Digital Bolex weeks away

Thanks.

Global shutter looks great, but as Bloom points out, more a personal camera than a production one.

Would not disagree with that assessment. There are many cameras that are easier and more forgiving to shoot for video. Perhaps more efficient for fast pro video production work flows.

Though the D16 can be very efficient in the field if you shoot it as you would film instead of video. Here is a link to a short fun piece that Indy Skint did last weekend before heading out for a 9 week feature gig. 44 setups in five hours from his comments.


And a link to a review and personal documentary by Curt Lancaster:

http://kurtlancaster.com/05/03/2014/digital-bolex-d16-review-with-a-celebration-of-clays-life/
 
Last edited:
Even Bloom confirms, what i was saying all along. LCD is worthless and CCD causes horrible light streak, probably the worst light streak I have ever seen. No removable battery and media is also a huge drawback. All were very predictable problems and looking at the feedback easy to avoid in the design phase. Too many cons for the price point.
 
Interesting review. My sense is overall positive review, but with some caveats. Bloom is very good at articulating what this camera is and is not ("NOT a low light camera!" :)
His review pretty much aligns with my experience with the D16, which is a camera that produces some great filmic images (with a pleasing grain structure) that may fall short as a full production camera but shines as a boutique camera that is a joy to shoot with. It's a shame that some people will dismiss the D16 out of hand because you really need to pick it up and shoot with it to get a sense of how ergonomic and classic its design is. BTW some of the issues that Bloom mentions with the sensor have been addressed with an update that came out after his review (pink highlights for instance).

And I own a BMCC and I love it and will stick with it but whenever I can I will steal by friends D16 to just have some fun shooting again.
 
Interesting review. My sense is overall positive review, but with some caveats. Bloom is very good at articulating what this camera is and is not ("NOT a low light camera!" :)
His review pretty much aligns with my experience with the D16, which is a camera that produces some great filmic images (with a pleasing grain structure) that may fall short as a full production camera but shines as a boutique camera that is a joy to shoot with. It's a shame that some people will dismiss the D16 out of hand because you really need to pick it up and shoot with it to get a sense of how ergonomic and classic its design is. BTW some of the issues that Bloom mentions with the sensor have been addressed with an update that came out after his review (pink highlights for instance).

And I own a BMCC and I love it and will stick with it but whenever I can I will steal by friends D16 to just have some fun shooting again.

"Boutique camera"...perfect description. I can think of many promos (network), commercials, POV etc where this camera can find a place in a professional environment. Maybe not as an A cam (I agree to a degree with Bloom) but when you need a certain look s16 etc, there's no other camera on the market. And yes, it's the perfect camera for film students, indes and even pros who need a certain look. Cameras are tools, The d16 captures different than any new cam out right now, that alone has to account for something.
 
Even Bloom confirms, what i was saying all along. LCD is worthless and CCD causes horrible light streak, probably the worst light streak I have ever seen. No removable battery and media is also a huge drawback. All were very predictable problems and looking at the feedback easy to avoid in the design phase. Too many cons for the price point.

The LCD provides monitoring as a convenience. It was never intended to be your primary monitor, mainly for menus. We discussed this in the development stage. Joe's conclusion was you could buy any one of a number of aftermarket EVF's or monitors that would do the job better and cheaper than it would cost them to develop a good one. He was trying to hold target price and we all preferred adding things like really good audio preamps and better external monitor output support. Realistically how many people never use external monitors of some kind anyway, either because you need something bigger or the on camera versions are lacking in one way or another?

IMO the benefits of the CCD outweigh its shortcomings. No aliasing or moire issues, color rich data to work with, no rolling shutter mush. The interline CCD has the means built in to control streaking better, that will improve as the firmware evolves.
Most top line 3-chip broadcast cameras are still CCD for sound image quality reasons. For time coherent image integrity they still outperform CMOS sensors.
 
Last edited:
The bottom line, as with any camera, is that the D Bolex is not for everyone. Some will appreciate the raw and the IQ in with good lighting, others may opt for a camera with better abilities in available light and/or a different work flow. Personally, I love the images from the D16...but all the other things add up to not being right for me.

All I'd really want is a camera that has a largish sensor that has decent functionality and shoots a nice codec and 4:2:2 to cards. Surprisingly in this day and age of affordable big sensors, raw and 4K...no one has made the camera I'm talking about in the $5,000 price point.
 
IMO the benefits of the CCD outweigh its shortcomings. No aliasing or moire issues,utperform CMOS sensors.
??? Aliasing (and attendant moire issues) have nothing whatsoever to do with CCD vs CMOS. A CCD with an improper (or worse, no) OLPF
Will alias exactly the same as a CMOS would.
 
I think Joe and Elle did a fantastic job, given their inexperience and low budget. I would be impressed even if such a camera came from one of the giants, especially for such a low price.

Joe did a great job hewing close to his passion and vision of a modern-day Bolex.
 
Most top line 3-chip broadcast cameras are still CCD for sound image quality reasons. For time coherent image integrity they still outperform CMOS sensors.
Come on! That's just not true, is it? And surely large sensor broadcast single chip are a better comparison anyway - both being more comparable to the (single chip) Digital Bolex and where most of the recent R&D money and attention has gone anyway.

Or are you saying that ALL out of Canon (C300/500), Sony (F5/55), Panasonic (4K s35 Varicam), Red, have ALL got it wrong by moving to CMOS for such!?

Is the truth not simply that Digital Bolex had to make do with buying a pre-existing, off the shelf, no longer state of the art sensor - which after another two years development have gone by is looking even more long in the tooth?

And as Barry accurately says, aliasing performance is primarily a function of photosite count - totally independent of whether it's CCD or CMOS. It CAN be modified with an OLPF - but inevitably with an effect on sharpness. The only way round that is to make the sensor resolution greater than system resolution. Start with a 1920x1080 sensor, and don't expect much better than 720 unaliased resolution. Which may well be good enough for some users - but please don't make this camera out to be something far better than it is.

Digital Bolex still don't seem to have produced any proper test charts, but it's interesting to compare the stills you earlier posted - the true frame grab and the downconverted 1024x576 you earlier posted. You need to scale up the latter to compare properly, but with the same canvas sizes, the frame grab is nowhere close to showing the improvement I'd expect. Consequently, I'm pretty convinced that my original estimates were close - the Digital Bolex is only capable of around 650-700lpph. OK, not bad as it's a fairly cheap camera, but nowhere near the 1000+ figure a good 1080 camera should give, before we even start to think about 4K camera performance.......
 
I thought that DB chose that Kodak (now Truesense) sensor because of the look. I am no expert but there's an opinion that CDDs have more of a organic film look then CMOS. With what I have seen from the DB so far I agree but am not sure if it has more to due with Grading vs sensor.

I remember
aaton was going to make a CCD cine camera, I think it did get made but not sure what happened to it after aaton closed its doors.
 
Come on! That's just not true, is it? And surely large sensor broadcast single chip are a better comparison anyway - both being more comparable to the (single chip) Digital Bolex and where most of the recent R&D money and attention has gone anyway.

Or are you saying that ALL out of Canon (C300/500), Sony (F5/55), Panasonic (4K s35 Varicam), Red, have ALL got it wrong by moving to CMOS for such!?

Is the truth not simply that Digital Bolex had to make do with buying a pre-existing, off the shelf, no longer state of the art sensor - which after another two years development have gone by is looking even more long in the tooth?

And as Barry accurately says, aliasing performance is primarily a function of photosite count - totally independent of whether it's CCD or CMOS. It CAN be modified with an OLPF - but inevitably with an effect on sharpness. The only way round that is to make the sensor resolution greater than system resolution. Start with a 1920x1080 sensor, and don't expect much better than 720 unaliased resolution. Which may well be good enough for some users - but please don't make this camera out to be something far better than it is.

Digital Bolex still don't seem to have produced any proper test charts, but it's interesting to compare the stills you earlier posted - the true frame grab and the downconverted 1024x576 you earlier posted. You need to scale up the latter to compare properly, but with the same canvas sizes, the frame grab is nowhere close to showing the improvement I'd expect. Consequently, I'm pretty convinced that my original estimates were close - the Digital Bolex is only capable of around 650-700lpph. OK, not bad as it's a fairly cheap camera, but nowhere near the 1000+ figure a good 1080 camera should give, before we even start to think about 4K camera performance.......

Well DB could have used a much cheaper off the shelf CMOS sensor like BMD did if they chose too. There are plenty of those out there. The Truesense CCD in the Bolex cost more at wholesale than the entire production cost of a BMPCC.

http://www.panasonic.com/business/provideo/P2HD-VariCam-cinematography-camcorder.asp

http://pro.sony.com/bbsc/ssr/cat-broadcastcameras/cat-hdstudio/product-HDC3300R/

http://pro.sony.com/bbsc/ssr/cat-broadcastcameras/cat-hdstudio/product-HDC1700L/

http://pro.sony.com/bbsc/ssr/cat-camerascineproduction/cat-cineprodcinealta/product-F23/

The above CCD cameras are still the bread and butter work horse cameras of most of high end broadcast production, especially live sports.

CCD's have their limitations, frame rate at high resolution being one of them which is why 4k motion cameras are CMOS. But they are still the only sensor tech that produces a completely time coherent frame like a film camera and do so with a much higher fill factor for their surface area due being analog devices without on chip processing circuitry that occupies in many cases up to half the surface area of the chip. They are not used in low end cameras for the most part due to their high cost, not only for the sensor but for the associated camera circuitry needed to support them and the added camera development time required. They are still the first line choice for scientific/industrial imaging applications where image integrity is critical, like the Mars Rovers. I am certainly not arguing that the D16 is going to match the pixel resolution of $15k and up production cameras shooting low compression codecs. But for cameras in the D16's price range it will hold up well in that respect against any of them for clean artifact free resolution and it is superior to virtually all of them in its ability to reproduce fine tonal scales and fine color discrimination when properly exposed thanks to the unique quality of its Bayer filters.

Aliasing performance is governed by pixel spacing, fill factor, and the absence or presence of a properly matched OLPF. Down scaling from high resolution over sampling certainly helps as do large high pixel resolution sensors. But the camera is just half the equation. Here is a link to a high resolution PDF of an ISO resolution chart.

In Adobe reader you will need to size it up to 120% to approximate correct image size on a 1080p display. What you will see is that for any 1080p LCD technology display you get maximum clean resolution without aliasing or other reproduction artifacts in the 600-700 line range. DLP projectors do significantly better, but 800-850 lpph is the practical upper resolution limit of the 1080p format regardless of source resolution.

https://www.dropbox.com/s/jdrjf2a9c69kmv4/ISO_12233-reschart.pdf
 
Thanks.

Global shutter looks great, but as Bloom points out, more a personal camera than a production one.

It was a great review. Bloom basically said, is this a great production cam? No. Do I still want one despite that because it's fun to use? Yes.
I think it's a camera for someone that is just interested in having fun creating pretty images, not comparing camera specs or competing for the next industrial or commercial. I think there is a market for that and you can't underestimate the "fun" factor. There's a lot of hobbyists out there and plenty of hobbies use equipment that cost a lot more. Not everyone is shooting to put food on the table or deliver broadcast images that are on a deadline and up to spec.

What I'm describing is someone like Andy Warhol and the quirky films he made back in the day. Can you imagine Andy Warhol reading camera specs, and caring about 4:2:2 color space and bit rates? What Warhol did with a 16mm camera then, he could do today with the Bolex. There's nothing complicated going on here. It just needs to be organic.

 
Last edited:
The above CCD cameras are still the bread and butter work horse cameras of most of high end broadcast production, especially live sports.
Not a single 2/3" camera that was released within last 12-16 months has CCD in it, all of them have went with CMOS. Also your argument about NASA using CCD's- actually they purchased several XF300 and 305, again a CMOS sensor cameras. Quit spreading FUD.
 
"What I'm describing is someone like Andy Warhol and the quirky films he made back in the day. Can you imagine Andy Warhol reading camera specs, and caring about 4:2:2 color space and bit rates? What Warhol did with a 16mm camera then, he could do today with the Bolex. There's nothing complicated going on here. It just needs to be organic."
Ha ha ha great point! Can't imagine Warhol pixel peeping when he's shooting "Empire" or "Blo*job". Of course he would have left it to Paul Morrissey to do the "techie" stuff :)... I think Warhol would have dug the D16...
 
It was a great review. Bloom basically said, is this a great production cam? No. Do I still want one despite that because it's fun to use? Yes.
I think it's a camera for someone that is just interested in having fun creating pretty images, not comparing camera specs or competing for the next industrial or commercial. I think there is a market for that and you can't underestimate the "fun" factor. There's a lot of hobbyists out there and plenty of hobbies use equipment that cost a lot more. Not everyone is shooting to put food on the table or deliver broadcast images that are on a deadline and up to spec.

What I'm describing is someone like Andy Warhol and the quirky films he made back in the day. Can you imagine Andy Warhol reading camera specs, and caring about 4:2:2 color space and bit rates? What Warhol did with a 16mm camera then, he could do today with the Bolex. There's nothing complicated going on here. It just needs to be organic.


Excellent points. To put cost in perspective, my current GL2 CCD camera package was about $2800 with accessories new 8 years ago now. Lovely color, good lens, surprisingly good DR if exposed properly, not that far off what you get out of current similarly priced AVC cameras, effectively a 100 ISO camera for best image quality. Still works like a champ after many hundreds of hours and paid for itself with one industrial doc.

HD resolution is nice, but all the compressed codec low end cameras in this general price range still have similar frustrating issues with video IQ that the GL2 does, which matters not a bit to most. Indeed in some ways they are worse. All of the layers of automated consumer functions that have made them seem simpler to use for those not skilled in the fundamentals of photography get in the way of full manual control of the most basic functions that get one the best performance. This goes for digital still cameras as well as video cameras.

To get into a high quality production video camera that can produce images with similar qualities to what I am seeing from the D16's raw output means at least an FS700 or F3 class camera with an external recorder. Forget what people post on the web from any of these cameras, it tells you nothing of significance about what the camera is ultimately capable of.
Uncompressed raw and the Kodak CCD are not the easiest things to deal with, but that is part of the challenge and the fun. PB repeatedly said the D16 is not a low light camera, but he also repeatedly said it produces lovely organic images too and he shot quite a few with it himself. So you need to use a lens shade or a matte box and don't point it at the sun. With the other cameras you don't try to shoot sharp deep focus shots pointed at anything with a fine repeating pattern from clothing to brick walls to rippling water.

The BMC cameras do have nice DR and decent color, but in most aspects they are cheap and crude. Raw or not they still have the same IQ issues with rolling shutter, aliasing and moire that have made AVC cameras not so interesting to me.
 
I disagree with saying you'd need an FS700. I've seen nothing from the D16 that surpasses the best C100/AF100 footage. Each has their strengths and weaknesses, but when properly shot they all look very good and they can all look very bad in the wrong situation as well...it just comes down to which compromise one wants to make (in this price range)
 
Back
Top