DIY Photographer Builds Full-Frame Camera, Open-Sources the Project (dpreview.com) 27
Boston-based engineer and photographer Wenting Zhang built his own full-frame camera and open-sourced the project on GitLab for anyone else to build upon. The camera, named Sitina S1, features a 10MP CCD sensor, custom electronics, and a 3D-printed body. Digital Photography Review reports: Zhang says he started the project in 2017, and it's not finished yet. "Engineers are usually bad at estimating how long things will take. I am probably particularly bad at that. I expected this project to be challenging, so it would take a bit longer, like probably one year. Turned out my estimation was off," he says. He makes clear to point out that this is a hobby project, purely for fun, and that his camera isn't going to achieve the level of image quality found in commercially available products from established companies. Despite that, his project provides a fascinating look into what's involved in building a camera from the ground up.
Although CMOS has become the dominant sensor technology in consumer cameras, owing to factors like speed, lower power consumption and cost, Zhang's camera is built around a 10MP Kodak KAI-11000CM CCD sensor with a global electronic shutter, which he selected for a rather pragmatic reason: it was easy to source. "Most manufacturers (like Sony) aren't going to just sell a sensor to a random hobbyist, so I have to buy whatever is available on eBay. This 10MP CCD turned out to be available," he explains. The choice of sensor has a useful benefit. As he explains in one of his videos, designing and building a mechanical shutter is complicated and beyond his area of expertise, so his DIY design is based on using an electronic shutter. For similar reasons, he chose to use an LCD screen as a viewfinder rather than a prism-based optical design, resulting in a mirrorless camera.
Zhang wanted his design to be compatible with existing lenses. His mirrorless design, with a short flange distance, provided a great deal of flexibility to adapt different lenses to the camera, and he's currently using E-mount with active electrical contacts. And that's just the start. Zhang also needed to integrate a CCD signal processor with an ADC (analog to digital converter), a CPU, battery, an LCD screen and buttons. He also designed and built his own circuit board with a power-only USB port, flash sync terminal, power button and SD card slot, and create the software and user interface to tie it all together. Finally, everything fits inside a 3D-printed enclosure that, to my eye, looks rather attractive.
Although CMOS has become the dominant sensor technology in consumer cameras, owing to factors like speed, lower power consumption and cost, Zhang's camera is built around a 10MP Kodak KAI-11000CM CCD sensor with a global electronic shutter, which he selected for a rather pragmatic reason: it was easy to source. "Most manufacturers (like Sony) aren't going to just sell a sensor to a random hobbyist, so I have to buy whatever is available on eBay. This 10MP CCD turned out to be available," he explains. The choice of sensor has a useful benefit. As he explains in one of his videos, designing and building a mechanical shutter is complicated and beyond his area of expertise, so his DIY design is based on using an electronic shutter. For similar reasons, he chose to use an LCD screen as a viewfinder rather than a prism-based optical design, resulting in a mirrorless camera.
Zhang wanted his design to be compatible with existing lenses. His mirrorless design, with a short flange distance, provided a great deal of flexibility to adapt different lenses to the camera, and he's currently using E-mount with active electrical contacts. And that's just the start. Zhang also needed to integrate a CCD signal processor with an ADC (analog to digital converter), a CPU, battery, an LCD screen and buttons. He also designed and built his own circuit board with a power-only USB port, flash sync terminal, power button and SD card slot, and create the software and user interface to tie it all together. Finally, everything fits inside a 3D-printed enclosure that, to my eye, looks rather attractive.
Full frame camera? (Score:2)
For non-photographers, what is a full frame camera and how does it differ from a regular DSLR camera, or the camera that's built into modern phones?
And why would you want one? What does this get you that the other cameras I listed above don't have?
Re: (Score:3)
> A full frame camera captures the image in one single frame...
This definition is incorrect. Another comment here provides the correct definition and also a link to Wikipedia.
Re: (Score:1)
Re: Full frame camera? (Score:1, Insightful)
Re: (Score:1)
A full frame digital camera is equipped with a sensor that has the same dimensions of a 135 (35mm) film negative frame, while most consumer or prosumer cameras have APS-C or smaller sensors. The full frame sensors has better low-light capabilities beyond other advantages, and it means your lenses will have the correct focus length.
None of this contradicts what I said.
Re: (Score:2)
As others have mentioned, you are 100% wrong on this one. Not even close to being correct.
Re: (Score:1)
Re: (Score:1)
no one asked for your opinion either, please take your own advice and stick your foot deeper in your mouth hole.
The comment I responded to literally asked for replies:
For non-photographers, what is a full frame camera and how does it differ from a regular DSLR camera, or the camera that's built into modern phones?
And why would you want one?
What does this get you that the other cameras I listed above don't have?
You're telling me that you're so illiterate that you don't understand what a question mark means? Just holy shit.
Re: (Score:2)
"In rolling shutter mode, different lines of the array are exposed at different times as the read out 'wave' sweeps through the sensor, whereas in global shutter mode each pixel in the sensor begins and ends the exposure simultaneously, analogous to the exposure mechanism of an interline CCD."
https://andor.oxinst.com/learn... [oxinst.com]
Re: (Score:2)
Cool story, bro. [youtube.com]
Re:Full frame camera? (Score:5, Informative)
It's the sensor size [wikipedia.org]
Larger sensors, everything else being equal, tend to be better (resolution, dynamic range, light gathering capability etc). Increased die area costs though. High end dslr's tend to be full frame. Cheaper ones a smaller format with about half the area.
Cameras on phones tend to have tiny sensors because you couldn't fit the appropriate optics for a large one.. and it would dramatically increase the cost of the phone. Most image quality improvements in phones in the last decade have not been from physical improvements but from heavy reliance in computational photography [wikipedia.org] along with some AI to recognize patterns and fudge things to look nicer.
Re:Full frame camera? (Score:5, Informative)
To add to our post one of the most important factors for full frame is the relationship between focal length and depth of field. Given the larger sensor a comparatively longer focal length is required. How this manifests in pictures is the the longer focal length on the full frame camera (for an equivalent field of view) creates a narrower depth of field which is often a very pleasing result in an image. Smartphones (to your point of computational photography) go to a lot of effort to fake this by applying blur to the image, but in simple physics terms you achieve this with a larger sensor and a longer focal length for the same field of view.
Example comparison https://www.reddit.com/r/Filmm... [reddit.com]
Sensor size (Score:4, Interesting)
"Full frame" means the sensor it the same size as a standard 35mm 135 film frame (24x36mm). With a larger sensor area, individual pixels are larger, so you get lower noise. This improves performance with low light or fast shutter speeds. The main disadvantage is that you need larger lens assemblies.
Re:Sensor size (Score:5, Informative)
The main disadvantage is that you need larger lens assemblies.
Which of course also means increased weight.
Tangent - sometimes full-frame lenses that are considered good but not stellar on a full-frame camera will give great results when used with a smaller sensor (since often the lense's shortcomings manifest more towards the edges).
Re:Full frame camera? (Score:4, Informative)
The first response to your question is wrong. "A full frame camera captures the image in one single frame ..." is incorrect, as the subsequent respondents pointed out. I am adding a comment thinking you might be interested in why there is a distinction. Going from film to digital, over a roughly 20 year period 1995 - 2015, meant developing the technologies - and that was not trivial. Various low quality cameras came out, but for the top tier camera makers like Nikon, Canon, Sony, Minolta, etc., the cameras had to be as good as traditional film.
The sensors are the most unique and finicky item in a digital camera. To meet market needs, sensors had to be big enough (physical size), big enough (pixels), fast enough, and sufficiently un-noisy, etc., to meet the needs of pro photographers. I got my first digital camera in 1998, an Olympus. Quality was good, but the image was only 1024 x 768, hardly a thumbnail image by today's standards. Year by year, model by model, the specs improved, as you might expect. However, image quality and size were not good enough that pros and industry, like quality magazine publishers, were ready to accept digital images or even abandon film until after circa 2010. Here in Arizona, it was a big deal when our beloved Arizona Highways magazine allowed digital submissions, circa 2010-2012 as I recall. I got my first FX body in 2012 (Nikon D800, 36 MP).
But there is another issue. The camera makers, and Nikon is especially famous for it, want to make sure that every lens they ever made will fit and work with any body they make, old or new. For it all to work best, the digital sensor ideally should be same size as the frame on a 35mm film camera, 36x24 mm. That way, the image projected by the lens onto the film plane, and the film frame are a match, as they are engineered to be. But in the early days, when sensor chips could not be made reliably at that size, the industry settled on a smaller sensor size, 24x16 mm. Nikon called it the DX format, and it is only 44% the area of full frame, which was designated FX when it finally arrived.
If you mount a standard lens on a compatible body, regardless the sensor size, the light image on the film plane is a size that will cover an FX sensor 36x24. But if the sensor is DX, then only the center 44% of the light image is recorded, which effectively increases focal length by narrowing or zooming in on the center of what the lens actually sees, but at the expense of low pixel count and resolution compared to what you would get if a full frame sensor could be made using the same technology.
By about 2010, sensors could be made 36x24, full frame, with high pixel counts. With counts such as 6000 x 4000, and ever since on the increase, it started to be easier to talk about sensors in "megapixels" instead of h x w counts. Putting a standard lens on a full frame FX sensor allows the camera to capture the full image as projected by the lens, which is 2.25 times what a DX sensor captures. Of course, some lenses were made to be DX specific, so if you now use them on an FX sensor, the image is a circle in the center, the peripheral 60% of the frame just black. Nowadays, all of the main camera and lens makers are making only FX equipment, or at least introducing new items that are only FX, or adaptations of such for the new generation of mirrorless cameras.
Regarding the posted article, using FX or full frame for his DIY project simply indicates that he was serious about this project, not cutting corners or settling on older cheaper bargain bin technology (as best as possible, as he describes using a CCD), and allowing the camera to be interoperable with standard lenses. It was definitely a cool project.
Full Frame explanation + benefits (Score:3)
For non-photographers, what is a full frame camera and how does it differ from a regular DSLR camera, or the camera that's built into modern phones?
And why would you want one? What does this get you that the other cameras I listed above don't have?
Full frame only means 35mm size, same size as film cameras. The benefit? Your phone camera takes pictures that look like shit in comparison because a FF gets a lot more light on the sensor because it's so much larger. Phone pics have low-detail, poor color, and a lot of noise. Now I am phrasing that to be provocative, admittedly. :). However, I have an iPhone Pro 15 Max and a Canon r6 full frame. I have done the side-by-side test many times. Anyone who thinks the images are the same either is viewin
Re: (Score:2)
"Full Frame" refers to a sensor that's the size of a 35mm film frame, which is 24mm tall x 35mm wide. Most sensors for dSLRs are around what we call "APS-C" sized which is 16.7mm tall by 24 mm wide, or about 2/3rds the size. There are many reasons to want "full fr
Vignette like from the early steam age. (Score:2)
But nice nonetheless. It has a decent body for such a project which is also good. However, the photography thing itself does need work before it's viable and quality is up to modern standards. But being open source perhaps some others will pitch in and help.
Overall a neat hardware project. Extra brownie-points for trying this and seeing the first iteration through IMHO.
Re: (Score:2)
Honestly, the thing that makes me most interested in this camera is that the developer included the option to make a build for left-handed users.
I'm not terribly confident in my ability to build the custom circuit boards on my own, but hopefully someone will do a manufacturing run for nerds who haven't done anything in the EE realm since they were undergrads.
Re:I wonder why CCD vs CMOS sensor. (Score:4, Informative)
I believe some people think CCD create better images, but is that still the case?
"Better" is a word that is doing a lot of heavy lifting. There are a lot of different ways in which to describe image quality. CMOS is objectively better in some areas: it's far easier to create higher resolution sensors, it consumes lower power (which funny enough leads to an interesting point about noise, CCD is usually cited as the lower-noise option all else being equal, but localised sensor heating raises the noise of the sensor meaning CMOS actually may produce the lower noise). There's also some very interesting issues with reading out CCDs, get it wrong and you end up with blooming and smearing effects - something which astronomers know well since astronomy specific cameras often give the user control over how to read out the sensor.
That said speaking of astronomy, CCDs are "better" for that as they are more linear sensing devices.
But back to cameras most people aren't interested in pure photon counting. They are interested in reading out 120 frames per second, or firing the shutter at 1/8000th of a second. When you design something to read out CCDs quickly virtually all their benefits disappear so you're left with CMOS being the dominant technology for virtually all general purpose photography / video cameras on the market.
That said CCDs have a special place in the scientific world, and that does make them far more accessible at low volumes. - I suspect this may have had to do with why a CCD was used, in TFS it is mentioned that he was fundamentally limited by what was available.
Re: (Score:2)
CCDs have a few benefits over CMOS. The first one is important when recording motion pictyures - global shutter. A CCD will have a global shutter because all those charges need to be moved to a "dark zone" on the CCD so they can be read out. CCDs are read out serially one pixel at a time, shifting pixels down the line to be read. This requires a global shutter.
CMOS sensors can be read out cell by cell, so they do not need a global shutter - basically at readout they transfer the reading to the data line an
Re: (Score:2)
Your views are very outdated. Firstly CCDs don't have a global shutter. That process is left up how the readout method works, and while the earliest examples worked as you described, there are various implementations. Reading out the CCD the way you described is not a very high performance approach so many sensors are not used that way anymore. Speaking of various implementations, CMOS sensors have had global shutter abilities for over a decade at this point. RED - well known cinema brand - has multiple CM
A wonderful project (Score:2)
It would be great if more people would contribute Linux style to this open source project, it could become a really slick camera with potentially innovative features.