Ad
Ad
Ad
Pages: [1] 2 3 ... 8 »   Bottom of Page
Print
Author Topic: Open Source CMOS Medium Format Camera/Back - An initial exploration  (Read 13514 times)
eronald
Sr. Member
****
Offline Offline

Posts: 4205



« on: February 24, 2014, 07:58:49 PM »
ReplyReply

Hi -

There are a variety of reasons why an open medium format camera platform makes sense. In this topic we could discuss whether there would be people interested in investing time and money or contributing such skills in such a product, and what is reasonably attainable in a reasonable amount of time.

In order to fix expectations, I would like to remind people that Linux Torvalds was not very skilled technically when he started Linux, but he got the project off the ground and the pre-existing GNU software suite made the system viable.

Also, the Linux project was  initially a big success for scientists and servers run by professional sysadmins, but  failed to convince the users of conventional desktop PCs. Of course things changed, and now a substantial fraction of the world's phones and tablets run Linux.

In the same way, I believe it would be feasible to do an open source tethered setup in limited time; external processing of the files could then take over: Many on this forum are skilled software engineers. I am not so sure that a mobile camera is an attainable short time goal. The Magic Lantern project has proved very useful to the film-making community, but what we discuss here starts from the sensor, not from an existing working and integrated platform.

Let's discuss -

- what you'd like to see as useful features
- what platform should be a base for the mechanics/optics
- what skills you could bring to the table, and how much time you could contribute


Edmund
« Last Edit: March 07, 2014, 10:06:55 AM by eronald » Logged
Doug Peterson
Sr. Member
****
Offline Offline

Posts: 2867


WWW
« Reply #1 on: February 24, 2014, 08:07:41 PM »
ReplyReply

I'm glad to volunteer our inventory of gear for any relevant comparisons or tests. If you come to our NYC studio I'll gladly put you in a room with any of our gear for the day.

This may seem counter to our company's at-the-end-of-the-day purpose of selling medium format backs I maintain that any interest in medium format or high-quality imaging increases demand for our high-end products. So in addition to the personal/academic interest I think it's business-purpose supportable as well.
Logged

DOUG PETERSON (dep@digitaltransitions.com), Digital Transitions
Dealer for Phase One, Mamiya Leaf, Arca-Swiss, Cambo, Profoto
Office: 877.367.8537
Cell: 740.707.2183
Phase One IQ250 FAQ
pixjohn
Sr. Member
****
Offline Offline

Posts: 673


« Reply #2 on: February 24, 2014, 08:08:23 PM »
ReplyReply

With all do respect, It sounds more like wishful thinking then reality. To many factors, and to small a market would make it unrealistic.  The other factor is money. When you can have a bunch of 16 year old kid's working on coding for free or paying out $100,000's in RD what is more likely? Maybe try the TV show shark tank Smiley
« Last Edit: February 24, 2014, 08:12:35 PM by pixjohn » Logged
eronald
Sr. Member
****
Offline Offline

Posts: 4205



« Reply #3 on: February 24, 2014, 08:12:54 PM »
ReplyReply

Doug,

 I am very glad you would like to help make this a fun party Smiley
 Quite a few companies have found open source to be a good business model.
 I would expect that if something like this takes off there would be some specialist vendors like you who would qualify the hardware and distribute it. 

Edmund

I'm glad to volunteer our inventory of gear for any relevant comparisons or tests. If you come to our NYC studio I'll gladly put you in a room with any of our gear for the day.

This may seem counter to our company's at-the-end-of-the-day purpose of selling medium format backs I maintain that any interest in medium format or high-quality imaging increases demand for our high-end products. So in addition to the personal/academic interest I think it's business-purpose supportable as well.
« Last Edit: February 24, 2014, 08:16:24 PM by eronald » Logged
matoqui
Newbie
*
Offline Offline

Posts: 40


WWW
« Reply #4 on: February 24, 2014, 09:00:26 PM »
ReplyReply

These people built an open source camera to use it in computational photography. Not medium format, however.

http://graphics.stanford.edu/papers/fcam/
Logged

bpepz
Full Member
***
Offline Offline

Posts: 203


« Reply #5 on: February 24, 2014, 11:51:55 PM »
ReplyReply

Glad to see interest in this. I like the idea of starting simple on a tethered solution. If a simple working prototype could be made, it would be prime material for a kickstarter project to bring costs down and go even further. Not much I can really contribute though asides from cash if a kick starter is ever made. But I can say what I would like to see, and I will try to keep it realistic. I am not too worried about specs, but more or less the "vision" or direction this could go in.

I would like to see an open and modular platform using medium format sized sensors that can be adapted in various ways at a reasonable price. But don't we all right? Just having open firmware would be amazing. I can see an entire ecosystem of custom firmwares for specialized purposes and continuous improvements from the community. Also exciting would be, once you have a foundation built, custom 3rd party accessories or modular addons could completely change the way it can be used too. Maybe one it has developed a bit, and moved from tethered to a mobile solution, you could install a "CF card" module, or if you don't mind the extra size and weight, a SSD module instead. Being able to customize and specialize would make it worlds over more attractive then closed 35mm and MF systems today. I can probably rant on forever about this, but that is sort of where I would like to see it go. Basically I want to see something like magic lantern for MF, but not just on the firmware side, the hardware side too. I think it would make an amazing combination.
Logged
Sareesh Sudhakaran
Sr. Member
****
Offline Offline

Posts: 547


WWW
« Reply #6 on: February 25, 2014, 01:01:10 AM »
ReplyReply

Count me in.


In the same way, I believe it would be feasible to do an open source tethered setup in limited time; external processing of the files could then take over: Many on this forum are skilled software engineers. I am not so sure that a mobile camera is an attainable short time goal.

Why rush? A half-baked product won't sell, even if you get it to work.

I believe one should first focus on the market. Who will buy it, why will they and for how much, and how will it be shipped?

A tethered chip to custom software is what you'll find on many researchers' and students' tables. It's not that hard to do. A 'product' is a much more complicated thing. The Digital Bolex Camera is an excellent case study.

Quote

Let's discuss -

- what you'd like to see as useful features

I think we need to start looking at the market in depth before a single electron is allowed to misbehave.

Personally, all I could ever hope for is a Mamiya 7 with a replaceable back, with room to grow until it is 6x7. Small, minimal, easy to calibrate, and a rangefinder already built in (no need to design an EVF or mirror mechanism). Maybe it can start with a fixed leaf shutter lens, an 80mm f/2.8.

No tethering and no mirror - why focus on 'satisfied' MFDB owners? The smart thing to do is go for the Pentax 645D or DSLR or Leica M crowd. Tethering means you need software and a computer. Eliminate the variables and focus on the basics - sensor, firmware, metering, color, body and one lens.

DNG files on CF or SD cards (licensing fees apply from SD). Maybe Sony will have a better deal on their media. Zeiss could supply the lens. Once the camera is up and running, the CF card can be wi-fi enabled for wireless tethering. Let someone right the tethering tools later, right?

The 44x33 Sony sensor definitely does video, but that is another ball game entirely.

Where will it be made and which country is the most cost effective in the immediate sales cycle?

Which license will be be under? Open source is a generic term. Who will look after patenting, accounting and admin?

What would be the long term goals of this enterprise? What structure will it take?

Quote

- what skills you could bring to the table, and how much time you could contribute


Marketing (both real world and Internet), Documentation, Web Development and Support is what I'm good at. I also know a wee bit of Python and created an image processing application in C in college (ages ago, and I can't recall what I did or how I did it). I can commit time, definitely, but only if the core team is in it for the long haul. If a project thinks it will take X time, it will really take 3X+. From experience.

Maybe we can confab a couple of hundred times to know for sure.
Logged

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

Posts: 1703



WWW
« Reply #7 on: February 25, 2014, 03:45:11 AM »
ReplyReply

With all do respect, It sounds more like wishful thinking then reality. To many factors, and to small a market would make it unrealistic.  The other factor is money. When you can have a bunch of 16 year old kid's working on coding for free or paying out $100,000's in RD what is more likely? Maybe try the TV show shark tank Smiley

Hey, don't complain, hopefully it'll confine all the geeks to one thread  Wink
Logged

Ken R
Sr. Member
****
Offline Offline

Posts: 532


WWW
« Reply #8 on: February 25, 2014, 07:26:21 AM »
ReplyReply

Hi -

There are a variety of reasons why an open medium format camera platform makes sense. In this topic we could discuss whether there would be people interested in investing time and money or contributing such skills in such a product, and what is reasonably attainable in a reasonable amount of time.

In order to fix expectations, I would like to remind people that Linux Torvalds was not very skilled technically when he started Linux, but he got the project off the ground and the pre-existing GNU software suite made the system viable.

Also, the Linux project was  initially a big success for scientists and servers run by professional sysadmins, but  failed to convince the users of conventional desktop PCs. Of course things changed, and now a substantial fraction of the world's phones and tablets run Linux.

In the same way, I believe it would be feasible to do an open source tethered setup in limited time; external processing of the files could then take over: Many on this forum are skilled software engineers. I am not so sure that a mobile camera is an attainable short time goal. The Magic Lantern project has proved very useful to the film-making community, but what we discuss here starts from the sensor, not from an existing working and integrated platform.

Let's discuss -

- what you'd like to see as useful features
- what platform should be a base for the mechanics/optics
- what skills you could bring to the table, and how much time you could contribute


Edmund

Let's not forget about Jim Jannard and RED. His discontent with the current state of affairs of digital cinema propelled him to do something about it and change it forever. (At the time the RED ONE was introduced for 20k, Sony's Digital Cinema Cameras were 10 times the price) He did have a lot of money to throw into the project though. Smiley
« Last Edit: February 25, 2014, 07:31:28 AM by Ken R » Logged
bpepz
Full Member
***
Offline Offline

Posts: 203


« Reply #9 on: February 25, 2014, 08:25:07 AM »
ReplyReply

Count me in.

Why rush? A half-baked product won't sell, even if you get it to work.

I believe one should first focus on the market. Who will buy it, why will they and for how much, and how will it be shipped?

A tethered chip to custom software is what you'll find on many researchers' and students' tables. It's not that hard to do. A 'product' is a much more complicated thing. The Digital Bolex Camera is an excellent case study.

I think we need to start looking at the market in depth before a single electron is allowed to misbehave.

Personally, all I could ever hope for is a Mamiya 7 with a replaceable back, with room to grow until it is 6x7. Small, minimal, easy to calibrate, and a rangefinder already built in (no need to design an EVF or mirror mechanism). Maybe it can start with a fixed leaf shutter lens, an 80mm f/2.8.

No tethering and no mirror - why focus on 'satisfied' MFDB owners? The smart thing to do is go for the Pentax 645D or DSLR or Leica M crowd. Tethering means you need software and a computer. Eliminate the variables and focus on the basics - sensor, firmware, metering, color, body and one lens.

DNG files on CF or SD cards (licensing fees apply from SD). Maybe Sony will have a better deal on their media. Zeiss could supply the lens. Once the camera is up and running, the CF card can be wi-fi enabled for wireless tethering. Let someone right the tethering tools later, right?

The 44x33 Sony sensor definitely does video, but that is another ball game entirely.

Where will it be made and which country is the most cost effective in the immediate sales cycle?

Which license will be be under? Open source is a generic term. Who will look after patenting, accounting and admin?

What would be the long term goals of this enterprise? What structure will it take?

Marketing (both real world and Internet), Documentation, Web Development and Support is what I'm good at. I also know a wee bit of Python and created an image processing application in C in college (ages ago, and I can't recall what I did or how I did it). I can commit time, definitely, but only if the core team is in it for the long haul. If a project thinks it will take X time, it will really take 3X+. From experience.

Maybe we can confab a couple of hundred times to know for sure.

Maybe we are over thinking this? I don't really see the point of making an entire camera system with shutter and new lens when there are heaps of great cameras like the Mamiya 7 as you mentioned, and of course stuff like the RZ67, Hasselblad V cameras, and Fuji GX680. It would just be a matter of making a sensor "box" with a modular mount and a couple of sync ports. Also, maybe it does not need to be a huge scale product with distribution chains, offices in every major city, and 24/7 tech support. I am exaggerating a bit, but I like Paul Buffs model of just being able to buy directly from the maker and not not worrying about middlemen, or even better the bare bones Chinese ebay model, my personal favorite.

However, that being said, a single rangefinder, non-tethered back would still be really really cool.

Logged
Rob C
Sr. Member
****
Offline Offline

Posts: 12213


« Reply #10 on: February 25, 2014, 08:33:07 AM »
ReplyReply

I'm no geek, unfortunately, so can't contribute anything there, but I can contribute to the wish-list: Hasselblad 500 series-compatible large-sensor square backs that don't cost the Earth would do just fine. The bodies could, I'm sure, be recreated again, and the user interface can't be bettered. IMO.

Why reinvent the wheel when all it needs is a new hub?

Rob C
Logged

powerslave12r
Full Member
***
Offline Offline

Posts: 117



WWW
« Reply #11 on: February 25, 2014, 08:35:32 AM »
ReplyReply

With all do respect, It sounds more like wishful thinking then reality. To many factors, and to small a market would make it unrealistic.  The other factor is money. When you can have a bunch of 16 year old kid's working on coding for free or paying out $100,000's in RD what is more likely? Maybe try the TV show shark tank Smiley

I think you are underestimating the talent pool willing to work on open source stuff. The list of open source accomplishments today are mind boggling and if you started at the beginning, trying to envision an open source project of such importance from ground zero, you would be forgiven for thinking it's not possible.

This is an insanely great idea. The truth remains that open source hardware projects are harder than open source software projects.

Until more affordable 3D printers!
Logged

http://www.flickr.com/garagenoise
DP2M | X-M1 | 6D | TS-E24IIL | EF24-105L
Doug Peterson
Sr. Member
****
Offline Offline

Posts: 2867


WWW
« Reply #12 on: February 25, 2014, 08:40:11 AM »
ReplyReply

My dad once told me "the best way to compete, is not to". I don't know if that's attributed to someone he was quoting, and I'm sure he wasn't the first to have the thought, but it always stuck with me.

Meaning if you make a digital back in the conventional sense you compete against the new models on the market as well as every used back out there.

If you make something completely different then you don't have to compete with those.

Don't get me wrong; I'm not saying you couldn't compete if you go head to head. It just seems more simple not to compete at all.

From this point of view a module built around the Mamiya 7 would seem like a winner. Many people ask for it and no solution is provided by any big player, nor seems forthcoming. It's an entirely different form factor and set of requirements/priorities than existing backs (which are all built with the assumption that there is space behind the film plane to place a thick digital back unit). It also seems a great place for CMOS medium format since being able to shoot at high ISO is a welcome addition to rangefinder applications.

If there was a kickstarter for that, and it looked well thought through, I'd fund it with personal money.
Logged

DOUG PETERSON (dep@digitaltransitions.com), Digital Transitions
Dealer for Phase One, Mamiya Leaf, Arca-Swiss, Cambo, Profoto
Office: 877.367.8537
Cell: 740.707.2183
Phase One IQ250 FAQ
torger
Sr. Member
****
Offline Offline

Posts: 1590


« Reply #13 on: February 25, 2014, 10:14:49 AM »
ReplyReply

- what you'd like to see as useful features
- what platform should be a base for the mechanics/optics
- what skills you could bring to the table, and how much time you could contribute

I'm a software guy, work professionally with it, sell software, and also contributed lots to the open source community in various projects, including raw converters. Software folks is easier to come by though than those that can do hardware.

I think the target market should be users of tech cameras (and 4x5") and legacy MF cameras (V, RZ). Supporting Hassy H and 645DF+ would be less interesting I think, those interested in those system generally can pay the price for them. Integrating with mechanical-only cameras is also easier, just a flash sync and you're done. User-changeable mount would be nice.

I don't think tether-only is acceptable, although that would be much easier to design and build. Pro studio users can do with tether only, but customers of this type of back would mostly be amateurs and they want to shoot outdoors. So there must be a CF card and some sort of display with GUI.

Sensor is the big question. If I could choose freely I think a 56x56mm ~7.2um pixel sensor would be ideal for this type of project (60 megapixels), but custom sizes would probably be hard to fund. In lack of that a standard sensor like Dalsa's 48x36mm 48 megapixel CCD would be fine for tech cams and digital lenses, but 6x6 and 6x7 MF users may think it's a bit too small. I think it would be feasible to make a digital back targeted at only tech cam users though, thanks to the huge landscape photography interest there is in the amateur space.

I don't think CMOS is ready yet for this type of back. Too small, too tech cam unfriendly.

End customer price, no more than $6000-$8000 would be nice.
Logged
JV
Sr. Member
****
Offline Offline

Posts: 688


« Reply #14 on: February 25, 2014, 10:28:25 AM »
ReplyReply

Thinking along the same lines, support formats other than 645 (6x6 and 6x7) which is currently not being done by the existing players.

Big question is obviously the availability (and price) of 56x56 sensors.

Output files should be in a standard format, none of that IIQ or 3FR nonsense.

IT guy but not very technical anymore, not sure how much I would be able to contribute.
Logged
Manoli
Sr. Member
****
Offline Offline

Posts: 685


« Reply #15 on: February 25, 2014, 10:56:22 AM »
ReplyReply

I don't think tether-only is acceptable, although that would be much easier to design and build. Pro studio users can do with tether only, but customers of this type of back would mostly be amateurs and they want to shoot outdoors. So there must be a CF card and some sort of display with GUI.

Couldn't that be solved by offering an 'out' socket so that those who want tethering (or screen size) could output to a tablet ? As a design 'principle' - I'm in favour of modularity.
Logged
torger
Sr. Member
****
Offline Offline

Posts: 1590


« Reply #16 on: February 25, 2014, 11:02:40 AM »
ReplyReply

More fantasies:

- DNG raw format, no more closed format b*llsh*t.
- Compact Flash support
- modern touch screen, more important to be strong in outdoor light and viewing angle than having one billion pixels though
- being open source, writing "apps" for the back should be possibly by anyone, like integrating a DoF app or LCC app or whatever, customizing the GUI
- USB3, to enable in the future for anyone interested in tethering, not needed in first release though (amateurs don't tether much)
- most gui effort put into making image review quick and effective. Quickly producing an image from the raw data might not bee too easy
- Real RAW histograms with true raw clip indicators for the ETTR fanatics.
- android-based OS?
- some basic weather-proofing

"Bad" things that could be accepted if necessary:
- some boot time (allows standard OS like android)
- fan (simplifies cooling challenge)
- limited ISO range (CCD)
- limited exposure time due to hot pixels (CCD, no hard limit though)
- no live view (CCD)
- somewhat battery hungry is okay
- could be ugly (ie Imacon Ixpress style)
- does not need to be class-leading in image performance, but must be competitive, especially in the DR, the 'geek' favourite property.

The availability and cost of sensor decides much about the product direction though. If 56x56 and a decent-looking ISO1600 could be feasible it would be a stronger focus on Hassy V and Mamiya RZ et al, with smaller sizes and/or worse ISO performance tech cams may be a stronger focus.

Unfortunately I don't think 56x56 is feasible, it's probably not too hard to come by a CCD in about that size but it would be some sort of astronomy or cartography sensor with strange color response and in need extreme cooling to perform well.
« Last Edit: February 25, 2014, 11:07:56 AM by torger » Logged
Rob C
Sr. Member
****
Offline Offline

Posts: 12213


« Reply #17 on: February 25, 2014, 11:43:19 AM »
ReplyReply

So, far, it seems several of us here think that enabling the old Hassy V system with a square back is a good idea. I suppose that even if the sensor has to be a very little bit smaller than 56mm x 56mm, it would still do, just as long as the viewfinders were accurate and still large enough to give good viewing. Using the best part of a lens isn't a problem just as long as there's still a good bit of it left usable and you are not effectively dropping down a format size.

Perhaps effort in the direction of square sensor manufacture is the way to go; I'm not sure that a rangefinder such as the Mamaya 67 is the way ahead for digital focussing unless you are willing to resort to using a rear screen, which pretty much makes the camera's point redundant. I can't see an eye-level camera as something I particularly want to use on a tripod. If you really want to turn an rf body like that into a form of reflex viewer via a fake EVF viewfinder instead of an optical one, that's not my idea of a good idea. I see it as an amateur-aimed decision.

Rob C
Logged

jerome_m
Sr. Member
****
Offline Offline

Posts: 656


« Reply #18 on: February 25, 2014, 12:17:44 PM »
ReplyReply

I would like to point out that there are already a few open source CCD camera projects running for astronomy cameras. They may use smaller CCD chips, but there is some expertise in these projects on how to run the necessary electronics. A good start is:
http://opencamera.wikispot.org (check the related links: although some are dead, they give the names of other projects). The original project was the Audine camera project which was quite successful amongst astronomers.

There are manufacturers of open-source astronomy cameras, for example Apogee. They have several CCD sizes available and can probably be convinced to use a bigger chip.

Astronomy cameras are tethered only.

On another directions, there are large CCDs available from some manufacturers. I know that Fairchild imaging has a 8cm x 8cm 18 mpix chip. These chips are mainly made for X-ray machines and I suppose that there are other manufacturers.

Last but not least, there is open source software for existing cameras, mainly Canon and Panasonic.
Logged
Chris Livsey
Sr. Member
****
Offline Offline

Posts: 303



« Reply #19 on: February 25, 2014, 01:29:41 PM »
ReplyReply

Another example is this project;
http://m8raw2dng.de/
This uses the hacked, just a series of button pushes, to open the Leica M8 "service" menu. this gives access to a RAW file in the raw state before any manipulation. 14 bit data instead of the default 8 bit RAW output. There is also a tool to personally re-map the sensor.
Demonstrates what can be done.
Logged

Chris Livsey

http://www.flickr.com/photos/red_eyes_man/

It's not what you look at that matters, it's what you see.
Pages: [1] 2 3 ... 8 »   Top of Page
Print
Jump to:  

Ad
Ad
Ad