BTHPhoto wrote:
In the late '90s or early 2000's you could buy a 35mm "film cartridge" that had something like a 1mp sensor in it.  I remember thinking that what you're suggesting was exactly where digital was going then - we'd have a sensor customized to match portra, one customized to match TMX, one customized to match FP4+, etc. - and that it was surely only a matter of time before they produced the 120 version that would fit my Mamiya 645.

That was vaporware, and none were ever actually made.

David Parsons wrote:
That was vaporware, and none were ever actually made.

I've never actually seen one, so you could be right, I distinctly recall two different companies, Imagetek and Silicon Film, advertising them.

Edit: seems you're correct.
http://photo.net/bboard/q-and-a-fetch-msg?msg_id=0008BW
http://www.imaging-resource.com/NEWS/1032460933.html

And there was even an April Fools version in 2011.
http://petapixel.com/2011/04/04/35mm-ca … o-digital/

David Parsons wrote:

That was vaporware, and none were ever actually made.

but they had wooden mock-ups and were just a few hundred million short in R&D

BTHPhoto wrote:
...it doesn't matter how much sense it would make or whether it could be done, it would NEVER make as much money as selling brand new DSLRs that get "better" every market cycle.  As long as profit is the motive of corporations, we'll never get the chance to find out whether your idea makes sense or not.

Yep - this ^^^
The obsolescence cycle is just too attractive for camera companies. Now that they've got you on the treadmill, they'll never let consumers off.

SKITA Studios wrote:
+1 Ricoh GXR.
Pretty much exactly as you described.

.

Not really.  If I recall that camera correctly, it was a lens/matched sensor module.  Only one chip to buy and you bought it with the attached lens, right?

That is different than my thought.  With my contraption, a consumer would have just a chip module and might own several of them for specific purposes.

As with lenses, the "tuned" chips would tie you closely with a particular brand making switching potentially less likely.  As new bodies came out with new features - über plus autofocus, say- you might buy a new body on a replacement cycle.  Chips would be continuously refined and new "tonalities" introduced routinely.  Continous sales.

In today's model, I am not sure that selling ONE body with a fixed chip to a consumer every four years (say) is inherently more profitable than selling one body every 6-8 years but selling a bunch of high margin chips too. 

Apart from the very real possibility that this system will never be built because of technical or financial reasons, if it ever was, I think I would be interested.  I would love to be able to swap out a color chip for a tuned b&W chip with specific ISO performance ranges rather than carry around a completely separate body (aka Leica Monochrome) .......

Digitoxin wrote:
Can anyone envision a day when we are able to swap the sensor in a dSLR, Mirrorless, or rangefinder camera like we swapped film?  Will chips ever get cheap enough that a camera maker will create a line of chips tuned for certain tasks?

B&W tuned for low ISO settings
B&W tuned for high ISO settings
Infrared
Various pixel counts
Various tonalities (replicating certain films for instance).

Each chip could simply be swapped into the camera as needed. 

This day is likely many years off .  I can see this day but I am wondering if anyone would buy into a system like this (inexpensives chips, swappable, no need to buy a new body for each type of tuned chip)?

Would ya?

Why not?  Samsung is allowing for upgrading our TVs   maybe some day...

Jon Gilchrist wrote:
I'd like to see the camera become more modular.  Reduce the body to a lens mount and a shutter/sensor assembly, then have a separate plug-in module for storage (could be SD, CF, Wireless, HD, USB, etc), another for viewfinder, another for system control.  You could update the sections you wanted, and retain the parts that still worked (has anyone ever worn out their viewfinder?).

This would certainly make weathersealing difficult, but some of us don't need weathersealing.

Many MF systems operated this way.

I think RED might be doing something along the lines you want ...
It's a bit expensive tho ...
http://www.red.com/products

(I was slow, see post above.)

Gary Melton wrote:
Remember you heard it here FIRST...this is what the next BIG thing in digital cameras will be:

                             Liquid Sensor Pixels!

Here's the way it will work - the camera will have a thin empty "tank" in it in place of the sensor...that you will fill it with the appropriate "liquid sensor pixels" (LSPs) - and there will be a variety of them available.  They'll have different sensitivities, different levels of sharpness, different bokeh algorithms, different color patterns, etc.  When you need different characteristics, you simply drain the "tank" and pour in different "LSPs".

Once these "LSPs" cameras are even somewhat perfected, all camera improvement development will cease pretty much completely...and a whole new world of "LSPs" development will become the "next big thing" - with thousands of variations for EVERY occasion being available within 2 years of the first cameras hitting the market.

The GIANT of this whole new world will be a newly created division of Holga...

No.  Thixotropy.

Michael Zahra wrote:
No.  Thixotropy.

Well...thixotropic yogurt to put a finer edge to it...

Gary Melton wrote:
Remember you heard it here FIRST...this is what the next BIG thing in digital cameras will be:

                             Liquid Sensor Pixels!

Here's the way it will work - the camera will have a thin empty "tank" in it in place of the sensor...that you will fill it with the appropriate "liquid sensor pixels" (LSPs) - and there will be a variety of them available.  They'll have different sensitivities, different levels of sharpness, different bokeh algorithms, different color patterns, etc.  When you need different characteristics, you simply drain the "tank" and pour in different "LSPs".

Once these "LSPs" cameras are even somewhat perfected, all camera improvement development will cease pretty much completely...and a whole new world of "LSPs" development will become the "next big thing" - with thousands of variations for EVERY occasion being available within 2 years of the first cameras hitting the market.

The GIANT of this whole new world will be a newly created division of Holga...

And you won't be able to purchase LSPs, they will only be available for use via a never-ending monthly subscription. If you stop paying, you're existing LSPs will cease functioning.

Digitoxin wrote:
Not really.  If I recall that camera correctly, it was a lens/matched sensor module.  Only one chip to buy and you bought it with the attached lens, right?

That is different than my thought.  With my contraption, a consumer would have just a chip module and might own several of them for specific purposes.

Yes.  The GXR was a P&S w/ a similar concept so the lens was included.
The closest thing to what you're thinking of really is a digital back.  It has to be designed as a system as someone mentioned (big ass sensor plus slow processor = crappy performance, so you're going to at minimum have a sensor plus processor).

But the main issue is that will scuttle your idea is the mechanical tolerances involved.  It was hard to get digital backs perfectly aligned.  The D800 sensor will lose focus if you slightly dent your mounting flange by bumping a lens into it.

pdxROCKpix wrote:
No. I work In the semiconductor industry and I can't see a way to package the chip where it would be user changeable and yet protected enough to not ruin it and still perform properly. It's not like swapping a CPU in your computer, which many people can't do without ruining something.

How does the medium format backs do it?

Digitoxin wrote:

How does the medium format backs do it?

it's a completely different design, cost point and level of maintenance.

are you willing to have your DSLR double in size just to be able to swap the sensors? Pay 5x as much just for the sensor? Pay 5x as much to have it maintained and repaired?

DSLR is what it is because it provides a high quality image from a relatively compact size that is easy to use and maintain. being able to swap sensors would take away many if not all of those advantages.

you are talking about two completely different systems.

Michael DBA Expressions wrote:
OK, I just gotta ask: in what way is swapping sensors remotely analogous to swapping film???

The analogous swap would be to trade memory cards. And given the ease of sharing images and the technical difficulties in trading cards -- to avoid problems, one needs to reformat the card before shooting anything -- I don't see a reason to attempt such a thing.

Film had certain characteristics, sensitivities, and tonalities.  You swapped film to acheive specific results. Chips could too.

Can you envision a chip that rendered images like velvia?  Or how about something that we haven't considered?

Can you envision a chip that was tuned amazingly well to get rich blacks but only in an ISO range of 50 - 400 and another tuned to provide amazing results at 3200 -12,800?

The chip sees the light.  The memory card just stores it as 0's and 1's.

Does that answer your question?

pdxROCKpix wrote:
it's a completely different design, cost point and level of maintenance.

are you willing to have your DSLR double in size just to be able to swap the sensors? Pay 5x as much just for the sensor? Pay 5x as much to have it maintained and repaired?

DSLR is what it is because it provides a high quality image from a relatively compact size that is easy to use and maintain. being able to swap sensors would take away many if not all of those advantages.

you are talking about two completely different systems.

Yes, it is a different design.  The cost point is based on the size of the sensor and current manufacturing costs.  In the future, dSLR chips might fall markedly in price.... Say $15 each wholesale.

There are many people, including myself, who cannot see a way to do something.  Yet, often times I am wrong and people invent a way to do things that I would not have thought possible. The "digital film cartridge" that never came to fruition might be a starting point that was simply before its time.  Or, perhaps not.

I guess the point here is that, base on our experience in the tech industry, you don't see that we will "ever" (I.e. never) be able to swap a sensor like we could swap film in something about the size of a current dSLR (or smaller). I am not sure that I think we will either.  But, if it is ever created, I think I would be an interested customer.

Digitoxin wrote:

Yes, it is a different design.  The cost point is based on the size of the sensor and current manufacturing costs.  In the future, dSLR chips might fall markedly in price.... Say $15 each wholesale.

There are many people, including myself, who cannot see a way to do something.  Yet, often times I am wrong and people invent a way to do things that I would not have thought possible. The "digital film cartridge" that never came to fruition might be a starting point that was simply before its time.  Or, perhaps not.

I guess the point here is that, base on our experience in the tech industry, you don't see that we will "ever" (I.e. never) be able to swap a sensor like we could swap film in something about the size of a current dSLR (or smaller). I am not sure that I think we will either.  But, if it is ever created, I think I would be an interested customer.

Bigger chips are actually easier to make. This is to me of the ironic parts of the mid size sensor costs. The pixel density of a mid sensor is actually less than say a D800 sensor. It is MUCH harder to manufacture the higher density, smaller sensor.