Bolex D16: RAW, 16/Super 16 sensor size, costs less than MkIII

Razz16mm said:
The sensor is not outdated. Kodak CCD's are the first choice in many scientific imaging applications where maximum quality data acquisition and refined images are essential, like the Mars Rovers for instance.
Click to expand...

Do you have a source to back up your statement? If this was a case NASA would not buy a bunch of Canon XF300/305 for the space program (these are the cameras used on Space Station). If CCD was used is most likely due to the fact that CCD are less prone to have dead pixels due to radiation, and integrations with existing remote systems. The second would back up the statement about CCD being outdated technology.
 
Yes, that's what the rover is using, but to me that doesn't mean much today: it means it was great tech a long time ago (designing a rover takes years, and you don't go changing sensors when the project is almost finished, you stick to your original choice and let it be)
 
snowblindfilm said:
Do you have a source to back up your statement? If this was a case NASA would not buy a bunch of Canon XF300/305 for the space program (these are the cameras used on Space Station). If CCD was used is most likely due to the fact that CCD are less prone to have dead pixels due to radiation, and integrations with existing remote systems. The second would back up the statement about CCD being outdated technology.
Click to expand...
They are preferred for critical imaging applications for three major reasons, higher fill factor, global shutter, and more accurate color rendition, especially in the reds. But they are four times the price of an equivalent CMOS sensor to build a camera around due to the extra needed off chip circuitry and added labor hours for design and calibration. They are still widely used in scientific and industrial applications where a time coherent image capture is essential, for motion analysis, or in astronomy for long exposures. Top of the line broadcast cameras are still 3-CCD models and that is typically all you see used for major live sports broadcast production.
CMOS has taken over the consumer camera market because they are cheap and require much less support circuitry in the camera since the analog gain and AD conversion is on chip. But you pay the price in rolling shutter artifacts, greater tendency for aliasing artifacts due to lower fill factors caused by more dead space between pixels for the added circuitry, and generally poorer color rendition.
CMOS has advantages in potentially higher frame rates. With most recent generations they have a stop or two more sensitivity and better average signal to noise ratios, though this is offset significantly by gain variations between individual pixels and typically higher defective pixel counts. Just because they happen to be popular doesn't make alternative technologies obsolete or inferior.
 
Samuel H said:
Yes, that's what the rover is using, but to me that doesn't mean much today: it means it was great tech a long time ago (designing a rover takes years, and you don't go changing sensors when the project is almost finished, you stick to your original choice and let it be)
Click to expand...
Bingo! For such projects you stick with a proven technology. Broadcast moves with a speed of a glacier, they just switched to digital a few years ago. If CCD was the way to go I bet Phantom, Arri, Red would have slapped a CCD into their cameras. The superiority of CCD over CMOS sensor is on the same level as people keep repeating US has "the best health care system". Just because people say something, doesn't make it so.
 
snowblindfilm - your opinion about health care aside, I bet Arri and Red have real good reasons for why they don't use CCD but that doesn't devalue anyone else's decision to use it. It's not like deciding to use a CCD means CMOS isn't any good, nor does it mean that Arri and Red choosing CMOS doesn't mean CCD's aren't any good.

Each chip has it's own strengths and challenges, just like any tool used in filmmaking.
 
snowblindfilm said:
Bingo! For such projects you stick with a proven technology. Broadcast moves with a speed of a glacier, they just switched to digital a few years ago. If CCD was the way to go I bet Phantom, Arri, Red would have slapped a CCD into their cameras. The superiority of CCD over CMOS sensor is on the same level as people keep repeating US has "the best health care system". Just because people say something, doesn't make it so.
Click to expand...

Yeah I suppose the Leica M9 is only an average camera ?
 
The Leica only shoots stills though. Each sensor will have different capabilities for given applications. I would suspect the majority of "boutique" cinema cams have gone CMOS for the faster readout/dr/cost per unit advantages of the sensors available to them of a given size (S35, S16)...not theoretical reasons.
 
The primary reason that CMOS is chosen over CCD these days is that they are cheaper and easier to develop cameras for. This certainly applies to just about any camera under $10k that is sold through retail distribution.
But any serious imaging application that requires a progressive frame that records all of the data for light that impacted the sensor during the frame interval still uses global shutter CCD's.
They are the only technology that actually captures a full frame image that is temporally coherent in the same way film is.
The trade off for CCD's vs CMOS is generally less sensitivity and lower maximum frame rates. But motion quality on CCD's is far superior to CMOS.

global-shutter.png
 
No, based on your $10,000 platform and doing Apples to Apples in the world in which we are speaking (not UV, infared, or medical applications which are irrelevant to cinema): "serious" imaging applications such as cinema cams, Red has chosen CMOS for their line, Arri has chosen CMOS for their line, Sony has chosen CMOS for the F5 and F55 AND F65, Canon has chosen CMOS for the C300 and C500. Phantom has chosen CMOS. ALL over $10,000.

Who does that leave using CCD for newly developed S16 and above cinema cameras that cost over $10,000? Not a whole lot.
 
Last edited:
pulpfiction007 said:
No, based on your platform: "serious" imaging applications such as cinema cams, Red has chosen CMOS for their line, Arri has chosen CMOS for their line, Sony has chosen CMOS for the F5 and F55 AND F65, Canon has chosen CMOS for the C300 and C500. Phantom has chosen CMOS. ALL over $10,000.
.
Who does that leave using CCD for newly developed S16 and above cinema cameras that cost over $10,000? Not a whole lot
Click to expand...

When I say serious, I mean scientific and industrial applications where the image data of each frame is critical, like the Mars rovers or monitoring manufacturing processes. Entertainment video of any kind doesn't qualify.
RED just released a new lens mount with the Tessive time filter technology built in for Epic and Scarlet which makes them global shutter cameras. Arri does the same with the mechanical mirror reflex shutter for the top of the line Alexa studio. Sony's top line 3-chip 2/3' broadcast cameras are all still CCD.
Phantom uses CMOS sensors because that is the only way to currently get extremely high frame rates. CCD's don't do high frame rates.

CMOSIS is currently making new CMOS sensors with global shutter capability, but there is a 2 stop loss of DR and reduced maximum frame rate compared to programming the same sensor for rolling shutter. CMOS is just a common manufacturing technique used for many types of chips, not just sensors.

Sure CMOS cameras are capable of making decent high quality video within their limits, no one is saying they aren't. But they are not inherently superior in every respect just because they are more popular with mass manufacturers. Price, especially in the under $5k consumer markets, plays a major part in the choice.
Given that the cost of manufacturing for a mass market retail electronic device of any kind is typically well under 1/5th the selling price, the cost for the raw KAI4050 scientific engineering grade CCD being used for the D16 is very likely higher than the total manufacturing cost for the new BMC pocket cam.

Joe has chosen to develop a true S16mm 2k format/16mm 1080P format high image quality digital camera for a small but appreciative market of 16mm film aficionados that is not currently being served by any mass market manufacturer. He deliberately chose the highest possible quality off the shelf components he can get for the intended price range. The design emphasis is on things that really directly impact image quality, sound quality, operational simplicity and rugged reliability over fundamentally unnecessary but often popular features in other mass market cameras. The only other camera that is comparable is the Ikonoskop D2, which uses the Kodak KAI4020 1080p CCD sensor. This line of sensors was specifically developed to closely match the spatial resolution, image integrity, and color rendition of Kodak's own color negative film stocks with comparable sensitivity and usable DR.
Neither the D16 nor the D2 will be obsolete for their intended market for many years to come. They are both a refreshing breeze of cool clear mountain air in a prosumer market dominated by rolling shutter jello, aliasing and moire polluted CMOS smog.
 
Last edited:
David W. Jones said:
OK, over a year and 124 pages later... What's the status of said camera?

From the outside looking in, the camera the Bolex is being compared to in the thread title can now shoot raw, and the Bolex is ?
Click to expand...

They are still doing sensor calibration and firmware development. They have two operational manufacturing prototypes with final components.
 
I've been reading this thread for ages, and more so I've been reading Razz's posts for a while. And my opinion changes on them as I grow up. Razz, sorry, no offense taken, this is all one big big-up.

I understand this camera isn't out. I understand it's been years, and other cameras have come out in the meantime.

I do agree with Razz about the 16mm digital film market. It's something I would've always jumped into if I could afford it. And years ago, just as the acamdii was my dream camera, it still is today. It's what I want out of it.

Still, my reason for this post is that I realized something reading razz's post #1231.

The reason we probably get all this speculation wrong is that we don't see the process past our own perspective. I was way more like this, and as a result, for a long time, I couldn't put myself in others' shoes, I couldn't make my films to be legible to anyone who didn't know the story, would be angry when ppl didn't notice small technical achievements, I would even emphasize moves in videos that shouldn't show rather be felt...

... and I could never understand Razz's posts on this topic. I thought he was blatantly defending a camera he liked and wanted.

But reading about all the factors that really never cross our minds when it comes to the sensor for example, specifically this line...

'When I say serious, I mean scientific and industrial applications where the image data of each frame is critical, like the Mars rovers or monitoring manufacturing processes. Entertainment video of any kind doesn't qualify.'

... I kinda saw the whole thing for a second. And that's true. When I look up imaging, entertainment content just isn't there. I see medical this, space that, surveillance this.

And then I think, yes, I still would take the a-cam over anything else out there. Even the bmc. It's what I want. I want a CCD. I don't care for 4k.

But then I think of everyone who watches the movies. They have no clue what we're doing. And that's what it's about. It's an incredibly wonderful ego-smashing field to be in if you go that far.

It is linking directly to how we analyze the small fraction of the factors that go into making something to sell.

I think if we can all use this thread, and Razz's posts to just realize and come to terms, peacefully and confidently that there's so much out there that we have no clue about that goes into making a camera, and that we pick what we need coz we can, if we can afford it...

... then i think we'll make films differently. I think it'd be easier to tell story to someone who doesn't know the story, or know the author personally. I think our films can work better.

Does anyone see a link between 'Why won't they make cameras like x,y,z, the parts only cost x,y,z, and we're the ones that are using them exclusively' and 'why does no one get the story of this film? I know it, why don't they?'

I dunno, bit of a nice realization at Razz's post.

Thanks man.
 
Personally, I am rooting for the D-Bolex people. Joe is a nice guy/smart businessman and I think the Bolex could be an awesome documentary camera (where the BMC falls short) I also love the idea of a 'hand made" camera in an era of giant faceless companies pumping out mass produced pieces of plastic... BUT they do need to deliver the camera soon. Patience is running thin and watching their ustream the other day did not exactly instill confidence. Pouring over the minutia of the cameras technical development may interest some but I personally just want the camera in my hands. Also, I understand that Joe and Elle are trying to undertake a democratic approach to developing the camera but I think at this point they just need to make some executive decisions and move on. Just my two cents...
 
scorsesefan said:
in an era of giant faceless companies pumping out mass produced pieces of plastic...
Click to expand...

well my fellow queensman, while i'm no fanboy of large corporations, i would hardly state them all as that! We have the cameras we have today because of large corporations who dump bucketloads of cash into developing them into incredibly intricate machines that have remarkably low failure rates.
 
scorsesefan said:
Personally, I am rooting for the D-Bolex people. Joe is a nice guy/smart businessman and I think the Bolex could be an awesome documentary camera (where the BMC falls short) I also love the idea of a 'hand made" camera in an era of giant faceless companies pumping out mass produced pieces of plastic... BUT they do need to deliver the camera soon. Patience is running thin and watching their ustream the other day did not exactly instill confidence. Pouring over the minutia of the cameras technical development may interest some but I personally just want the camera in my hands. Also, I understand that Joe and Elle are trying to undertake a democratic approach to developing the camera but I think at this point they just need to make some executive decisions and move on. Just my two cents...
Click to expand...

The basic camera is locked for production, they have all the parts to build the first 100 for the Kickstarter backers. But there is no way to short cut sensor calibration and color science. It takes the time it takes until you see the result you want. Their goal is to ship those in August sometime at this point. Even though that is a year past their first anticipated release date, the camera is much better for the extended development time and that would still have to be a record of some sorts for rapid development of a completely new camera from the ground up. How long did it take Canon to do a minor update to the 5DII? Most cameras go through a 3 to 5 year development cycle. We are just not privileged to watch or participate in the process.
 
You must log in or register to reply here.
Back
Top