained from a large (APS-C or bigger) sensor.

I guess I'd prefer two separate cameras: a new GX-300 and a GRD IV with an APS-C sensor. But in their wisdom, Ricoh has chosen to create a single system that allows one to do both.

Because I like their products, I hope their vision is proven sound.

Thinking about this system a bit more, I feel that perhaps a better way to provide a real choice to photographers is to offer different APS-C sensors such as Bayer for high ISO, a true B&W sensor with perfectly acceptable 12,800 ISO results and a great dynamic range, Foveon sensor for fidelity, another sensor optimized for video and free of jellos, and so on. If the sensor was a separate part, then this would be the equivalent of having different films. Each sensor itself can be sold for about $200 with a healthy profit for Ricoh and interesting choices for consumers. With the current design though, you would be hard pressed to justify the cost of adding another APSC lens to the system because you would be paying twice for the same sensor. If they made their mount another module, then you would be able to use your existing lenses too. The system would really take off then - especially if they can come up with clever ways to help with manual focus with old lenses. You would buy Ricoh's own lenses for autofocus and sensor optimized performance but you would have the option of using your own lenses too.

From my own experience, when I am carrying both a compact camera and my DP-1, I always reach for the DP-1 since I value its image quality over the compositional flexibility of a compact zoom camera. I suspect it would be tough for people to justify buying more than one module for this Ricoh camera in its current form unless they can come up some clever ways to use different sensors and different lenses.

For the record, I resent video capability in my camera because camera design is all about compromises.

If video is a high-priority feature, it means that when the designers are weighing things like speed of reading from the sensor vs. noise, the decision is likely to go in a direction against the best interests of what I want the camera for — still photography.

Yeah, like that. I WANT to like it, it's new and different and points towards more flexibility in at least some ways. Possibly they'll be able to make some highly specialized modules that couldn't finance an entire camera, too; that could be neat (IR, B&W, even UV).

But so far I'm not liking it; it seems to me to tie together exactly the things I want to mix and match, the lens and the sensor.

This camera system smacks of desperation. I can't even remember the last time I saw a Ricoh in a store.

I think they are about to go the way of Contax, Yashica, Minolta, and so many other companies who did not have the ability to compete effectively in the transition to an all digital market.

I cannot imagine that sensor cost, especially when you get to the aps-c size, is low enough to justify this approach. What about the shutter, are they going to use an electronic shutter? I don't think that has been very successful for larger sensors.

As for the competition angle, I think that is working with Canon-Nikon since they both are large companies that have deep enough pockets to support the surge in expenses. But if you pair smaller companies like Sigma-Pentax, I doubt either one could support massive expenditures on sensor development.

One of the earlier commenters made a good point that we should not assume that the base is itself permanent. Depending on what flexibility they designed into the lens/sensor-base interface, you can imagine video base, closed-circuit TV security base, usb-tethered base, third-party reversed engineered base, all kinds of different combos of lenses/sensors to match them up with, p&s zoom lens/sensor, high quality-high rez portrait lens/sensor, chain store portrait studio industrial base with single purpose lens/sensor, printer base interface cradle, etc.

Might never happen, of course, but fun to speculate.

We should keep in mind that a lot of the comments on a site like this come from people with DSLRs, some pros, long-time serious camera users and photographically-interested people. There are lots of other people on earth with entirely different needs and desires.

Let it roll boys. Ricoh is dipping deeper in the well and hopefully it will unleash trend of new thought about what a digi cam should or could be, doesn't have to be right the very first time. I think it's incredibly exciting..these guys are taking risks.

The concept of interchangable sensors is very appealing to me. Interchangable lens - sensor combos, are much less appealing.

If you told me I could have a high resolution sensor that performed well within a normal range (say, perhaps, up to ISO 800) and a second sensor that gives up some resolution, but takes very usable pictures up to some very high ISO (perhaps 6400 or 12800), I'm in. This would have the advantage of a control and menu layout that wouldn't change at all. When you tell me I'm locked in to a given lens with the sensor, I'm not reaching for my wallet.

Ctein wrote: "In principle Ricoh can closely match lens and sensor characteristics."

This is a loose parrot of Ricoh's marketing. What I'm seeing is that all the media is now repeating that line as if it means something. I call BS on Ricoh. No one to date has been able to provide a convincing argument of why you need a sealed lens/sensor combo to "more closely match" the two, let alone what the words "more closely match" actually refers to. Even Leica's offset microlenses is a match to lens position (mount), not to specific lens.

Isn't anyone asking the tough questions anymore?

I can see myself buying a single module and body if it was spot on for what I want and enjoy (and my tastes are much the same as Mike's I think)but it doesn't seem economical to me when a Canon 1000D and 50mm f/1.8 or 35mm f/2 is incredibly cheap and capable whilst having a lot of potential for upgrading to other bodies and collecting more lenses as you need them.

I am tempted by the Micro 4/3's stuff because of it's smaller size but I like low light photography so my 40D rarely goes below ISO 800 and is usually higher so right now micro 4/3's looks like more money for less capability in the one area I need it to perform. I expect they'll get better though so I'll wait.

This Ricoh concept looks like it's trying to solve a problem that isn't there. Again I'll wait and see but I don't expect it to be anything except a toy for rich photographers who see a problem they don't have.

It seems that Ricoh has been going for the high-end compact camera market for quite some time. Maybe they felt they were missing the boat on all the Micro 4/3 developments, and I wonder if the quickest way for them to get back into that game was to develop this hybrid system. That way they only had to develop a single new APS-C lens/sensor combo, and could still use their existing zoom/sensor designs in the meantime. Instead of having to develop a whole range of new primes at once, or a more complicated compact large sensor zoom system. If that's the case, it seems like a rushed attempt to get their foot in the large-sensor-compact door and buy themselves more time.

Thanks. Your thoughts basically reflect my own.
Also, I really enjoyed this 'conversation' format. I'd be interested in seeing more posts like this in the future. Even if the topic isn't as quite as unique as the GXR.

That was a good back-and-forth. A couple of half-baked thoughts came to me.

If Ricoh had just come out with an APS-sensored, fixed standard lens camera, wouldn't we, the seekers of pocket-sized quality, be celebrating instead of scratching our heads? I don't recall this great uproar about instant obsolescence when the Sigma DP1 was announced.

And if it's possible to perfectly match a sensor to a specific lens and get better image characteristics than an interchangable lens system could achieve, isn't that something certain photographers are going to want? Maybe Ricoh thinks this segment is large enough to justify the considerable risk of unleashing a crazy-looking idea on the camera world.

I may have missed it in the previous set of comments or in this conversation, but there is one factor that no-one seems to have mentioned: the GXR could just be a way to lock people into an OEM-only system. The (potential) benefit to Ricoh is obvious, but it requires consumers to bite. If people shy away because they don't like the idea of having their consumer choice thus limited, or their camera's range of capabilities thus limited, it could be a huge backfire.

"One module would have the perfect sensor [or, um IS], another module would have the perfect lens. And never the twain would meet."

You mean, like using a Panasonic lens on an Olympus body? Yeah, no one would want THAT travesty.

After all, it might hurt resale value or something ... having those "mismatched" nameplates. How unsightly! How unseemly!

-With belated sarcasm, the Devil's Advocate

Regarding economies of scale etc., it appears Ricoh already has that covered in the various modules' asking prices.

Regarding what-were-they-thinking, that's their biggest concern. Whenever you intro a new product and it's not clearly evident what the benefit is, healthy marketing effort beyond "look at this" is needed to propel.

Tick-tock, Ricoh. Don't make me *think* of a need for your widget. Chances are at least 50-50 I won't be able to, regardless of price. Tell me more.

I'm starting to feel like I have a plausible guess as to what they may be thinking.

Some photographers carry a small point and shoot for fun shots on the go, and take an entirely different system when they want to do something serious. Switching from one camera to another can be a pain due to the different controls and capabilities to tweak the images. Maybe Ricoh is catering to this idea.

They gave us two things -

1) a mid-range point and shoot module that will be small enough for people to carry around and "get the shot" when they don't want to work too hard or carry too much for their photos.

2) A more serious module with a lens that provides true macro and a sensor that should give better low-light capabilities. The 50mm-e lens and the larger format (aps-c) sensor should give them better quality for enlargements, and probably much better high-iso quality than the point and shoot version (testing will tell us more) Yet this 50mm-e is still very portable compared to a dslr, and they don't have to think about the controls. They are the same.

For someone who likes the controls that they offer and finds comfort with the Ricoh system, upgrading to a new module without having to learn a new camera could be appealing over the span of time. I have heard a lot of people get really intimidated over the controls offered with a dslr, as well as just the fact that it is different from what they are used to. Maybe Ricoh heard these people too.

I'm skeptical. I can't help but look at the lenses and wonder "how much am I paying for the sensor, and how much for the lens?" I know this is in some ways a silly question, but nonetheless, it comes to mind. Nevermind that for the body and 50mm (equivalent? I assume? they don't specify. The marketing text implies it is - "an angle of view nearly the same as the human eye") lens I'm looking at $1378. And for a case to carry said lens, an additional $97. And it's hard enough to learn one camera (I assume, having not yet learned to use the one I've owned for several years as well as I ought) imagine having (essentially) a completely different camera with each lens. Each lens has a different processing engine. Lord.

Newthink is great, when it's great - but it needs to offer some fundamental advantage over the old. Micro 4/3 does that, I think. This doesn't seem to. It looks like it'll add clutter and expense, and I find myself searching for what the advantage might be.

I cut my teeth on a Ricoh KR-5, so I have a totally irrational soft spot for the company - but I'll be amazed if this is something that flies.

My real worry about this "concept" camera is that too much time and money hasw been sunk into a camera of unproven marketability and a very innovative and responsive company may be dragged into a financial abyss.

The issues I have with "Mike and Ctein Discuss the Ricoh GXR" follow:
Item 01:
Quote: "Seems to me that for every photographer who'd be excited about a modular camera, there will be two who won't be able to get the sensor/lens combination they want"
- How would a potential buyer know? Right now, a potential buyer looks across the compact camera landscape and does not have a choice of several sensors and lenses for the same body (or backplane or camera abstraction layer). At best, camera bodies can have several lenses (to choose from). So the Ricoh GXR gets demoded because it restores one of the favorite items of film cameras: one could change film type. Having the same option as medium format bodies [of having the option of changing digital (sensor) backs] in a very compact body is a choice that no other manufacturer is yet providing [or promising to provide].

Item 02:
Quote: "A reader named Charles Hueter put it even better in the comments: "Optimizing sealed modules is a nifty idea," he wrote, "but not for someone who disagrees with the 'marriages' offered." "
- Which means you are not a potential buyer for this camera. To make this camera and sensor lens combo into a system a buyer would agree with, said potential buyer would have to know beforehand what would be the ideal sensor/lens combination. How to potential buyers know that *now*?.
They don't. They choose among the camera/ sensor/ lens combinations available now. They choose the closest combination, if that ever makes it into the camera buyers' equation.
The Ricoh GXR allows the well heeled to gain the experience of differing sensor/ lens combinations with the same body and user interface.

Quote: "They've have to get everything right for me to get me to lay real money down, and the chances of that happening are probably slim to none."
- Apart from me repeating that this means you would never buy [into] this system, I ask: how do you know that a grayscale sensor would *not* work for you? At all? Ever?
Converseley, what *would* work for you [in regards to greyscale sensor]? The whole discussion fails to mention what *would* work as possible modules that would make the GXR system attractive for both of you.
The discussion is exclusive [with issues] without being inclusive [of redeeming options that could be made available to improve the GXR system].

Item 03:
Quote: "I'm still trying to figure out the logic behind the combined lens/sensor module"
- Where do you get the idea that all modules have to be sealed units? If the GXR system provided a unit that was a Micro 4/3 [mount] with no lens, and the sensor was greyscale, would this be 'outside the box'?
Because they are sealed *now* does not preclude the sensor units from never having lens mounts.
- How about a UV sensor with a lens coated to suppress IR and nuissance visible spectrum wavelengths?
- Marketing wise: carry several sensor/ lens combos and not worry about dust, in a smaller volume that carrying several cameras. When did this become bad?

I've covered the bigger points and issues I have with this discussion. I appreciate the opportunity to express/ discuss them.

Correct me if I'm wrong, but isn't the majority of the cost of a digital camera the lens and sensor? If so, how much would I be saving by buying the GXR with two modules, as compared with buying two equivalent high-end point-and-shoot cameras like the Sigma DP-series, the new Leica X1, Panasonic LX3? There must be an extra cost associated with breaking the camera in half as Ricoh has done, not to mention the "cost" of being tied to a particular camera system rather than having the flexibility to buy any appealing point-and-shoot camera.

Having no photographic knowledge to add, I keep thinking back to the first Sony Walkman which arrived on these shores ready to meet a need no one knew they had. It worked and grew for many years. Sony has since been displaced in the MP3 wars, but could Ricoh have that kind of entry here? Could this be the camera that brings a whole new demographic into photography? Admittedly, the entry price point is quite high here.

It seems to me that this camera is aimed at *me* - that is a person with aspirations of being a "good photographer" who generally "just" shoots a bunch of snapshots. This kind of person now has to ideally buy two cameras: The point and shoot for the snapshots and the DSLR for the serious stuff.

This new system (especially the first two lensors) seems tailor made for this. The zoom and small sensor would be used for most snapshots, but when I am on the rim of the Grand Canyon at sunset, out comes the APS-C for the serious stuff.

Of course, since each lensor costs as much as a small camera, I don't know that it will go very far.

A simpler way to look at it....Ricoh announced two cameras yesterday...

- a small, lightweight, fixed-focal length APS-C camera with very good low-light performance. At $1400, it undercuts the upcoming Leica X1 significantly, has a larger sensor than competing (albeit slightly cheaper) M43 competitors, and should be significantly more usable than Sigma's DP2. Plus, it's future-proofed with swappable lens/sensor units.

- a successor to the GX100 and GX200. Small sensor, good optics. At $1000, it's a premium price when compared the Canon G11 and Panny LX3, but Ricoh has the best user controls in the business. Plus, it's future-proofed with swappable lens/sensor units.

Viewed this way, the APS-C camera's pricing almost makes sense. As a former owner of a GX100, I'm still wondering if the "future proofing" of the zoom is worth the price premium.

I love my GX200, but I'm thinking of adding/moving to the MFT format because the lens offerings and larger sensor size seem a great compromise between my GX200 and my big, heavy Nikon gear. This GRX option hasn't got me drooling at all yet - their prime lens looks huge and slow compared to the Panasonic 20mm 1.7, the zoom is a limited range, and I can't graft on Leica et al lenses if I wanted. I'm scratching my head along with you. I have a fair amount loyalty to Ricoh for their photographer-centric thinking, but I haven't grokked this offering yet.

Thats a great conversation, thanks!

I was wondering, maybe it will succeed more as a very specialized camera - weren't some other Ricohs very specialized as well?. I can think of a microscope adapter module, a pinhole module, some modules with some unusual lenses, and so on. I'd buy that, if the price was not too exorbitant.


Ok, I just had a brain-flash. Suppose they take the modular approach a bit further:

1. They develop a lensless, MFT-friendly sensor module that you can mount a MFT lens or Leica etc. lens on so you don't have to rely on their lens/sensor combinations.

2. They develop a back that you can slide your iPhone or iPod Touch into, and have it function as a super-high def viewing/focusing screen, touch controller, and memory module, with the mounting case providing some dials and other interface controls.

THAT would be a game-changer! I'd jump all over it.

Doesn't the current Ricoh combination offer the smallest form factor with the least compromise? By having different sensor sizes, the lens can be minimized yet still achieve the highest relative IQ. Don on the zoom lens, it is as small as a point a shoot. Don on the fixed lens, it is still smaller than the current M4/3 models. It's like carrying a point and shoot with zoom and a M4/3 camera with fixed lens at the same time, but in a very small form factor. Plus, all the images are in one memory card, makes photo management that much easier. I think the oldthink is that most people here want flexibility and high IQ, not the maximum in portability. For somebody who refuses to carry anything larger than a typical point and shoot camera, wouldn't this combo make sense?


The IR filter on Sigma SLR's is in front of the mirror box and user-removable. No modifications necessary. They are mid-range in price, new.

Plus, you could see for yourself whether the lack of an AA-filter or Bayer array makes an appreciable difference.

I'm not sure I really buy into the advantage of having lenses so closely "optimized" for a given sensor. People seem to get some mighty good results pairing all sorts of vintage and mismatched optics with existing camera bodies.

I hope, as with others, this design is just a prelude to separate lens and sensor/ processing modules. That would be much more appealing.

I can't for the life of me get past the problem that the "body" of the GXR is the cheapest part of any camera, while the lens/sensor assembly is the most expensive. Ricoh could probably have made entire cameras that were as inexpensive as the back+module.

The multiple sensors don't seem to solve anything, either -- are current sensors so bad that we need a different sensor for every purpose? There are a few niche things where different sensors might be useful (IR and...?), but again, building a whole camera wouldn't cost much more, since you're only duplicating the cheapest part.

And basically, two cameras with interchangeable lenses (m4/3) are better than one back plus relatively bulky module, because then you have a backup. If anything goes wrong with the "body" of the GXR, it doesn't make any difference how many modules you have -- you're out of luck.

The whole thing sort of baffles me. I can see no *real* advantages, and lots of disadvantages. Among the few advantages I can see, from Ricoh's point of view, is that the camera gets lots of comment and the company gets credit for NewThink, but that's all PR. PR won't help shoot a photograph.

The strange thing is, this camera would be great if the lenses were interchangeable. It doesn't even have to be small, it would be a true system camera. I really do hope Ricoh will change their mind and some day release a sensor unit sans lens, ment for adapting lenses. The beautiful part of this concept is, that such a thing would be even possible!

I hope Ricoh sends you a review unit, Mike. After the initial fuss dies down, the big questions will be about how good the Ricoh is to use in the real world and about its optics: your forte.

Anyone comparing the price of this oddball unit to a Canon G11's hasn't looked closely enough at the dpreview.com samples. The Canon and its tiny sensor aren't in the same league, even after the move back to 10MP.

I like the Mike & Ctein format, BTW—great match. More!

In the future, one might be able to order a customized lens/sensor combination... long, wide, fast or slow lenses, primes or zooms, large sensors or small sensors, color or B&W, anti-aliased, IR, 3-D, etc.

Ctein says, "I think most one-company technologies are one-company technologies because they're not honestly interesting enough to develop competition. I don't recall that Foveon was in any way restricted to dealing with Sigma alone. They couldn't drum up other interests. (And, by the way, I agree with you that full-color integrated pixels are a lot more elegant than Bayer arrays.)"

I am not privy to the negotiations that went on between Foveon and the Japanese camera manufacturers in the early part of this decade but I have read that Foveon executives' pushy approach didn't go down very well with the Japanese executives. They figured that consumers would not be able to tell the difference between a Foveon sensor image and a Bayer sensor image. Add to that the fact that they had already invested R&D funds in developing the Bayer sensor technology further. If Foveon had more sensible marketing, it is entirely possible that we would be seeing a different landscape today. As it stands today, Bayer sensor technology has made great advances in increasing the resolution and improving high ISO performance, both areas where Foveon is sorely lagging behind. We'll find out in next few years if Sigma's acquisition of Foveon last year is a boon for all of us or a death knell for a promising technology.

It's interesting to watch, even over the course of just a day or two on a single blog, people actually working to wrap their heads around a new concept. So an excellent idea, Mike and Ctein, to expose your own thoughtful and informed process.

Of course, it's no surprise that in a world where photographers stake out passionate positions on things like viewfinders and lens cap designs or frame rates, there will be many who simply won't or can't try to grok a new approach.

I wish we could hear some thoughts from industrial engineers and designers, and managers, too, because I think the advantages for Ricoh are quite significant.

First of all, to be able to devote limited resources over a development cycle to designing, prototyping and testing just an imaging chain, for example, rather than an entire camera, has to have some serious impact.

Second, judging from the list of planned modules, Ricoh obviously sees an opportunity here to leverage its expertise and capabilities in a number of areas--imaging, printing, IT, communications--without some of the typical costs and complications of bringing different teams together on a single product.

A comment about John Camp's comment: if this idea proves viable, the prices are likely to change significantly. An APS-C sensor might today be costing Ricoh today twice what it might cost in a year's time, given sufficiently robust sales. (The sensor design would be a year older by then and Ricoh would be in a stronger negotiating position.) The idea would only survive if, say, a 12MP sensor had a decent lifespan—I'd say it does.

Thom Hogan mentions that Leica's offset microlenses are a response to the lens mount rather than the lens. True, but aren't both the lens mount and the sensor (Leica only did this for the full-frame Kodak sensor, after all) just a part of the Ricoh module? If so, the argument about optimisation is vaild.

I received this new system with a smile, because the concept has been in my mind for a couple of years (truly :). The idea is not perfect but encompasses some clear advantages and provides, to some point, a unique combination of features, quality and convenience (as has been already pointed out, the idea of just picking the right combination for the day -holiday, portraiture session, street reportage,... - is very appealing, specially on the move; probably you will not mix choices within a project). I think current alternatives to what the gxr system provides are not so cheaper in relation to the ricoh's offer, and they lack the versatility.
Of course, the cost of the individual aps-c-like sensors in each lens is against the system's viability. That's true, if quite a relative factor. Sensors are getting cheaper... but the system has to be viable now. Sensor-only modules (or interchangeable front-lens elements, like in cameras from the past) could be a solution.
We are looking now at ricoh´s first two offerings: a general purpose one, and a premium prime at a premium price, the later probably made more to impress than to mass-sell. Maybe they decided to just launch the system that way.
I think I could (progressively) buy on the system if I could get quality 24-70/2.8 plus 35e/2/apsc for under $1500. With ricoh g-handling.
Of course I d'ont dare to say if the system will succed or not, but the concept makes (some) sense to me.

I'll echo previous comments on the discussion format... more please.

This discussion evoked comments that are remind me of previous posts on other blogs about a modular camera. In those discussions I alluded to my Mamiya 645 and the beauty of modular design. This discussion/comment thread clearly shows that the concept is the next evolution in camera design and it's time for a wake up call. Which I think is exactly what the Ricoh is. A potential game changer. Not new think, we've been throwing this out for a while, but new execution, which is far more valuable.

My Nikon D700 is rugged, well built, and if my D200 is indicative, will be useful for a long time. So why should I have to buy a whole new body when the D700x/s comes out? If I want video, I should be able to "plug" in a video optimized sensor to the already perfectly great body. If I want higher resolution, then I should be able to buy that in a sensor.

Kudo's to Ricoh. Bring on the change. It's well past the time for this.

Mike, sounds like a whole separate discussion could be generated about a modular camera.

Mike writes: One module would have the perfect sensor, another module would have the perfect lens.

Logic similar to Ricoh's has been used successfully in the audio world for some time, by way of amplified speakers. Done well, this allows the engineer to design the amplifier(s), crossovers, and the speaker drivers as a single system with high performance. See Genelec's line of studio monitors as a long, successful line embodying this design philosophy.

I'll grant that the above comparison is incomplete; speakers aren't swapped out nearly as often as camera lenses. However, Ricoh may gain additional leverage through this model by reducing body design costs. There will be a common hardware and software protocol by which the lens/sensor modules talk to the body. If we allow ourselves to consider each combination as a separate camera, Ricoh can now produce an entire product line of different cameras without having to redesign the body once. Firmware changes, if any, should be modular and incremental, reducing QA costs.

I might have missed it, but I'm wary of the idea of an interchangeable sensor. The assumption seems to be that we could do this at home. How would you ensure it is installed flat? Would it need its own housing, like the combined lens/sensor? Are people really just asking for a 35mm or APS-c back a la medium format? What about environmental sealing? Is this an example of people wanting even more choice? I have always believed in simplicity of design, because I am fundamentally thick, and find simple processes/operations easier to work with.

The comments to this entry are closed.



Blog powered by Typepad
Member since 06/2007