Is there time for an interchangeable sensor DSLR?

Started Oct 19, 2012 | Discussions
Astrophotographer 10 Forum Pro • Posts: 12,868
Re: Trying to answer

Good post, thanks.

Greg.

 Astrophotographer 10's gear list:Astrophotographer 10's gear list
Sony a7R II Sony a7R III Sony FE 55mm F1.8 Zeiss Batis 85mm F1.8 Zeiss Loxia 21mm F2.8 +1 more
marcio_napoli Senior Member • Posts: 1,451
Re: Trying to answer
1

Hi Grevture

Thanks for clarifying !

Reading your post, I tend to agree, specially on the not commercially doable part.

So true... after all, these companies need to make money.

But if you allow me, I'll throw in one last argument on why I believe major makers never tried it.

For each Nikon F6 sold, several backs are flying out the door.

You keep the body for 10 years (just like you keep the lenses), and upgrade backs.

But the problem is that those backs would not generate large profit for Nikon. A considerable part goes to the sensor maker, in this case Sony, Kodak, Aptina... whoever is behind the sensor itself.

That's not a really great deal for Nikon, and Nikon could go out of business selling so few bodies.

So that adds to the equation, and it seems it really isn't commercially doable.

-- hide signature --

Marcio Napoli

www.marcionapoli.com

Grevture Veteran Member • Posts: 4,188
Re: Trying to answer

marcio_napoli wrote:

Hi Grevture

Thanks for clarifying !

Reading your post, I tend to agree, specially on the not commercially doable part.

The commercial aspect is where most ideas which show up in forums tend to fail

So true... after all, these companies need to make money.

Sad, but true.

But if you allow me, I'll throw in one last argument on why I believe major makers never tried it.

For each Nikon F6 sold, several backs are flying out the door.

You keep the body for 10 years (just like you keep the lenses), and upgrade backs.

But the problem is that those backs would not generate large profit for Nikon. A considerable part goes to the sensor maker, in this case Sony, Kodak, Aptina... whoever is behind the sensor itself.

Now, keep in mind many of Nikons sensors are their designs, meaning they own much of the intellectual properties (IP) within it. Which also mean they own much of the income from making them.

That's not a really great deal for Nikon, and Nikon could go out of business selling so few bodies.

The thing is, the camera body represent a smaller part of the total value, it is the sensor package (sensor + processor + firmware) which is the expensive part, not the body and its buttons, wheels, screens and whatnots.

Read one of Joesephs posts which is quite enlightening:

http://forums.dpreview.com/forums/post/50108700

-- hide signature --

I don't suffer from insanity, I enjoy every moment of it!
By the way, film is not dead.
It just smell funny

 Grevture's gear list:Grevture's gear list
Nikon D70s Nikon D3 Nikon D3S Nikon AF-S Nikkor 17-35mm f/2.8D ED-IF Nikon AF Nikkor 50mm f/1.4D +7 more
marcio_napoli Senior Member • Posts: 1,451
Re: Trying to answer

I understood what you mean, Grevture, but if you say so, doesn't it sound like good business to Nikon after all?

Well, so they could (for example) sell a FF digital back for 1.5k.

Imagine you could buy the D800 for 3k, and a module sporting the same 36mp sensor for 1.5k, ready to use on your F5.

Nikon would sell tons of both, I believe.

And following your logic, they could have tremendous profit doing so.

So back to my original question, why is it not commercially doable?

And under the risk of hijacking this thread, I'm gonna explore Silicon Film's concept even further.

I've found this info, right from DPreview's PMA 2001 show report.

"the unit itself has a built-in capacity for 24 images (64MB) after which time it must be inserted into the E-Box and its contents either transferred to a computer or CompactFlash card.

The second limitation is that the relatively small 1.3 megapixel CMOS sensor uses only about 30% of the center of the frame, this means that when looking through the viewfinder you have a small field of view (marked out by a supplied rub-on transfer) which equates to a 2.58x focal length multiplier, thus a 28mm lens becomes 72mm.

Lastly it only currently supports certain camera models: Nikon F5, F3, N60/N90 and Canon EOS-1N, EOS-A2, EOS-5."

It's ancient 2002 tech.

I mean, max storage of 24 shots per run? 1.3mp? 2.6x crop?

It's clear why it never went into production. 2002 tech did not allow a proper implementation.

What if you had these 2012 specs:

- internal storage for 2.000 shots.

- 24mp

- FF sensor

- could be used on a variety of popular film cameras (probably we all have some of those in our closets): F100, F5, F6... Leica M6 anyone ? etc

My point is: with the original specs, it was clearly headed downhill right from the start, even in 2002.

The old D100 runs circles around it.

But... what about these very doable 2012 specs?  Seems very easy to sell.

-- hide signature --

Marcio Napoli

www.marcionapoli.com

glacierpete Senior Member • Posts: 1,917
Re: Is there time for an interchangeable sensor DSLR?

bgbs wrote:

It will solve many problems. For instance, why should I be limited to 14mp sensor on a D4 body? Why can't I put 36mp sensor on it? So many people complain that 12mp is too small, 36mp is too much, and 24mp on a D600 body although looks desirable, but because it is a semi-pro body, with scenes dial, less fps, less speed, less....you name it, it starts to miss the mark of what you want in a perfect body.

I think buying a camera that you want with the sensor that you want (we do this with lenses already, or MF cameras do it with digital backs) would make for a more tailored solution. Nikon would win by selling sensors, lenses, and cameras together or separately. Say I have an aged camera, but I do not care to upgrade to the latest camera, but I'm only interested in the latest sensor, would it not be great to only purchase the new sensor for my older body.? Or say I want to quickly convert my camera to DX format, I just swap the sensor and get more reach for wildlife.

The modular approach from a customer loyalty perspective, will tie a customer to a brand for life.

It would be a financial nightmare for camera manufacturers.

Finally they are in the beautiful world of "upgraders". Just buy a new camera every 2 years. As nice as it sounds,  it is not in their financial interest.

And behind the sensor is a lot of electronics in a body processing the image for this specific sensor.

But being on the consumer side, I like the idea.

lovEU Veteran Member • Posts: 3,135
Re: More B&W and IR...

Joseph S Wisniewski wrote:

The CFA reduces DR

Actually, it increases it by 1-2 stops.

Interesting.  It still strikes me as counterintuitive since I gain DR by suppressing the blue and green channels with a magenta filter.

Struck me that way, too. The magenta filter trick (Bernes and Taplin, I go back a long way with this sort of stuff) actually decreases dynamic range. It also decreases the workload on the software, because all the channels tend to blow at the same time, so the algorithms that reconstruct a blown channel from the remaining ones don't cut in as often.

I don't understand how "The magenta filter trick (...) actually decreases dynamic range." Not sure if Iliah would agree on this.

", ... because all the channels tend to blow at the same time, so the algorithms that reconstruct a blown channel from the remaining ones don't cut in as often"

This is true, obviously. But doesn't it only mean one has to watch out even more than usual not to overexpose a shot when using a magenta filter (in cases where the green channel is fairly ahead of the other channels)?

-- hide signature --

regards, eric

OP Theodoros Fotometria Senior Member • Posts: 2,090
Re: Trying to answer

The thing is Marcio, that you keep talking about a Digital Back that could fit a film camera too... Interchangeable sensor is different... It retains the processor and memory/buffer as well as the recording media (cards), while you only change the sensor.... this is much less costly, more flexible, while at the same time it gives the opportunity to the user to adapt the camera to his needs... a digital back still retains  a unique sensor with frame, cooling system, screen, memory, buffer and recording media, which makes it a camera with n shutter or viewfinder... Clearly the OP's aim is for a discussion about the "interchangeable sensor camera" not for an interchangeable back one....

-- hide signature --

Theodoros
www.fotometria.gr
www.fotometriawedding.gr

Zardoz
Zardoz Senior Member • Posts: 1,249
Re: Trying to answer
4

Theodoros Fotometria wrote:

this is much less costly

[citation needed]

Clearly the OP's aim is for a discussion about the "interchangeable sensor camera" not for an interchangeable back one....

Referring to yourself in third person isn't a good sign.

Grevture Veteran Member • Posts: 4,188
I think you might have misread slightly
2

marcio_napoli wrote:

I understood what you mean, Grevture, but if you say so, doesn't it sound like good business to Nikon after all?

Well, so they could (for example) sell a FF digital back for 1.5k.

Imagine you could buy the D800 for 3k, and a module sporting the same 36mp sensor for 1.5k, ready to use on your F5.

But to make it work on a F5 you need a clunky solution akin to the old Leica DMR which was truly horrible. And you would only get - at best - something like a DX sensor to fit in a film camera, not a FX one. And you would get the same immense problems controlling the shutter as Leica DMR had (basically one shutter button tripping the shutter (in the camera) and one one triggering the sensor readout (in the sensor module). And you need a lot of connections from teh F5 that does not exists, like from the exposure meter and the AF to confirm to the sensor module when they are done so you can take a picture. And you need a lot of space for the sensor (which is a whole lot thicker then film, and you want a LCD screen behind it too) ... And where to we put the battery necessary and the rest of the circuit boards, the memory cards ...

Agian you would end up with something like the DMR who had the most abysmal handling I have ever experienced. And that camera did not even have to support AF (since Leica R never had AF).

Nikon would sell tons of both, I believe.

They are selling tons of the D800 which isn't all that expensive all things considered. Looking at the size of the potential target audience for the contraption you want, how many units could you sell, realisticly? Considering you can buy excellent image quality in one single much nicer and lighter little unit for around 650 USD? (the D3200). Not to mention getting a D600 or D800?

And following your logic, they could have tremendous profit doing so.

So back to my original question, why is it not commercially doable?

See above, and I think you slightly may have misunderstood what I wrote

There is a huge difference between the price of a sensor package built to work inside a digital camera with all the supporting infrastructure in place, and on the other hand a sensor package which need to connect to an old mechanical contraption not designed to house a digital sensor in the first place and all the necessary supporting technology around the sensor.

And under the risk of hijacking this thread, I'm gonna explore Silicon Film's concept even further.

I've found this info, right from DPreview's PMA 2001 show report.

"the unit itself has a built-in capacity for 24 images (64MB) after which time it must be inserted into the E-Box and its contents either transferred to a computer or CompactFlash card.

The second limitation is that the relatively small 1.3 megapixel CMOS sensor uses only about 30% of the center of the frame, this means that when looking through the viewfinder you have a small field of view (marked out by a supplied rub-on transfer) which equates to a 2.58x focal length multiplier, thus a 28mm lens becomes 72mm.

Lastly it only currently supports certain camera models: Nikon F5, F3, N60/N90 and Canon EOS-1N, EOS-A2, EOS-5."

It's ancient 2002 tech.

I mean, max storage of 24 shots per run? 1.3mp? 2.6x crop?

It's clear why it never went into production. 2002 tech did not allow a proper implementation.

What if you had these 2012 specs:

- internal storage for 2.000 shots.

- 24mp

- FF sensor

- could be used on a variety of popular film cameras (probably we all have some of those in our closets): F100, F5, F6... Leica M6 anyone ? etc

My point is: with the original specs, it was clearly headed downhill right from the start, even in 2002.

The old D100 runs circles around it.

But... what about these very doable 2012 specs?  Seems very easy to sell.

The thing is, they are not doable, not even today. First because you would never get beyond  DX sized sensor, and probably not even that big. Probably more like Fourthirds size (with a crop factor of 2). The space within a film camera is literally 24x36 mm while a FX sensor is usually significantly larger then that.

And again, consider how large the target audience really - realistically - is when we have cameras like D3200, D5100, D7000, D600, D800 and D800E to compete with (not mentioning all the cameras from other brands like the excellent little NEX cameras, Canons rather impressive lineup, the very good m4/3 cameras and so on ...

So, realistically: a small (I'll say tiny) audience, a technically complex product (making it connect with various old cameras will be no trivial task) making the digital film rather clumsy and thus - very, very expensive. Not easy to sell ...

Some previous posts from Joeseph to ponder, belive me, he knows what he is talking about:

About the digital film:

http://forums.dpreview.com/forums/post/50109809

About the physical size requirements for a FX sensor and why it is not easy to make replaceable:

http://forums.dpreview.com/forums/post/50108408

About why the replaceable sensor in a digital camera (as the OP want) is problematic, here Joeseph and signature 'kb2zuz' together recount a number of problems to deal with:

http://forums.dpreview.com/forums/post/50109727

-- hide signature --

I don't suffer from insanity, I enjoy every moment of it!
By the way, film is not dead.
It just smell funny

 Grevture's gear list:Grevture's gear list
Nikon D70s Nikon D3 Nikon D3S Nikon AF-S Nikkor 17-35mm f/2.8D ED-IF Nikon AF Nikkor 50mm f/1.4D +7 more
Joseph S Wisniewski Forum Pro • Posts: 34,157
Digital backs, may not be reading this right...
2

marcio_napoli wrote:

I understood what you mean, Grevture, but if you say so, doesn't it sound like good business to Nikon after all?

Well, so they could (for example) sell a FF digital back for 1.5k.

Imagine you could buy the D800 for 3k, and a module sporting the same 36mp sensor for 1.5k, ready to use on your F5.

There are two things wrong with that line of reasoning, unfortunately.

First is your price. There's only about $900  worth of "camera" in a D800, the other  $2,100 is the "digital back". Now, digital backs for an existing SLR won't sell in as high quantities as new DSLRs. (I'll get to why in a second) so their prices will be even higher than the $2,100 you'd expect from the difference in the prices of D800 and F100.

Second, you have to differentiate between a digital back for a brand new SLR designed specifically to accept one, and a digital back designed to work with an existing F5.

  • Existing film SLRs can't take FF digital backs. The sensor frame protrudes past the image plane and has to go through the camera's film gate. It's about 8mm bigger than the sensor, so 1.3x crop is the limit for digital backs on existing SLRs, like the Kodak DCS-760 or Leica DMR.
  • Existing film SLRs can't do TTL flash with digital sensors, because the sensors don't reflect like film does. The Nikon F5 that Kodak used for their DCS series was not stock, it was torn down, the shutter curtain repainted to reflect better, and the flash system modified to use a "preflash" that metered off that curtain. This worked horribly.
  • The F5 focusing system is flat-out not up to the demands of a D800 sensor. I tried using an F5 for high resolution film work with Tech Pan film: it was a dismal failure.

Nikon would sell tons of both, I believe.

I did a design study and coordinated the market research for a company designing a pretty cool digital back, and the market research shows that when the credit cards actually hit the table, very few customers won't put up with all the suckage of a back/body combination in order to save a few $ or keep a sentimental "favorite"camera alive a few more years.

And following your logic, they could have tremendous profit doing so.

Believe it or not, DSLRs are basically loss-leaders that get you buying the profitable items, lenses, batteries, vertical grips.

So back to my original question, why is it not commercially doable?

Which? A new SLR built for digital backs, a digital back for F6 (and another for F5, another for F100, etc. etc. etc.), or the OP's concept?

And under the risk of hijacking this thread, I'm gonna explore Silicon Film's concept even furtherGogol

Go for it. I'm a bit of a historian on that scam. And Kirsten Schmidt is an old friend.

The dpReview editor puked again. Nothing good past this point. I'll follow up in another post.

"the unit itself has a built-in capacity for 24 images (64MB) after which time it must be inserted into the E-Box and its contents either transferred to a computer or CompactFlash card.

The second limitation is that the relatively small 1.3 megapixel CMOS sensor uses only about 30% of the center of the frame,  looking through the viewfinder you have a small field of view (marked out by a supplied rub-on transfer) which equates to a 2.58x focal length multi

Lastly it only currently supports certain camera models: Nikon F5, F3, N60/N90 and Canon EOS-1N, EOS-A25."

It's ancient 2002 tech.

I mean, max storage of 24 shots per run? 1.3mp? 2.6x crop?

It's clear why it never went into production. 2002 tech did not allow a proper implementation.

What if you had these 2012 specs:

- internal storage for 2.000 shots.

- 24mp

- FF sensor

- could be used on a variety of popular film cameras (probably we all have some of those in our closets): F100, F5, F6... Leica M6 anyone ? etc

My point is: with the original specs, it was clearly headed downhill right from the start, even in 2002.

The old D100 runs circles around it.

But... what about these very doable 2012 specs?  Seems very easy to sell.

-- hide signature --

Marcio Napoli

www.marcionapoli.com

-- hide signature --

Rahon Klavanian 1912-2008. Armenian genocide survivor, amazing cook, scrabble master, and loving grandmother. You will be missed. Ciao! Joseph www.swissarmyfork.com

 Joseph S Wisniewski's gear list:Joseph S Wisniewski's gear list
Nikon D3 Nikon D2X Nikon D90 Nikon D100 Canon EOS 5D Mark II +43 more
Joseph S Wisniewski Forum Pro • Posts: 34,157
The lingering silicon film...
4

marcio_napoli wrote:

And under the risk of hijacking this thread, I'm gonna explore Silicon Film's concept even further.

The original thread was destroyed by the OP himself, so don't worry.

"the unit itself has a built-in capacity for 24 images (64MB) after which time it must be inserted into the E-Box and its contents either transferred to a computer or CompactFlash card.

The second limitation is that the relatively small 1.3 megapixel CMOS sensor uses only about 30% of the center of the frame, this means that when looking through the viewfinder you have a small field of view (marked out by a supplied rub-on transfer) which equates to a 2.58x focal length multiplier, thus a 28mm lens becomes 72mm.

Lastly it only currently supports certain camera models: Nikon F5, F3, N60/N90 and Canon EOS-1N, EOS-A2, EOS-5."

It's ancient 2002 tech.

I do not believe so. I believe it's outright fraud. Have a look at their SEC filings for the 1999-2002 timeframe. No engineering expenses, aside from payments to a local machine shop. How could the things on display in 02 be anything other than solid blocks of plastic and metal, totally non-functional.

The whole concept of shooting 24 shots without a review image until you download the whole cartridge makes for a perfect staged demo.

Imagine that they ...

  • Set up a background in their office and shot 24 shots of their own people posing with a real camera, Kodak, to get a similar sensor to what they claimed to be using.
  • Took that computer, that background, those lights, the same people in the same outfits to the show, then shot more or less the same poses with a non-functional cartridge.
  • Then they displayed the images they brought with them, as the "downloaded" images from their cart.

I mean, max storage of 24 shots per run? 1.3mp? 2.6x crop?

It's clear why it never went into production. 2002 tech did not allow a proper implementation.

Of course it did, if you had $25M of real money.

What is clear is that there is not a $25M business case.

What if you had these 2012 specs:

- internal storage for 2.000 shots.

- 24mp

- FF sensor

You can't have FF. Nothing bigger than 1.3x crop. That was covered elsewhere in this thread.

- could be used on a variety of popular film cameras (probably we all have some of those in our closets): F100, F5, F6... Leica M6 anyone ? etc

Can't happen. They have different relationships (on the lateral axis and the depth axis) between canister and film plane.

My point is: with the original specs, it was clearly headed downhill right from the start, even in 2002.

The old D100 runs circles around it.

But... what about these very doable 2012 specs?  Seems very easy to sell.

No, it's really not. I worked up the numbers around 2007. I really wanted to build it.

  • Plastic molded back that would be done custom for each camera, the only part that varies from camera to camera. The other three components snap into the back.
  • Sensor unit, snapped onto the front of the back, which protrudes through the film gate, and is obviously limited to 1.3x crop.
  • Power electronics unit, snaps onto the front of the back and occupies the battery chamber.
  • Control unit, snaps onto the back of the back, and has and LCD and 8 buttons.
-- hide signature --

Rahon Klavanian 1912-2008. Armenian genocide survivor, amazing cook, scrabble master, and loving grandmother. You will be missed. Ciao! Joseph www.swissarmyfork.com

 Joseph S Wisniewski's gear list:Joseph S Wisniewski's gear list
Nikon D3 Nikon D2X Nikon D90 Nikon D100 Canon EOS 5D Mark II +43 more
Royce - Into the Night Photography
Royce - Into the Night Photography Forum Member • Posts: 56
Re: Is there time for an interchangeable sensor DSLR?

But then Nikon and Canon couldn't sell us the next generation camera, and make more money! As a professional photographer, I kept the same Nikon film camera bodies for about 10 years before upgrading. With digital I've had to upgrade about every 2-3 years to a new body

However, I've able to keep the same 2 lenses I bought for my first FX camera. It's still working great after almost 10 years, and I expect another ten!

-- hide signature --

Royce Bair
Into-the-Night Photography
(featured photographers, tips, tutorials, and reviews)
http://intothenightphoto.blogspot.com/

Grevture Veteran Member • Posts: 4,188
Re: Is there time for an interchangeable sensor DSLR?
1

Royce - Into the Night Photography wrote:

But then Nikon and Canon couldn't sell us the next generation camera, and make more money! As a professional photographer, I kept the same Nikon film camera bodies for about 10 years before upgrading. With digital I've had to upgrade about every 2-3 years to a new body

Well, as have been pointed out elsewhere in this thread, the body part is the less expensive part, more or less a shell to the valuable parts. And very probably, considering the extra engineering to make a replaceable sensor package work, that package would very likely be more expensive to build (and thus for us to buy) then those complete new cameras ... Which sort of defeats the point of such a design.

However, I've able to keep the same 2 lenses I bought for my first FX camera. It's still working great after almost 10 years, and I expect another ten!

This is where the modular design does work: Lenses can be designed fairly independent on sensors, and thus work fine to have as separate and replaceable parts. Same with flashes, tripods and other similar things. But separating the camera body from a sensor module would very likely cause more problems then it solves.

Think of the camera body as a fairly low cost shell holding the important image processing parts together and protecting them. When you want to replace the image processing part, you might  as well get a new protective shell for it while you are at it.

-- hide signature --

I don't suffer from insanity, I enjoy every moment of it!
By the way, film is not dead.
It just smell funny

 Grevture's gear list:Grevture's gear list
Nikon D70s Nikon D3 Nikon D3S Nikon AF-S Nikkor 17-35mm f/2.8D ED-IF Nikon AF Nikkor 50mm f/1.4D +7 more
OP Theodoros Fotometria Senior Member • Posts: 2,090
Re: Digital backs, may not be reading this right...

Joseph S Wisniewski wrote:

marcio_napoli wrote:

I understood what you mean, Grevture, but if you say so, doesn't it sound like good business to Nikon after all?

Well, so they could (for example) sell a FF digital back for 1.5k.

Imagine you could buy the D800 for 3k, and a module sporting the same 36mp sensor for 1.5k, ready to use on your F5.

There are two things wrong with that line of reasoning, unfortunately.

First is your price. There's only about $900  worth of "camera" in a D800, the other  $2,100 is the "digital back". Now, digital backs for an existing SLR won't sell in as high quantities as new DSLRs. (I'll get to why in a second) so their prices will be even higher than the $2,100 you'd expect from the difference in the prices of D800 and F100.

Second, you have to differentiate between a digital back for a brand new SLR designed specifically to accept one, and a digital back designed to work with an existing F5.

  • Existing film SLRs can't take FF digital backs. The sensor frame protrudes past the image plane and has to go through the camera's film gate. It's about 8mm bigger than the sensor, so 1.3x crop is the limit for digital backs on existing SLRs, like the Kodak DCS-760 or Leica DMR.
  • Existing film SLRs can't do TTL flash with digital sensors, because the sensors don't reflect like film does. The Nikon F5 that Kodak used for their DCS series was not stock, it was torn down, the shutter curtain repainted to reflect better, and the flash system modified to use a "preflash" that metered off that curtain. This worked horribly.
  • The F5 focusing system is flat-out not up to the demands of a D800 sensor. I tried using an F5 for high resolution film work with Tech Pan film: it was a dismal failure.

Nikon would sell tons of both, I believe.

I did a design study and coordinated the market research for a company designing a pretty cool digital back, and the market research shows that when the credit cards actually hit the table, very few customers won't put up with all the suckage of a back/body combination in order to save a few $ or keep a sentimental "favorite"camera alive a few more years.

And following your logic, they could have tremendous profit doing so.

Believe it or not, DSLRs are basically loss-leaders that get you buying the profitable items, lenses, batteries, vertical grips.

So back to my original question, why is it not commercially doable?

Which? A new SLR built for digital backs, a digital back for F6 (and another for F5, another for F100, etc. etc. etc.), or the OP's concept?

And under the risk of hijacking this thread, I'm gonna explore Silicon Film's concept even furtherGogol

Go for it. I'm a bit of a historian on that scam. And Kirsten Schmidt is an old friend.

The dpReview editor puked again. Nothing good past this point. I'll follow up in another post.

"the unit itself has a built-in capacity for 24 images (64MB) after which time it must be inserted into the E-Box and its contents either transferred to a computer or CompactFlash card.

The second limitation is that the relatively small 1.3 megapixel CMOS sensor uses only about 30% of the center of the frame,  looking through the viewfinder you have a small field of view (marked out by a supplied rub-on transfer) which equates to a 2.58x focal length multi

Lastly it only currently supports certain camera models: Nikon F5, F3, N60/N90 and Canon EOS-1N, EOS-A25."

It's ancient 2002 tech.

I mean, max storage of 24 shots per run? 1.3mp? 2.6x crop?

It's clear why it never went into production. 2002 tech did not allow a proper implementation.

What if you had these 2012 specs:

- internal storage for 2.000 shots.

- 24mp

- FF sensor

- could be used on a variety of popular film cameras (probably we all have some of those in our closets): F100, F5, F6... Leica M6 anyone ? etc

My point is: with the original specs, it was clearly headed downhill right from the start, even in 2002.

The old D100 runs circles around it.

But... what about these very doable 2012 specs?  Seems very easy to sell.

I suppose that if there was market for a DB to fit any film 35mm camera, and if such back could cost a sensible price, then it wouldn't have to be Nikon to it for the F5, it could be anybody... even Phase one or Sinar or... Kodak... and he wouldn't make it for F5 only, but he would make variations that could apply for any film camera that exists! Now that's good business for Marcio, he can pick a few friends (don't count on me) ...go to the bank with the idea and next day we will be talking with a multimillionaire...

However, since converting an F5 to digital costs more than ...a MFDB! (the reason is that there is no provision for correct fitting, cooling and communicating with the unit as there is with a modern MF camera designed with an MFDB in mind) I suggest to come back to the topic that is "interchangeable sensor"...  and that can only cost some.... hundreds!

Now imagine this....:

1. You buy your D800 for 2500, it has an interchangeable sensor capability but bares no sensor in it!

2. You have the following choices:

A. 36mpx sensor module for 1000, B. 24mpx sensor module for 800, C.16mpx sensor module for 700, D. A new sensor module from RED for superb 2K video at 1000, E.Another one from Arri for 4K video at 2000, F. Another sensor module from Ilford for 28mpx B&W at 900, G. An IR sensor module from KODAK for 1600, H. A special Astrophotography 26mpx sensor module from Orion for 1200....

.....would you like that?

-- hide signature --

Theodoros
www.fotometria.gr
www.fotometriawedding.gr

OP Theodoros Fotometria Senior Member • Posts: 2,090
Re: Is there time for an interchangeable sensor DSLR?

Grevture wrote:

Royce - Into the Night Photography wrote:

But then Nikon and Canon couldn't sell us the next generation camera, and make more money! As a professional photographer, I kept the same Nikon film camera bodies for about 10 years before upgrading. With digital I've had to upgrade about every 2-3 years to a new body

Well, as have been pointed out elsewhere in this thread, the body part is the less expensive part, more or less a shell to the valuable parts. And very probably, considering the extra engineering to make a replaceable sensor package work, that package would very likely be more expensive to build (and thus for us to buy) then those complete new cameras ... Which sort of defeats the point of such a design.

However, I've able to keep the same 2 lenses I bought for my first FX camera. It's still working great after almost 10 years, and I expect another ten!

This is where the modular design does work: Lenses can be designed fairly independent on sensors, and thus work fine to have as separate and replaceable parts. Same with flashes, tripods and other similar things. But separating the camera body from a sensor module would very likely cause more problems then it solves.

Think of the camera body as a fairly low cost shell holding the important image processing parts together and protecting them. When you want to replace the image processing part, you might  as well get a new protective shell for it while you are at it.

Ok... if we demand such a camera that takes a replaceable module which contains the sensor and a manufacturer will make it... don't buy it!

...I am sure you can get the plenty current ones that will be all over the place in the S/H market from the rest of us!

-- hide signature --

Theodoros
www.fotometria.gr
www.fotometriawedding.gr

Grevture Veteran Member • Posts: 4,188
Now we are talking!
2

Theodoros Fotometria wrote:

I suppose that if there was market for a DB to fit any film 35mm camera, and if such back could cost a sensible price, then it wouldn't have to be Nikon to it for the F5, it could be anybody... even Phase one or Sinar or... Kodak... and he wouldn't make it for F5 only, but he would make variations that could apply for any film camera that exists! Now that's good business for Marcio, he can pick a few friends (don't count on me) ...go to the bank with the idea and next day we will be talking with a multimillionaire...

However, since converting an F5 to digital costs more than ...a MFDB! (the reason is that there is no provision for correct fitting, cooling and communicating with the unit as there is with a modern MF camera designed with an MFDB in mind) I suggest to come back to the topic that is "interchangeable sensor"...  and that can only cost some.... hundreds!

Except for the very, very last part about "cost some ... hundreds" ... Now you write things which make sense!

I agree it does not make much sense trying to re-use old 35 mm film cameras for something they were not designed for to begin with (unlike MF cameras who where actually designed for replaceable film backs).


Now imagine this....:

1. You buy your D800 for 2500, it has an interchangeable sensor capability but bares no sensor in it!

2. You have the following choices:

A. 36mpx sensor module for 1000, B. 24mpx sensor module for 800, C.16mpx sensor module for 700, D. A new sensor module from RED for superb 2K video at 1000, E.Another one from Arri for 4K video at 2000, F. Another sensor module from Ilford for 28mpx B&W at 900, G. An IR sensor module from KODAK for 1600, H. A special Astrophotography 26mpx sensor module from Orion for 1200....

.....would you like that?

As I have said before, the idea is nice and intriguing. So like it, yes. But there are some serious caveats which make it very hard to implement:

a) As has been pointed out many times, it would not work just replacing the bare sensor. Several of the more interesting sensor options you bring up would require replacing major parts of the whole interior since the sensor designs are so vastly different. In short you would need to replace a rather substantial "image processing module" with sensor + processor + possibly some other stuff (like external A/D converters and so on).

b) Building a D800 sized camera house where you can replace some internal parts (the image processing module) would require the camera body to be opened up. I mean its not like you will be able to sneak that stuff in and out through the memory card compartment. And building a camera which can be opened up, and still remain reasonably weather sealed and reliable and where you do not end up with  a slightly misaligned sensor ... Would be a whole lot more expensive then 2500 USD. A lot more. Again look at the DMF cameras which have been very prone to failures and disconnects where you have to take them apart and put them together several times to make them work. And this because they are modular in design.

I am not saying your idea is bad or attacking you here (just a clarification :)), I am just saying it might not be as easy to make it into a commercially viable product (one that can be sold at a competitive price) as you seem to think.

My suggestion would be slightly different: Nikon has already given a hint by offering two D800 versions (the standard and the E) with slightly different sensor options. Why not take that concept further and offer more options? A bit like they did with the D200/Fuji S5 Pro. Or a bit like Canon does with the 1D X and the 1D C?

-- hide signature --

I don't suffer from insanity, I enjoy every moment of it!
By the way, film is not dead.
It just smell funny

 Grevture's gear list:Grevture's gear list
Nikon D70s Nikon D3 Nikon D3S Nikon AF-S Nikkor 17-35mm f/2.8D ED-IF Nikon AF Nikkor 50mm f/1.4D +7 more
Leif Goodwin Senior Member • Posts: 1,390
Re: Joseph, don't bother...
3

Wojciech Sawicki wrote:

This guy'll soon go Ad Hominem at you if he loses a logical debate.

Check out his exchange with me a few posts earlier. Apparently it's all a conspiracy and we're idiots.

Yes, instead of addressing the criticisms I raised, he started abusing me and others. It is easier to hide behind a smokescreen than to debate in an adult manner, which includes admitting if someone else is right, or has made a good point.

 Leif Goodwin's gear list:Leif Goodwin's gear list
Nikon D200 Nikon D600 Nikon AF Micro-Nikkor 60mm f/2.8D Nikon AF-S Micro-Nikkor 60mm F2.8G ED Nikon AF Micro-Nikkor 200mm f/4D ED-IF +4 more
Leif Goodwin Senior Member • Posts: 1,390
Re: Now we are talking!

Grevture wrote:

Theodoros Fotometria wrote:

I suppose that if there was market for a DB to fit any film 35mm camera, and if such back could cost a sensible price, then it wouldn't have to be Nikon to it for the F5, it could be anybody... even Phase one or Sinar or... Kodak... and he wouldn't make it for F5 only, but he would make variations that could apply for any film camera that exists! Now that's good business for Marcio, he can pick a few friends (don't count on me) ...go to the bank with the idea and next day we will be talking with a multimillionaire...

However, since converting an F5 to digital costs more than ...a MFDB! (the reason is that there is no provision for correct fitting, cooling and communicating with the unit as there is with a modern MF camera designed with an MFDB in mind) I suggest to come back to the topic that is "interchangeable sensor"...  and that can only cost some.... hundreds!

Except for the very, very last part about "cost some ... hundreds" ... Now you write things which make sense!

I agree it does not make much sense trying to re-use old 35 mm film cameras for something they were not designed for to begin with (unlike MF cameras who where actually designed for replaceable film backs).


Now imagine this....:

1. You buy your D800 for 2500, it has an interchangeable sensor capability but bares no sensor in it!

2. You have the following choices:

A. 36mpx sensor module for 1000, B. 24mpx sensor module for 800, C.16mpx sensor module for 700, D. A new sensor module from RED for superb 2K video at 1000, E.Another one from Arri for 4K video at 2000, F. Another sensor module from Ilford for 28mpx B&W at 900, G. An IR sensor module from KODAK for 1600, H. A special Astrophotography 26mpx sensor module from Orion for 1200....

.....would you like that?

As I have said before, the idea is nice and intriguing. So like it, yes. But there are some serious caveats which make it very hard to implement:

a) As has been pointed out many times, it would not work just replacing the bare sensor. Several of the more interesting sensor options you bring up would require replacing major parts of the whole interior since the sensor designs are so vastly different. In short you would need to replace a rather substantial "image processing module" with sensor + processor + possibly some other stuff (like external A/D converters and so on).

b) Building a D800 sized camera house where you can replace some internal parts (the image processing module) would require the camera body to be opened up. I mean its not like you will be able to sneak that stuff in and out through the memory card compartment. And building a camera which can be opened up, and still remain reasonably weather sealed and reliable and where you do not end up with  a slightly misaligned sensor ... Would be a whole lot more expensive then 2500 USD. A lot more. Again look at the DMF cameras which have been very prone to failures and disconnects where you have to take them apart and put them together several times to make them work. And this because they are modular in design.

Hey, I've had a revolutionary idea. The sensor and image processing hardware are coupled, so really need to be supplied in one unit. And the shutter and mirror mechanism are coupled to them too, in that sometimes you need a D4 grade unit and sometimes a D3100 grade unit, depending on the processing capability of the electronics. And opening up a camera exposes it to damage, consumer complaints and so on. And then there is the increased cost of seals, etc, as mentioned. So, given that the camera bits are much cheaper than the sensor, why not package the eletronics, sensor, shutter, mirror and viewfinder in one metal box, and have the lens connected by a screw thread, or some form of quick release mechanism? I am going out on a limb here, but it'd work.

I am not saying your idea is bad or attacking you here (just a clarification :)), I am just saying it might not be as easy to make it into a commercially viable product (one that can be sold at a competitive price) as you seem to think.

My suggestion would be slightly different: Nikon has already given a hint by offering two D800 versions (the standard and the E) with slightly different sensor options. Why not take that concept further and offer more options? A bit like they did with the D200/Fuji S5 Pro. Or a bit like Canon does with the 1D X and the 1D C?

Yes, these interchangeable sensor cameras would sell to a small number of users, in my opinion of course. But from what I have seen, IR enabled cameras do not sell many, and Canon seem to have only produced one such camera. The Fuji S5 Pro did not really sell well, despite having some advantages.

-- hide signature --

I don't suffer from insanity, I enjoy every moment of it!
By the way, film is not dead.
It just smell funny

Edited to correct typos.

 Leif Goodwin's gear list:Leif Goodwin's gear list
Nikon D200 Nikon D600 Nikon AF Micro-Nikkor 60mm f/2.8D Nikon AF-S Micro-Nikkor 60mm F2.8G ED Nikon AF Micro-Nikkor 200mm f/4D ED-IF +4 more
Leif Goodwin Senior Member • Posts: 1,390
Re: It's here
2

Theodoros Fotometria wrote:

Leif Goodwin wrote:

Theodoros Fotometria wrote:

Leif Goodwin wrote:

Theodoros Fotometria wrote:

I do think that TH is a mindless troll... but that's my opinion...!

Wow. There is not much more to say, I'm afraid. Have a nice day.

Oh! It was a "to the guru I believe" post from the start... why don't you say so... (from the start)...?

Well... let the "Guru" post all the "predictions" and "reasoning" himself ...no? ....What are you guys? ...mindless servants to the "Guru"? ...or is the Guru so mindless to let the "servants" do "the job for him"? ...LOL (for the ...guru!)

P.S. ....(and I wondered all this time "where is TOF Guy"... ?) ....LOL!

You dismiss comments by someone who is widely respected for his reviews, and Nikon camera guides, and who has inside contacts in the industry, including in Nikon, as a 'mindless troll'. I think you need to look in the mirror. You are a complete unknown, making claim after claim, without any justification or proof whatsoever. And when someone questions you, you insult them, on the grounds that they do not accept something simply because you say so. And as for my not having any brains, I have a PhD, unlike you.

In what? ...is it in reasoning? ...because if it is in reasoning and you present it... I' ll "shut up"...., otherwise..... some "others" have to "shut up"!

So, you tell me I do not have any brains, you say that a well known equipment reviewer and author is a mindless troll, and now you tell me to shut up. Are you really quite so offensive as you appear to be? It looks to me that you are avoiding answering the questions/criticisms I raised by hiding behind abuse. What a shame. I've explained why I think this idea is not commercially viable. And so have many other people.

 Leif Goodwin's gear list:Leif Goodwin's gear list
Nikon D200 Nikon D600 Nikon AF Micro-Nikkor 60mm f/2.8D Nikon AF-S Micro-Nikkor 60mm F2.8G ED Nikon AF Micro-Nikkor 200mm f/4D ED-IF +4 more
Tony Beach Forum Pro • Posts: 11,926
Re: More B&W and IR...
1

lovEU wrote:

Joseph S Wisniewski wrote:

The CFA reduces DR

Actually, it increases it by 1-2 stops.

Interesting.  It still strikes me as counterintuitive since I gain DR by suppressing the blue and green channels with a magenta filter.

Struck me that way, too. The magenta filter trick (Bernes and Taplin, I go back a long way with this sort of stuff) actually decreases dynamic range. It also decreases the workload on the software, because all the channels tend to blow at the same time, so the algorithms that reconstruct a blown channel from the remaining ones don't cut in as often.

I don't understand how "The magenta filter trick (...) actually decreases dynamic range." Not sure if Iliah would agree on this.

I know Iliah wrote that using a magenta filter added DR, and that's consistent with my experience as well.

Keyboard shortcuts:
FForum MMy threads