Is the megapixel race over?

Seems that the megapixel race is virtually over. I suspect the 5MP 1/1.8 has the smallest photosites that are practical, given that they are only about 2X the size of the wavelength of red light and you can't get much smaller. The same applies to an 8MP 2/3 chip. As it stands these sensors all have an ISO noise issue and there does not seem to be any way around this.

And there does not seem to be a lot of interest from DSLR manufacturers to create an APS sensor with more than 6MP. Is this just a cost issue, or is it just an admission that 6MP APS, or 11MP full frame, is about as far as you can go with existing optics and processing speeds - or is it just "good enough" for the majority of professional SLR uses? In other words, there is a lack of serious demand for more MP in this market.

Buying a film SLR was always a long term purchase, and it seems the same may now be true for their digital equivalents. The prices may well come down some, though I suspect chip yields for larger formats will remain an issue for some time and this will hold the baseline cost of DSLR cameras at somewhere around the $800 mark.
Steve

Steve Jacob said:

Seems that the megapixel race is virtually over. I suspect the 5MP
1/1.8 has the smallest photosites that are practical, given that
they are only about 2X the size of the wavelength of red light and
you can't get much smaller. The same applies to an 8MP 2/3 chip. As
it stands these sensors all have an ISO noise issue and there does
not seem to be any way around this.

Click to expand...

Not so far but who knows what is going on in some R&D lab. It would take a while to come to market in any event.

*isteve said:

And there does not seem to be a lot of interest from DSLR
manufacturers to create an APS sensor with more than 6MP. Is this
just a cost issue, or is it just an admission that 6MP APS, or 11MP
full frame, is about as far as you can go with existing optics and
processing speeds - or is it just "good enough" for the majority of
professional SLR uses? In other words, there is a lack of serious
demand for more MP in this market.

Click to expand...

Probably due to all your reasons. 6 MP is good for what I do (landscapes, panoramas, etc). Going larger for me would have disadvantages. More pixels up the processing load. Going larger without losses in noise performance would push me towards full frame. I would not like loosing the DOF. I'm already having some difficulty with DOF with a 1.6 miltiplier. Studio folks and others do use more pixels to good effect. Many seem to be happy with 11MP or so but some use the larger sensors (at a hefty price).

*isteve said:

Buying a film SLR was always a long term purchase, and it seems the
same may now be true for their digital equivalents.

Click to expand...

Not sure about this. I would say that there are still significant improvements to be had in other areas in DSLRs and consumer cameras. More speed, better AF, better histograms, more reliable flash, etc. are all improvement areas.

*isteve said:

The prices may
well come down some, though I suspect chip yields for larger
formats will remain an issue for some time and this will hold the
baseline cost of DSLR cameras at somewhere around the $800 mark.

Click to expand...

For a while. In a longer term, it depends on how many buy into the DSLR camera business. A large market will drive more R&D to improve processes and yield and will also permit more economies of scale. Over time, this happened in the film camera business.

*isteve said:

Steve

Click to expand...

--
Leon
http://pws.prserv.net/lees_pics/landscapes.htm

Steve,

Well it all depends....

on Canon or Nikon or maybe Olympus who knows? etc..
Whether they release as a 'suprise for 2004'
lets say the Nikon D200 at 8Mp
or the Canon 20D at 8Mp,
to grab and maintain sales dominance in the DSLR market.

As we all know if a D200 or 20D at even just 8MP ( mere 2MP increase) is released..

the buying public will not give the 10D and D100 another look especially if like Canon (D60) the last model is dis-continued..
as the public perseve (as do i) that the more MP the higher the resolutuion
etc, etc...the better the camera.

Especially if the price point is the same or lower, human nature etc...
So it all depends IMO who is 1st to break the current 6MP barrier,
if nobody then well its as you say..

But IMO i think the MP race is far from over.....we will just have to wait and see...

interesting post Steve..

regards Michael..

Steve Jacob said:

Seems that the megapixel race is virtually over. I suspect the 5MP
1/1.8 has the smallest photosites that are practical, given that
they are only about 2X the size of the wavelength of red light and
you can't get much smaller. The same applies to an 8MP 2/3 chip. As
it stands these sensors all have an ISO noise issue and there does
not seem to be any way around this.
And there does not seem to be a lot of interest from DSLR
manufacturers to create an APS sensor with more than 6MP. Is this
just a cost issue, or is it just an admission that 6MP APS, or 11MP
full frame, is about as far as you can go with existing optics and
processing speeds - or is it just "good enough" for the majority of
professional SLR uses? In other words, there is a lack of serious
demand for more MP in this market.
Buying a film SLR was always a long term purchase, and it seems the
same may now be true for their digital equivalents. The prices may
well come down some, though I suspect chip yields for larger
formats will remain an issue for some time and this will hold the
baseline cost of DSLR cameras at somewhere around the $800 mark.
Steve

Click to expand...

Steve Jacob said:

Seems that the megapixel race is virtually over. I suspect the 5MP
1/1.8 has the smallest photosites that are practical, given that
they are only about 2X the size of the wavelength of red light and
you can't get much smaller. The same applies to an 8MP 2/3 chip. As
it stands these sensors all have an ISO noise issue and there does
not seem to be any way around this.

Click to expand...

I agree with you so far; there doesn't seem much point to adding more photosites to this size sensors.

*isteve said:

And there does not seem to be a lot of interest from DSLR
manufacturers to create an APS sensor with more than 6MP. Is this
just a cost issue, or is it just an admission that 6MP APS, or 11MP
full frame, is about as far as you can go with existing optics and
processing speeds - or is it just "good enough" for the majority of
professional SLR uses? In other words, there is a lack of serious
demand for more MP in this market.

Click to expand...

Here, I disagree. I just think the technology isn't "there" yet. There are 16-20 MP solutions for medium-format digital backs, and a lot of interest for them. Today's best optics should be able to make use of 40 or even 60 MP.

Don't be too impatient. Big chips like the APS-sized (and bigger) sensors are difficult to design and difficult to make. The market is pretty small, too -- only Canon is selling them by the million. I'm pretty sure we'll see 8-10 MP APS-sized sensors and 16-20 MP full-frame ones by the next generation or the one after that: I wouldn't be at all surprised if this happens this year, and very surprised if it doesn't happen before the end of 2005.

However, I'm hoping that the next generation of chips will add more than just megapixels -- what we really need is more latitude, especially at the highlights.

*isteve said:

Buying a film SLR was always a long term purchase, and it seems the
same may now be true for their digital equivalents.

Click to expand...

Eventually, but not just yet. Give it 2-5 more years, and it'll happen.

*isteve said:

The prices may
well come down some, though I suspect chip yields for larger
formats will remain an issue for some time and this will hold the
baseline cost of DSLR cameras at somewhere around the $800 mark.
Steve

Click to expand...

Yep, I think this is quite true.

Petteri
--



Portfolio: [ http://www.seittipaja.fi/index/ ]
Pontification: [ http://www.seittipaja.fi/ ]

michael thompson said:

Steve,

Well it all depends....

on Canon or Nikon or maybe Olympus who knows? etc..
Whether they release as a 'suprise for 2004'
lets say the Nikon D200 at 8Mp
or the Canon 20D at 8Mp,
to grab and maintain sales dominance in the DSLR market.

Click to expand...

I thought about this. Chip yields are more to do with the chip area (how many chips to a wafer) rather than the site dimension. 8MP is clearly possible on an APS sensor, but at what cost (sensitivity/noise etc). Its clearly NOT a no brainer to get low noise on a large sensor as the Kodak 14N, D2H and Contax digitals show. If you want a high speed camera, its probably even harder.

Michael Thompson said:

As we all know if a D200 or 20D at even just 8MP ( mere 2MP
increase) is released..
the buying public will not give the 10D and D100 another look
especially if like Canon (D60) the last model is dis-continued..

Click to expand...

Unless there is a major price differential, then you may be right.

Michael Thompson said:

as the public perseve (as do i) that the more MP the higher the
resolutuion
etc, etc...the better the camera.

Click to expand...

Never been that convinced myself. Unless you can afford professional glass or use good primes, the lenses are already something of a limitation on a 6MP DSLR. For most of us, a complete set of L glass is a distant dream!!

Michael Thompson said:

Especially if the price point is the same or lower, human nature
etc...
So it all depends IMO who is 1st to break the current 6MP barrier,
if nobody then well its as you say..

But IMO i think the MP race is far from over.....we will just have
to wait and see...

Click to expand...

If not over, I think its certainly slowed up a bit..

Michael Thompson said:

interesting post Steve..

regards Michael..

Steve Jacob said:

Seems that the megapixel race is virtually over. I suspect the 5MP
1/1.8 has the smallest photosites that are practical, given that
they are only about 2X the size of the wavelength of red light and
you can't get much smaller. The same applies to an 8MP 2/3 chip. As
it stands these sensors all have an ISO noise issue and there does
not seem to be any way around this.
And there does not seem to be a lot of interest from DSLR
manufacturers to create an APS sensor with more than 6MP. Is this
just a cost issue, or is it just an admission that 6MP APS, or 11MP
full frame, is about as far as you can go with existing optics and
processing speeds - or is it just "good enough" for the majority of
professional SLR uses? In other words, there is a lack of serious
demand for more MP in this market.
Buying a film SLR was always a long term purchase, and it seems the
same may now be true for their digital equivalents. The prices may
well come down some, though I suspect chip yields for larger
formats will remain an issue for some time and this will hold the
baseline cost of DSLR cameras at somewhere around the $800 mark.
Steve

Click to expand...

Click to expand...

I don't think consumers want/need/can handle/ many more pixels. Even the majority of young people (teens and under) using digital cameras don't care a bit about megapixels.

Steve Jacob said:

Seems that the megapixel race is virtually over. I suspect the 5MP
1/1.8 has the smallest photosites that are practical, given that
they are only about 2X the size of the wavelength of red light and
you can't get much smaller. The same applies to an 8MP 2/3 chip. As
it stands these sensors all have an ISO noise issue and there does
not seem to be any way around this.
And there does not seem to be a lot of interest from DSLR
manufacturers to create an APS sensor with more than 6MP. Is this
just a cost issue, or is it just an admission that 6MP APS, or 11MP
full frame, is about as far as you can go with existing optics and
processing speeds - or is it just "good enough" for the majority of
professional SLR uses? In other words, there is a lack of serious
demand for more MP in this market.
Buying a film SLR was always a long term purchase, and it seems the
same may now be true for their digital equivalents. The prices may
well come down some, though I suspect chip yields for larger
formats will remain an issue for some time and this will hold the
baseline cost of DSLR cameras at somewhere around the $800 mark.
Steve

Click to expand...

--
'Don't believe anything you read on the net. Except this. Well,
including this, I suppose.' Douglas Adams

If you know something about photography and think like a photographer you can think different and clearly than a normal person who doesn't have any idea about photography(Just know to look from the LCD and hit the button).

But, what I see at the people is like "My car is better than yours mine has sooooo much horsepower" etc.... etc....

If you live in the city and go to work with a Ferrari or a Lada you reach to your work in the same time because of the traffic jamm..........

Now the new talk is "my camera has more MP than yours". May be it's strange for you but it's the new reality.

I know lot of people they have a 5-6 MP camera and they use it allways on Auto Mode once a month and believe me when they realy need it they are complaining because of their batteries are almost emty.
Because they leave them in their house somewere and they forgot it.
When they show you their few potos on their computer the photos are terrible.

They complain that they give so much euros to that S*it and it doesn't take good pictures. Post processing " WHAT'S THAT"

It's the same as everybody have the latest model Mobile Phone.

Megapixelmania will go on and will never ends.

If they have the money and jusst like the shape of the cam they will buy a 20-30 MP Camera and they will complain......

Comedy but reality....................

*isteve said:

Seems that the megapixel race is virtually over. I suspect the 5MP
1/1.8 has the smallest photosites that are practical, given that
they are only about 2X the size of the wavelength of red light and
you can't get much smaller. The same applies to an 8MP 2/3 chip. As
it stands these sensors all have an ISO noise issue and there does
not seem to be any way around this.
And there does not seem to be a lot of interest from DSLR
manufacturers to create an APS sensor with more than 6MP. Is this
just a cost issue, or is it just an admission that 6MP APS, or 11MP
full frame, is about as far as you can go with existing optics and
processing speeds - or is it just "good enough" for the majority of
professional SLR uses? In other words, there is a lack of serious
demand for more MP in this market.
Buying a film SLR was always a long term purchase, and it seems the
same may now be true for their digital equivalents. The prices may
well come down some, though I suspect chip yields for larger
formats will remain an issue for some time and this will hold the
baseline cost of DSLR cameras at somewhere around the $800 mark.

Click to expand...

If we talk about DSLR (only good cameras) I think it will end, when sensor resolution will reach lens maximum, that is about 20-25 megapixels. It is a good resolution to print A2 and even A1. Digital backs will reach MF film resolution, that is about 50 megapixels for 645.

what about snapshoot cams, I think the race will end at about 6 megapixels for 2/3" sensors and about the same resolution for 1/1.8" but surely, lust ones will have bad dynamic range, colors, everithong, as they do now.

As for me, I'll be glad to replace film SLR with 16-18 megapixel full sized sensor based camera, if it will have accaptable dynamic rangge, atleast 10 f-stops.

MF backs are still around the 25MP region and still not full frame. Digital sensors are just not that suitable for non-telecentric lens designs.

A question - would you rather have 2 more stops of DR or 20% more resolution? I'd go for the DR. I can't afford lenses that can pull the most out of the current sensors, let alone a 16MP FF.

Steve

Igor R. said:

If we talk about DSLR (only good cameras) I think it will end, when
sensor resolution will reach lens maximum, that is about 20-25
megapixels. It is a good resolution to print A2 and even A1.
Digital backs will reach MF film resolution, that is about 50
megapixels for 645.
what about snapshoot cams, I think the race will end at about 6
megapixels for 2/3" sensors and about the same resolution for
1/1.8" but surely, lust ones will have bad dynamic range, colors,
everithong, as they do now.
As for me, I'll be glad to replace film SLR with 16-18 megapixel
full sized sensor based camera, if it will have accaptable dynamic
rangge, atleast 10 f-stops.

Click to expand...

Megapixel madness might slow down but one race I am intererested to follow is sensityvity race (noiseless).

These will be a great race where IS lenses are going to be a thing of the past. Same for bright fast expensive lenses. Will see huge zooms 16X?

Most of the photography ignorant people take shots with lots of blurr, so they will appreciate this one.
Another interesting race would be Dynamic Range race.

All you have to do is Quantify D. Range and make sure put the numbers in big golden letters next to the Zoom label.

So i think the resolution (megapixel) race might slow a bit but there is huge room for improvement for sensors in other areas. Some new technologies may come that surpass the best film. Not necesarily CCD or CMOS.

It's not even close to over.

If a 2/3rds sensor can house 8MP, then a 4/3rds sensor could house 32MP. 32 very noisy MP, but it could house them, and FF could grow 40-60MP if processing power could deal with it.

So that brings up a few questions. What's the pixel pitch of a 3MP 2/3rds sensor? Is it around 5-6um ? Because that should make a 12MP 4/3rds, or at least an APS, more than possible, one with good ISO 800 plus performance too.

But then there are other things to consider. Foveon type, layered sensors; opponent process two color technologies that Fuji was playing with a while back, and multi sensor backs? Provided they get cheaper, why not two or three sensors in a camera (when power usage and battery tech permit it) ???

And finally, once someone works it out, true "digital" sensors, which given similar advances to "digital" amplification, would allow for dramatically better performance from small photosites.

Barramundi said:

Megapixel madness might slow down but one race I am intererested to
follow is sensityvity race (noiseless).

Click to expand...

There is no room for sensitivity race, most sensitive film is ISO3200 it can be pooshed to 6400, but people usually shoot ISO1600 film, everithing above is too grainy, same for digital. Canon 1D allready have ISO3200.

Barramundi said:

These will be a great race where IS lenses are going to be a thing
of the past. Same for bright fast expensive lenses. Will see huge
zooms 16X?

Click to expand...

No, there is no place fro highX zooms, and will never be. The more X of the zoom, the worse quality it has!

Barramundi said:

Most of the photography ignorant people take shots with lots of
blurr, so they will appreciate this one.

Click to expand...

Technology doesn't give a deam about photography ignorant people, they care about pros.

Barramundi said:

Another interesting race would be Dynamic Range race.

Click to expand...

Thats will be interesting for me too. As I can see, current digicams has dynamic range less than film, 1.8" cameras are just ugly. But even DSLRs with movie sized sensors (24*15mm) still ahve less dynamic range then film. I di test myself, ISO100 film has dynamic range of 11 stops. I didn't test DSLRs, but I guess it will be about 8 stops only. Let me remind you that 1 stop means 2 times brighter! 3 stops more means laot!

Barramundi said:

All you have to do is Quantify D. Range and make sure put the
numbers in big golden letters next to the Zoom label.

Click to expand...

Digcam manufacturers are never presenting there numbers because DR is a weak side of digital today.

Barramundi said:

So i think the resolution (megapixel) race might slow a bit but
there is huge room for improvement for sensors in other areas. Some
new technologies may come that surpass the best film. Not
necesarily CCD or CMOS.

Click to expand...

Maybe Foveon X3... It has better DR than Bayer!

A very simple way to advertise digital cameras:
16
(bits dynamic range)
8
(megapixels)

Whereas '16' is written in a large type, '8' in a bit smaller, and thingies in parentheses even smaller.
It could look like (in Didots):
12/13;7/8;10/11;7/8.
I hope you understand.

I imagine, no - I hope - the megapixel race for digital SLRs ends at 8.64 megapixels (3600x2400) for APS-sized sensors (at 6.67 µm pixel pitch) and at 13.5 megapixel (4500x3000) for full-frame ones (8 µm pixel pitch).

There's no point going over that. With those values any photographer will feel comfortable, even more with Foveon, and at 16 bit dynamic range, such cameras will yield enormous RAW files - 17 MB (APS) or a freakin' huge 27 MB for FF cameras. A Foveon will yield excessive 52 MB or a whopping 81 MB RAW file. Even with compression, values will not go down too much, if the sensor is fairly noiseless at that 16 bits and the image is fairly dynamic range sensitive. TIFF files? Without compression we're talking those Foveon sizes. Who will want to store up to 12 images on a 1 GB microdrive? Not to mention lengthy write times. No, going above such resolutions is not only pointless, but a waste of resources that could be put into improving current crop of sensors at their current resolution.

Matsu said:

It's not even close to over.

If a 2/3rds sensor can house 8MP, then a 4/3rds sensor could house
32MP. 32 very noisy MP, but it could house them, and FF could grow
40-60MP if processing power could deal with it.

So that brings up a few questions. What's the pixel pitch of a 3MP
2/3rds sensor? Is it around 5-6um ? Because that should make a
12MP 4/3rds, or at least an APS, more than possible, one with good
ISO 800 plus performance too.

But then there are other things to consider. Foveon type, layered
sensors; opponent process two color technologies that Fuji was
playing with a while back, and multi sensor backs? Provided they
get cheaper, why not two or three sensors in a camera (when power
usage and battery tech permit it) ???

And finally, once someone works it out, true "digital" sensors,
which given similar advances to "digital" amplification, would
allow for dramatically better performance from small photosites.

Click to expand...

toughluck said:

A very simple way to advertise digital cameras:
16
(bits dynamic range)
8
(megapixels)
Whereas '16' is written in a large type, '8' in a bit smaller, and
thingies in parentheses even smaller.
It could look like (in Didots):
12/13;7/8;10/11;7/8.
I hope you understand.

Click to expand...

Unfortunately bit depth has nothing to do with dynamic range. You can have a 2-bit image with equal dynamic range as a 48-bit one; the difference will be that the latter will have rather more steps between the darkest and the lightest shades that can be distinguished from black and white.

DR is measured in decibels. I suppose you could use stops for photographic purposes. "12 stops of DR" would get me drooling, for one...

Petteri
--



Portfolio: [ http://www.seittipaja.fi/index/ ]
Pontification: [ http://www.seittipaja.fi/ ]

You're right. My wrong about dynamic range... Thought about something else... Can't name it right now...
What about (for example) 16 f-stops with 16 bit accuracy? (accuracy?)

It would be possible, and it would mean every bit would mean a stop. You can do quite a lot with 16 bit colour depth...

And 16 stops... Even 8 are attractive, or those 12 you've mentioned, at 8/12 bits, it would mean quite a latitude...

Petteri Sulonen said:

toughluck said:

A very simple way to advertise digital cameras:
16
(bits dynamic range)
8
(megapixels)
Whereas '16' is written in a large type, '8' in a bit smaller, and
thingies in parentheses even smaller.
It could look like (in Didots):
12/13;7/8;10/11;7/8.
I hope you understand.

Click to expand...

Unfortunately bit depth has nothing to do with dynamic range. You
can have a 2-bit image with equal dynamic range as a 48-bit one;
the difference will be that the latter will have rather more steps
between the darkest and the lightest shades that can be
distinguished from black and white.

DR is measured in decibels. I suppose you could use stops for
photographic purposes. "12 stops of DR" would get me drooling, for
one...

Petteri
--



Portfolio: [ http://www.seittipaja.fi/index/ ]
Pontification: [ http://www.seittipaja.fi/ ]

Click to expand...

I wish it was, but manufacturer's seem to be intent on pushit it onwards.

I believe that the resolution available from the 5MP compact's and 6MP DSLR's is more than enough for the majority of the market, as most won't ever print larger than A4 anyway.

I believe that the manufacturer's should switch their attention to

improved dynamic range
decreased shutter lag
improved burst capability
better low light performance
reducing barrel and pincushion distortion
reducing CA with better lens optics

there are many other's as well, but that's my list for now, and would certainly give manufacturer's a lot to work on

Steve Jacob said:

Seems that the megapixel race is virtually over. I suspect the 5MP
1/1.8 has the smallest photosites that are practical, given that
they are only about 2X the size of the wavelength of red light and
you can't get much smaller. The same applies to an 8MP 2/3 chip. As
it stands these sensors all have an ISO noise issue and there does
not seem to be any way around this.
And there does not seem to be a lot of interest from DSLR
manufacturers to create an APS sensor with more than 6MP. Is this
just a cost issue, or is it just an admission that 6MP APS, or 11MP
full frame, is about as far as you can go with existing optics and
processing speeds - or is it just "good enough" for the majority of
professional SLR uses? In other words, there is a lack of serious
demand for more MP in this market.
Buying a film SLR was always a long term purchase, and it seems the
same may now be true for their digital equivalents. The prices may
well come down some, though I suspect chip yields for larger
formats will remain an issue for some time and this will hold the
baseline cost of DSLR cameras at somewhere around the $800 mark.
Steve

Click to expand...

--
See my gallery at http://www.pbase.com/stuartd

toughluck said:

You're right. My wrong about dynamic range... Thought about
something else... Can't name it right now...
What about (for example) 16 f-stops with 16 bit accuracy? (accuracy?)
It would be possible, and it would mean every bit would mean a
stop. You can do quite a lot with 16 bit colour depth...
And 16 stops... Even 8 are attractive, or those 12 you've
mentioned, at 8/12 bits, it would mean quite a latitude...

Click to expand...

Wow, imagine the possibilities:

You'd use about 5 stops of that for a normal scene. Imagine, + - 5 stops of post-exposure compensation! Shoot at ISO100, push it to 3200! Or shoot at 3200, pull it to 100! Or, even better, shoot at ISO100, underexpose 5 stops (!) then push it to... ISO102400!!!

That'd also mean ISO3200 about as clean as ISO200 on current cameras.

Basically, this would mean that exposure value becomes irrelevant. You could set your shutter speed and aperture based on your need to stop motion or control depth of field, and adjust exposure to taste in post-processing. It'd pretty much turn photography on its head.

OK, where do I put my name on the waiting list?

Petteri
--



Portfolio: [ http://www.seittipaja.fi/index/ ]
Pontification: [ http://www.seittipaja.fi/ ]

toughluck said:

I imagine, no - I hope - the megapixel race for digital SLRs ends
at 8.64 megapixels (3600x2400) for APS-sized sensors (at 6.67 µm
pixel pitch) and at 13.5 megapixel (4500x3000) for full-frame ones
(8 µm pixel pitch).

Click to expand...

I don't think it will end there. 8-12 MP is more than doable on current APS sensors right now! If we're talking the physics of it. Cost is probably not the limiting factor either, but rather processing power -- both in camera and on the average computer. But neither of those two things stands still at all! The computers we'll buy in 5 years time will look ridiculously powerful by todays standards. It seems that the real limiting factor for a camera ATM is the on-boards processing ability available via current embedded electronics. To get a reasonable fps out of a 12-20MP camera we're going to need faster kit, what yields 8fps at 4MP would barely produce 2fps at 12MP, electronics that produce 2FPS currently, would be basically unuseable for anything but a studio capture system.

toughluck said:

There's no point going over that. With those values any
photographer will feel comfortable, even more with Foveon, and at
16 bit dynamic range, such cameras will yield enormous RAW files -
17 MB (APS) or a freakin' huge 27 MB for FF cameras. A Foveon will
yield excessive 52 MB or a whopping 81 MB RAW file. Even with
compression, values will not go down too much, if the sensor is
fairly noiseless at that 16 bits and the image is fairly dynamic
range sensitive. TIFF files? Without compression we're talking
those Foveon sizes. Who will want to store up to 12 images on a 1
GB microdrive? Not to mention lengthy write times. No, going above
such resolutions is not only pointless, but a waste of resources
that could be put into improving current crop of sensors at their
current resolution.

Click to expand...

As I say, nothing stands still. Look at the exponential gains in storage capacity over the years. There's no evidence of that slowing at all. In two years time we'll be recording on the first blue laser systems with 20-40GB per disc, compared to DVD-RW at 4.7 which replaced the CDRW at only 700MB. Or your computer, which once measured HD storage in MB, and now does so in 100's of GB and will soon reach the Terabyte HDD storage mark.

The electronics seem to be holding back camera resolutions more than physics ATM. When the next generation of in-camera processing technology hits, you can fully expect a jump into the 12-20MP range for APS-FF DSLRs.

Why the heck wouldn't you want it? Computers and storage to take advantage of it won't be far behind, if at all behind, and as they catch up, prices invariable fall.

It's good times, yes, we'll have to buy new cameras to take advantage, but the old stuff will still produce good results and isn't irrelevant.

Do you realize that there's a project underway to bring "digital film" to hollywood? UHDTV 4000x2000 !!! I shudder at the thought of moving that much 24fps, 24 bit info around, but already some one somewhere is hard at work developing both it and the supporting computer technologies to make it workable. Might not see it for 7-8 years, but it will come.

Buck up. You won't ever have a final product that never gets eclipsed, but you will have some even better stuff to play with in the future.

I only see that as a good thing.

toughluck wrote:
[snip]

Suppose you had, say, a 60 MP CFA sensor that would outresolve almost any lens. This would mean that boogers like moiré (both for Foveon and CFA sensors) would be a thing of the past. You could downsample it to, say, a 10 MP image of unprecedented quality, and write that to the card. This would also average out the extra noise.

Petteri
--



Portfolio: [ http://www.seittipaja.fi/index/ ]
Pontification: [ http://www.seittipaja.fi/ ]