Ad
Ad
Ad
Pages: « 1 [2] 3 »   Bottom of Page
Print
Author Topic: Future Medium Format Camera  (Read 6197 times)
Stefan.Steib
Sr. Member
****
Offline Offline

Posts: 414



WWW
« Reply #20 on: November 14, 2011, 03:46:17 AM »
ReplyReply

Eric

I think this is purely political and as I think the task that Leica, Phase/Leaf, Pentax and Hasselblad have to solve, is to cooperate  !
(BTW this idea is spoken out by Henrik Hakansson / Phase One in an Interview he gave for German Profifoto 4/2010)
Because technically it is not a problem at all see here:
http://www.lincoln.ac.uk/news/2011/05/360.asp    which is even topped by this
http://www.engadget.com/2010/08/31/canon-develops-worlds-largest-cmos-sensor-shoots-60fps-video-i/

I think a MF CMOS with sizes up to 6/7 cm could be cheaper than a CCD of 4,5/6 size but of course it depends on the numbers produced.
What I donīt understand why Dalsa, Phillips and Kodak do not switch production or at least broaden their spectrum of offers for this kind of chips.
Of course the makers of smaller formats with even tinier pixel pitches than 5 Micron will tell everything works  wonderful, but for a sufficient number of electrons
to be collected something in the 6 Micron range would be the best compromise as of todays technology (see Canon 1DX  and I believe they know what they are doing)
so a 6/7 chip with around 60-80Mpix would be the perfect solution.
Any larger Resolutions can be done by superexposures, stitching or multishot.
A 6/7 chip would also allow resolutions of 4k video with Canons downsample method. Perfect !
And removing the shutter completely and replace with EVF and live view ,plus rolling or global readout of the CMOS chip (see andor!) would solve any problems which may arise optically, advanced retrofocus lenses could be used.And last but not least: Using a simple piezo Peltier cooling for longer exposures could give nearly noisefree 1 Hour shots (see astronomy - and Leaf had this already 17 years ago in the DCBīs !)


So there is the big question: why arenīt they doing it HuhHuh??

Greetings from Munich
Stefan
« Last Edit: November 14, 2011, 05:54:58 AM by Stefan.Steib » Logged

Because Photography is more than Technology and "as we have done this all the time"
www.hartblei.de     www.hcam.de    www.spectralize.com
BernardLanguillier
Sr. Member
****
Offline Offline

Posts: 7761



WWW
« Reply #21 on: November 14, 2011, 05:38:06 AM »
ReplyReply

So there is the big question: why donīt they do it HuhHuh??

At the risk of looking like an old broken record, is there any possible reason besides "they don't see the economic value of doing it considering where they are technologically today, the cost of R&D they would need to do and the very small series resulting from the high end pricing of most backs"?

The only case where small series can win in electronics is when the absolute level reached by mass products makes it un-appealing for consumer electronic companies to push the envelope farther. This is why high-end audio survives... because main stream audio is already so incredibly good that 99.9% of the population finds it to be enough. It leaves a small niche wide open for those artisans able to get a few more % of performance out of silicon and tubes.

Unfortunately for Dalsa the big names Nikon, Sony and Canon are still not happy about the current level of DRs/resolution of their cameras because they see that customers still want more and think they see the difference, or at least they can make money with high spec sensors. Nobody cares about 0.0001% of THD vs 0.01%, but people still care about 13 stops of DR vs 12.

The yellow cameras we are starting to see, the success of more compact cameras like 4/3 or Nikon 1 series,... are some sign of slow down of the race ahead, but there are still a few more years of high speed progress in the high end of DSLR segment. The more than happens, the smaller the application niches that can only be handled by MF and the lesser the revenue potential for highly priced devices, therefore the lesser appealing the investement needed to do more significant technological progress for MF sensors manufacturers.

Just look at what happened with high end graphic cards 10 years ago... nvidia and AMD are the only players left, their hardware was driven forward by mass market expectation so fast that Silicon graphics just got left behind. It is not going to be that clear for high end photography, but doesn't sound too good with the current pricing model. IMHO.

Cheers,
Bernard
Logged

A few images online here!
Stefan.Steib
Sr. Member
****
Offline Offline

Posts: 414



WWW
« Reply #22 on: November 14, 2011, 06:32:59 AM »
ReplyReply

Bernard

I fear you are right, and to be honest my question was kind of rethorical.
But there are some things that still seem not logical to me:

1.: if this is the status of the MF industry today, are they waiting to disappear without defense ?

2.: If even a University (probably with some 3rd party funding) can afford to have their own custom CMOS be made, why canīt a consortium of
 all involved back-/cameramakers as listed in my former mail ? (using it like the Cosworth machine in Formula one that was standard for all competitors for decades !)

and 3.: This concept could as well bring them back to a broader customer base, using modular chips of differing sizes and resolutions and allowing
more aggressive pricing which could also solve this hen and egg dilemma. As you said Canon, Nikon or Sony wonīt  bother, in some years itīs over,
leaving the actual model in a status like 8/10" Film today. And this is not only the back makers - there are the techcamera makers and a lot of the peripherals
used in this industry will also go down the drain, or maybe leaving a tiny niche for specialists.

And yes of course this also will affect my business, thatīs the reason why I speak it out here !

Regards
Stefan
« Last Edit: November 14, 2011, 06:36:45 AM by Stefan.Steib » Logged

Because Photography is more than Technology and "as we have done this all the time"
www.hartblei.de     www.hcam.de    www.spectralize.com
ondebanks
Sr. Member
****
Offline Offline

Posts: 820



« Reply #23 on: November 14, 2011, 09:15:21 AM »
ReplyReply

OK, nice question by Stefan.

My wishlist (overlaps a lot with previous replies, but has a few new ideas too):

* Mirrorless, thin body. As in, Sony NEX-thin. This thinness permits adapters to just about every lens on the planet; even some of those with insufficient image circle are interesting. It also permits shift-tilt adapters for all our MF lenses.

* Modular backs - digital and (lower priority) film

* 645-sized CMOS sensor in the digital back, with Canon levels of readnoise and dark noise, but not their Bayer CFA filters: rather Kodak's CFA filters (better colour discrimination)

* User-changeable filters over the sensor (choice of: AA + IR-blocking, plain IR-blocking, extended-red (680nm cutoff) IR-blocking, and IR-passing)...in the style of the old Kodak MFDBs and Mamiya ZD

* Live View, of course

* Fully pivoting LCD screen, with detachable WLF-style folding magnifying hood

* Built-in intervalometer (I mean, the Chinese make them as $20 handsets...how hard is it to integrate that into the body?!), and no upper limit to individual exposure times.

* Video? Nah, not a priority for me. [In nearly a year of using a 5DII, I've taken 3 movies, and 2 of those were accidental!]. There could be a second, more expensive version with video features. I think some people here on LuLa seem to forget that most of us, even now, still really only want to shoot very high quality still photos.

* Canon-style RAW format where "raw MEANS raw". No dark frame subtraction (unless specifically chosen by the user), no bias subtraction, and no long exposure median filtering or any such f*cking around by the firmware. DNG sounds fine as a vehicle for this format.

* Runs off high-capacity Li-Ion or 12V (run a cable from a large camping/astro battery, or your car's cigarette lighter port in a pinch).

Ray
Logged
marfa.tx
Newbie
*
Offline Offline

Posts: 27


« Reply #24 on: November 14, 2011, 09:33:21 AM »
ReplyReply

hope this doesn't kidnap the thread... but a practical Q:

is there a set of drawings or specs publicly available for the various lens mounts?

--- just noodling in CAD program now ----
Logged

-------
richard
bernhardmarks
Newbie
*
Offline Offline

Posts: 9


WWW
« Reply #25 on: November 14, 2011, 11:18:50 AM »
ReplyReply

For architectural photography:

EOS 5D with double size sensor (2 sensors side by side), 48x36mm, 42 mp or more
no (fast) autofocus needed
no video needed
size, weight do not matter
programs: av,tv,m, c
built-in-spirit level
adapted TS-E lenses 24, 40, 70, 100 with tripod connection
Logged

ErikKaffehr
Sr. Member
****
Online Online

Posts: 7251


WWW
« Reply #26 on: November 14, 2011, 01:47:46 PM »
ReplyReply

Hi,

I agree mostly. I don't really agree on the utility of a "waist level viewfinder loupe" as LCDs have selectable magnification and actual pixels are needed for live view.

Another point is that I'd suggest that Video is essentially free, it does not add complexity or cost. That said I see little use for MF video.

Best regards
Erik

OK, nice question by Stefan.

My wishlist (overlaps a lot with previous replies, but has a few new ideas too):

* Mirrorless, thin body. As in, Sony NEX-thin. This thinness permits adapters to just about every lens on the planet; even some of those with insufficient image circle are interesting. It also permits shift-tilt adapters for all our MF lenses.

* Modular backs - digital and (lower priority) film

* 645-sized CMOS sensor in the digital back, with Canon levels of readnoise and dark noise, but not their Bayer CFA filters: rather Kodak's CFA filters (better colour discrimination)

* User-changeable filters over the sensor (choice of: AA + IR-blocking, plain IR-blocking, extended-red (680nm cutoff) IR-blocking, and IR-passing)...in the style of the old Kodak MFDBs and Mamiya ZD

* Live View, of course

* Fully pivoting LCD screen, with detachable WLF-style folding magnifying hood

* Built-in intervalometer (I mean, the Chinese make them as $20 handsets...how hard is it to integrate that into the body?!), and no upper limit to individual exposure times.

* Video? Nah, not a priority for me. [In nearly a year of using a 5DII, I've taken 3 movies, and 2 of those were accidental!]. There could be a second, more expensive version with video features. I think some people here on LuLa seem to forget that most of us, even now, still really only want to shoot very high quality still photos.

* Canon-style RAW format where "raw MEANS raw". No dark frame subtraction (unless specifically chosen by the user), no bias subtraction, and no long exposure median filtering or any such f*cking around by the firmware. DNG sounds fine as a vehicle for this format.

* Runs off high-capacity Li-Ion or 12V (run a cable from a large camping/astro battery, or your car's cigarette lighter port in a pinch).

Ray
Logged

paul_jones
Sr. Member
****
Offline Offline

Posts: 565


WWW
« Reply #27 on: November 14, 2011, 02:13:07 PM »
ReplyReply

I've always wanted an evf or external monitor to shoot stills with. I've experimented with an external monitor for a while- through a 1ds mk3- this worked well- and fast. but the 5d is slow from monitoring to click, i also tried the 1dx and this is still a little slow.

but i still want an optical finder. the number one complaint i hear about the alexa (which doesn't get many complaints) is that the lcd evf is a little delayed, so when you quickly move the camera, its very annoying and hard to follow things well. i guess this is why alexa is going to offer a optical finder version the camera- theres nothing quite like seeing through the lens properly. having both options would be perfect.

also, i still want to focus by eye, and I'm not sure lcds are sharp enough yet to do this. i want to be able to control the amount, or position of the focal plane, not have a "focus point" do it for me.

but when i want focus points- i want them to be selectable over the whole frame- i often have subjects outside the area of focus points with my canon, and still have to move the camera sideways to use the closest point. the iPhone way of touching a screen- exactly where you want to focus is a perfect solution for me.

paul

Logged

check my new website
http://www.paulrossjones.com
Stefan.Steib
Sr. Member
****
Offline Offline

Posts: 414



WWW
« Reply #28 on: November 14, 2011, 02:24:47 PM »
ReplyReply

Erik

Imagine a soon to be had 4k display attached to that MF Video out for work and image composition - especially when used in production environments.

Paul

I believe this can be solved, it may be something like a one frame delay, for fast movements this may be annoying, but I guess if you want to do sports or
fashion shows or playing children itīs better to use a 35mm maybe a D1x. The usage target group of an MF system will probably not have this problem.
And the optical (mirror reflex) viewfinder will lock off all other options for a modern camera. So the choice is - Life view/EVF and wideangles, focusing when itīs dark or at 100 % or more zoom in and absolute focus accuracy, even with movements......... or a mirror.

greetings from Munich
Stefan
Logged

Because Photography is more than Technology and "as we have done this all the time"
www.hartblei.de     www.hcam.de    www.spectralize.com
ErikKaffehr
Sr. Member
****
Online Online

Posts: 7251


WWW
« Reply #29 on: November 14, 2011, 04:33:17 PM »
ReplyReply

Hi!

Regarding video I would say that an APS-C sensor is perfectly OK for 4K video, I think RED is already doing that. But I really don't see MF being used for video, it's hard to see the major advantage and DoF will be very limited.

Regarding MF for photography I'd say that the digital back is excellent invention. You can put it a different cameras as needed. But really, I'd suggest that DSLRs are good enough for any kind of shooting you can do without mirror lockup. You shoot without MLU, 75% of the pixels are essentially gone, or if you stop down to f/22 75% of the pixels are also gone! I presume that the main driver for MF is sharpness, and to utilize it takes exacting work.

Best regards
Erik

Erik

Imagine a soon to be had 4k display attached to that MF Video out for work and image composition - especially when used in production environments.

Paul

I believe this can be solved, it may be something like a one frame delay, for fast movements this may be annoying, but I guess if you want to do sports or
fashion shows or playing children itīs better to use a 35mm maybe a D1x. The usage target group of an MF system will probably not have this problem.
And the optical (mirror reflex) viewfinder will lock off all other options for a modern camera. So the choice is - Life view/EVF and wideangles, focusing when itīs dark or at 100 % or more zoom in and absolute focus accuracy, even with movements......... or a mirror.

greetings from Munich
Stefan

Logged

BJL
Sr. Member
****
Offline Offline

Posts: 5121


« Reply #30 on: November 14, 2011, 08:02:05 PM »
ReplyReply

Because technically it is not a problem at all see here:
http://www.lincoln.ac.uk/news/2011/05/360.asp    which is even topped by this
http://www.engadget.com/2010/08/31/canon-develops-worlds-largest-cmos-sensor-shoots-60fps-video-i/

I think a MF CMOS with sizes up to 6/7 cm could be cheaper than a CCD of 4,5/6 size but of course it depends on the numbers produced.

What I donīt understand why Dalsa, Phillips and Kodak do not switch production or at least broaden their spectrum of offers for this kind of chips.

I agree with Bernard that when multiple competing companies (sensor makers Teledyne-Dalsa, Kodak, Sony, Panasonic, Canon etc.) do not produce an often-asked for product, it is probably not a conspiracy, or stupidity, but instead is due to inadequate potential for return on the investment [ROI].

The sensors you mention above do not come close to proving that it is possible to produce a larger-than-35mm-format CMOS sensor suitable for MF usage at a competitive price. Firstly, they are of extremely low resolution, with 50nm to 100nm pixel pitch and so only abut 6MP on that huge Lincoln full wafer sensor. The far smaller feature sizes needed for a viable MF sensor require fabrication with steppers of smaller field size (a maximum of 26mm by 33mm has been the industry-wide standard for many years now), and so stitching together of more fields, with greater precision need in the stitching, reducing yields and thus increasing costs. Secondly the applications for those sensors can tolerate visible defects like join lines that would be unacceptable in a DMF camera. Dalsa in fact also offers huge CMOS sensors for X-rays and such, but they are made by butting together multiple sensor chips, leaving thin but visible join lines. Thirdly, we have no idea of the prices of those sensors: their customers might be willing to pay hundreds of thousands of dollars for a sensor, since they go into X-ray machines and telescopes that probably cost a million or more.

There is a big gap between technologically feasible and economically viable (except perhaps when the customers are primarily military/government funded, or more generally are spending lots of other peoples' money, not their own.)


P.S. the former Phillips sensor business was acquired by Dalsa (but still kept a lot of its facilities in the Netherlands), and Dalsa was in turn acquired by Teledyne. Also, Kodak had a CMOS sensor division, but shut it down.
Logged
uaiomex
Sr. Member
****
Offline Offline

Posts: 973


WWW
« Reply #31 on: November 14, 2011, 08:25:32 PM »
ReplyReply

I've asked the same questions over and over again. Why don't they do agressive pricing? They should proceed a hostile takeover on Dalsa! Cheesy

1. Why are DMF backs so expensive? Because the market is too small!
2. Why is the market so small? Because the DMF backs are too expensive!

Yes, the proverbial hen and egg which was first.

But really, do you believe number one? I don't. For many reasons. It is true that the MF market shrunk considerably with digital but also the world's population almost doubled since 1975 during the medium format film golden era. The world then was 4 billion. Today we are 7 billion. In 1975 there was Kodak, Fuji, Agfa, Ilford, Konica, etc., true powerhouses of film manufacturing and some other lesser ones. We also had Hasselblad, Mamiya, Bronica, Fuji, Rollei, Contax and some others I can't recall right now. They all were doing just fine competing for the same markets in a world of 4 billion.
Today, we have only 2 commercial sensor manufacturers. Dalsa and ISS ex-Kodak. MF cameras makers is mainly about Mamiya/P1 and Hasselblad. (Leica, Hy6 and Pentax market shares are still too small, they're newbies) So, how in this world (of 7 billion) 2 contenders in sensor making and 2 contenders in camera making can't find their ways to have substantial profits by selling enough gear to lower prices to human levels? I beats me, really.

I know there must be some other explanations to this riddle but in the meantime I have a (dreadful) theory. DMF manufacturers know their days are counted. (they know better than us, don't they?) They know that the smaller formats will eventually win the commercial photography race entirely. They know that in order to compete tomorrow they have to invest today impossible amounts of money and nothing guarantees success. So, they're just enjoying life to the max until lady in black knocks at the front door. This could also answer question number one.
Crazy? certainly, but let's remember that the main contenders are just survivors (remnants) from the film era and that mf sensor manufacturers main income is not from commercial photography but from science and the military.

But I dearly love medium format photography. So, I'd like to think and believe that the survival of MF resides somewhere in answering question number 2. If this is true, the only thing lacking is will and balls.

To all DMF manufacturers:
Please, form a consortium and fight for better and cheaper sensors (Stefan's idea). Get to be more creative, more intuitive, bolder. Do a real effort. Slash prices by half tomorrow and see how sales increase 10 fold the day after tomorrow. Hasselblad: Stop neglecting the V cameras. Just imagine how many CFV39 backs with a built-in rotating sensor would you sell at half the price. Phase One: Push in newer products the technology inside the P45+ back. It is the only back in the world that can do long exposures.

Please, please, make me a failed prophet.
Sincerely
Eduardo

« Last Edit: November 14, 2011, 09:11:43 PM by uaiomex » Logged
Sareesh Sudhakaran
Sr. Member
****
Offline Offline

Posts: 547


WWW
« Reply #32 on: November 14, 2011, 10:10:49 PM »
ReplyReply

You shoot without MLU, 75% of the pixels are essentially gone

Erik - can you give me some links that conclusively prove MLU actually helps in sharpening? I tried some amateurish tests on my DSLR earlier and the results showed MLU hardly made any difference. I have tried googling it but can't find any solid tests for it. Thanks for your time.
Logged

Get the Free Comprehensive Guide to Rigging ANY Camera - one guide to rig them all - DSLRs to the Arri Alexa.
BernardLanguillier
Sr. Member
****
Offline Offline

Posts: 7761



WWW
« Reply #33 on: November 14, 2011, 10:56:17 PM »
ReplyReply

I've asked the same questions over and over again. Why don't they do agressive pricing? They should proceed a hostile takeover on Dalsa! Cheesy

1. Why are DMF backs so expensive? Because the market is too small!
2. Why is the market so small? Because the DMF backs are too expensive!

Yes, the proverbial hen and egg which was first.

But really, do you believe number one? I don't.

The fact that Pentax is able to sell an extremely advanced and well built 645D for 9000 US$ is the obvious proof that there is zero reason why an IQ180 back should cost more than 12.500 US$ if its price were set in a way as to grasp a large market share. Granted, Pentax is able to re-use some of the technology they develop for the mass market, but the 645D sells for what it sells with a significant margin built-in...

Perhaps, that's the difference between the ambitious businessman running Red and the small players leading Phaseone and Hassy? :-)

The problem though, is that the 2 best MF platforms have been captured by these small players which render a huge disservice to the photographic community by essentially killing MF...

How about creating a movement "give MF back to the people"?  Grin That's what I would do if I were the Phasone CEO:

- make the development of their next platform an open source effort - publish all the drawings starting with the mount with its full specification,
- get rid of all the back except the IQ80 (cut its price in 3 or 4) and another one below with a smaller sensor and sell it for 6.000 US$ new (that is going to be their mass market product),
- open up the software also with an Android layer enabling the development of custom applications in back.

Cheers,
Bernard
« Last Edit: November 14, 2011, 11:08:16 PM by BernardLanguillier » Logged

A few images online here!
ErikKaffehr
Sr. Member
****
Online Online

Posts: 7251


WWW
« Reply #34 on: November 14, 2011, 11:13:34 PM »
ReplyReply

Hi,

Both are based on my own observations. Admittedly it may differ from camera to camera, tripod to tripod. The test I have done was a couple of years ago.

I made a series of exposures with a 200 mm lens from 1/15s to 1/125s

A) Camera on tripod MLU and cable release no IS
B) Camera on tripod no MLU, cable release no IS
C) Camera on tripod no MLU, cable release IS
D) Free hand, no MLU, hand release, IS

All images looked sharp at actual pixels!

After that I analyzed all 16 images with Imatest. The A series had best resolution that camera could achieve, essentially with no variation.

The B, C and D series lost about half the resolution, the free hand shots were pretty similar to the others. But I used to be a decent free hand shooter.

After doing the Imatest analysis I rechecked the images visually and could see that the B, C, D series were slightly softer, so we had half the resolution without being visually very obvious.

Cutting resolution in half wastes 75% of the pixels. I never wrote about the test, because I had problems keeping all the data and images apart. Let's say that I'm quite religious about MLU since than.

Lloyd Chambers has written much on the issue, like here: http://diglloyd.com/prem/prot/DAP/SHARP/publish/TheSharpestImage-ShootingMode.html , unfortunately Lloyd's articles are mostly subscriber only.

The effect of stopping down is demonstrated here: http://echophoto.dnsalias.net/ekr/index.php/photoarticles/49-dof-in-digital-pictures?start=1


Best regards
Erik


Erik - can you give me some links that conclusively prove MLU actually helps in sharpening? I tried some amateurish tests on my DSLR earlier and the results showed MLU hardly made any difference. I have tried googling it but can't find any solid tests for it. Thanks for your time.
Logged

ErikKaffehr
Sr. Member
****
Online Online

Posts: 7251


WWW
« Reply #35 on: November 15, 2011, 12:05:42 AM »
ReplyReply

Hi,

I would believe that Phase One is essentially a digital back company and that they went into the camera business after Hasselblad locked down their system. I also assume that Phase One sets their prices so they maximize return on investment while keeping risk down. They are not in the charity business...

I don't know that much about sensor prices but I would assume that a "full frame" 645 sensor is significantly more expensive than a cropped frame sensor, but you are probably right that if Pentax can sell a 40 MP digital back with and advanced MF SLR built in for 9000 US$ the sensor chip itself must cost far less than that.

I'm not that happy about cropped MF chips, however. Having a crop factor essentially eliminates the wide angle end. There are some new wide angles for cropped formate but they are very expensive. Taking your example, I guess I would rather pay 12500 US$ for your hypothetical high end back than buy your hypothetical low end back for 8000 US$ and a wide angle for another 4500 US$.

The other factor is that bigger is better. A bigger sensor collect more photons and will have better MTF. I guess the industry has settled on 645 size sensors and there are many fine lenses from Alpa, Phase One, Rodenstock and Schneider for that format. Add to that lenses from Mamiya and Hasselblad. I'd say that affordable full frame MFDBs would be very interesting!

On the other hand, 135 DSLRs are perfectly good enough for most jobs and have also some developments going on. Good enough at a fraction of the price is tough competition!

My main interest is the flexibility of the digital backs. You can put them on technical cameras, optical bench type cameras, MF SLRs. That's a great flexibility. Could they just keep prices down while adding live view, and CMOS sensor with on chip converters!

Best regards
Erik


The fact that Pentax is able to sell an extremely advanced and well built 645D for 9000 US$ is the obvious proof that there is zero reason why an IQ180 back should cost more than 12.500 US$ if its price were set in a way as to grasp a large market share. Granted, Pentax is able to re-use some of the technology they develop for the mass market, but the 645D sells for what it sells with a significant margin built-in...

Perhaps, that's the difference between the ambitious businessman running Red and the small players leading Phaseone and Hassy? :-)

The problem though, is that the 2 best MF platforms have been captured by these small players which render a huge disservice to the photographic community by essentially killing MF...

How about creating a movement "give MF back to the people"?  Grin That's what I would do if I were the Phasone CEO:

- make the development of their next platform an open source effort - publish all the drawings starting with the mount with its full specification,
- get rid of all the back except the IQ80 (cut its price in 3 or 4) and another one below with a smaller sensor and sell it for 6.000 US$ new (that is going to be their mass market product),
- open up the software also with an Android layer enabling the development of custom applications in back.

Cheers,
Bernard

Logged

Stefan.Steib
Sr. Member
****
Offline Offline

Posts: 414



WWW
« Reply #36 on: November 15, 2011, 06:18:34 AM »
ReplyReply

@BJL

now this is very interesting and logical information- THANKS !
Just one core question arises to me from that - if the pixel pitch of 5,2 Micron can be achieved on a 53.7 x 40.4mm chip size for a CCD, why canīt this be produced as a CMOS ?
I also remember that the registers of the actual 80mpix chips seem to be parted in 4 (there were several threads where people found this when extremely underexposing their 80mpix and pull up the channels the four parts of the chip became visible) so again:  if this is done on a CCD why not for a CMOS ? Are there visible lines on the CCDīs ?

And after reading this:    
http://www.siliconimaging.com/ARTICLES/CMOS%20PRIMER.htm      
http://www.siliconimaging.com/ARTICLES/Sensor%20Mag%20Article.htm
http://www.siliconimaging.com/ARTICLES/cmos_advantages_over_ccd.htm

see excerpt:

"...Integration - Because CMOS Imagers are created in the same process as processors, memories and other major components, CMOS Imagers can integrated with these same components onto a single piece of silicon. In contrast, CCDs are made in a specialized process and require multiple clocks and inputs. This feature limits CCDs to discrete systems, which in the long run will put CMOS Imagers at a cost advantage, as well as limit what kinds of portable devices CCDs can be integrated into....."   and

".....Reduced Power Consumption - because of all the external clocks needed to "bucket brigade" each pixel, CCDs are inherently power hungry. Every clock is essentially charging and discharging large capacitors in the CCD array. In contrast CMOS imagers require only a single voltage input and clock, meaning they consume much less power than CCDs, a feature that is critical for portable, battery operated devices.

Pixel Addressibility - CCDs use of the bucket brigade to transfer pixel values means that individual pixels in a CCD cannot be read individually. CMOS imagers on the other hand have the pixels in an x-y grid allowing pixels to be read individually. This means that CMOS imagers will be able to do functions such as "windowing", where only a small sample of the imager is read, image stabilization to remove jitters from camcorders, motion tracking and other advanced imaging techniques internally that CCDs cannot do.

Manufacturing Cost - Since CMOS imagers are manufactured in the same process as memories, processors and other high-volume devices, CMOS imagers can take advantage of process improvements and cost reductions these devices drive throughout the industry....."

...I have even more questions: Sorry if insisting but to me this seems more to be a decision thing and caused by company policies and shareholder interests than by technology !
What happened to Kodak- they have enthusiastically published these articles above in 2001 so why (besides cutting down anything they ever started) did they shut down CMOS fabrication ? Are there any technical reasons ?

TIA and greetings from Munich
Stefan
« Last Edit: November 15, 2011, 06:35:48 AM by Stefan.Steib » Logged

Because Photography is more than Technology and "as we have done this all the time"
www.hartblei.de     www.hcam.de    www.spectralize.com
EricWHiss
Sr. Member
****
Offline Offline

Posts: 2372



WWW
« Reply #37 on: November 15, 2011, 09:47:22 AM »
ReplyReply


I also remember that the registers of the actual 80mpix chips seem to be parted in 4 (there were several threads where people found this when extremely underexposing their 80mpix and pull up the channels the four parts of the chip became visible) so again:  

I was one that posted an image from the Aptus 12 earlier on LL.  So just to clarify:  The 80mp chips have 4 sets of gain electronics so people can sometimes see quarter lines in the resulting images, but the actual chips look to have 8 pieces or sections when you look at them.
Logged

Authorized Rolleiflex Dealer:
Find product information, download user manuals, or purchase online - Rolleiflex USA
Stefan.Steib
Sr. Member
****
Offline Offline

Posts: 414



WWW
« Reply #38 on: November 15, 2011, 06:07:24 PM »
ReplyReply

The clock is ticking - we have a 70 Mpix CMOS available with 3 frames per second, it was shown in October - the only problem is: Itīs 3.1 micron and only 31x21mm in Size.

http://www.cmosis.com/news/press_releases/2011/cmosis_chr70m_ultra-high-resolution_cmos_industrial_image_sensor

Iīm sure this is only the first of a new breed of chips to come.

Greetings from Munich
Stefan
Logged

Because Photography is more than Technology and "as we have done this all the time"
www.hartblei.de     www.hcam.de    www.spectralize.com
Sareesh Sudhakaran
Sr. Member
****
Offline Offline

Posts: 547


WWW
« Reply #39 on: November 15, 2011, 09:30:08 PM »
ReplyReply

Hi,

Both are based on my own observations. Admittedly it may differ from camera to camera, tripod to tripod. The test I have done was a couple of years ago.

I made a series of exposures with a 200 mm lens from 1/15s to 1/125s

A) Camera on tripod MLU and cable release no IS
B) Camera on tripod no MLU, cable release no IS
C) Camera on tripod no MLU, cable release IS
D) Free hand, no MLU, hand release, IS

All images looked sharp at actual pixels!

After that I analyzed all 16 images with Imatest. The A series had best resolution that camera could achieve, essentially with no variation.

The B, C and D series lost about half the resolution, the free hand shots were pretty similar to the others. But I used to be a decent free hand shooter.

After doing the Imatest analysis I rechecked the images visually and could see that the B, C, D series were slightly softer, so we had half the resolution without being visually very obvious.

Cutting resolution in half wastes 75% of the pixels. I never wrote about the test, because I had problems keeping all the data and images apart. Let's say that I'm quite religious about MLU since than.

Lloyd Chambers has written much on the issue, like here: http://diglloyd.com/prem/prot/DAP/SHARP/publish/TheSharpestImage-ShootingMode.html , unfortunately Lloyd's articles are mostly subscriber only.

The effect of stopping down is demonstrated here: http://echophoto.dnsalias.net/ekr/index.php/photoarticles/49-dof-in-digital-pictures?start=1


Best regards
Erik



Thank you for the info Erik. Appreciate it.
Logged

Get the Free Comprehensive Guide to Rigging ANY Camera - one guide to rig them all - DSLRs to the Arri Alexa.
Pages: « 1 [2] 3 »   Top of Page
Print
Jump to:  

Ad
Ad
Ad