Digital Bolex
Tuesday, March 13, 2012 at 1:47PM My inbox this morning was crammed with mentions of a new Kickstarter campaign for a 2K digital cinema camera called, with the blessing of the nominal company, the Digital Bolex. Philip Bloom has done a terrific job of summing up the camera’s specs and raison d’être in a blog post provocatively titled The Digital Bolex D16. Raw 2K for less than a cost of a 5Dmk3?. He’s even got an audio interview with the creators, Joe Rubinstein and Elle Schneider.
In the time it’s taken me to retrieve coffee and fight my way through the rain to my desk, the campaign has funded—but there are still open pledge spots at the $2,500 mark, which earns you a first-run D16 for $800 less than the planned retail price. Less than 24 hours after launching their campaign, the Digital Bolex is a success. Hooray! Right?
I hope so. But color me… skeptimistic.
Like Philip, I also have fond memories of shooting 16mm film with a Bolex. And I can see where Elle and Joe are coming from with their design philosophy:
There is no camera on the market that offers affordable RAW quality to consumers and independent filmmakers. The Digital Bolex will mean filmmakers who prefer an uncompressed and “film like” look won’t have to spend hundreds of thousands of dollars to achieve that. Isn’t it time for the digital generation to have image quality as good as our parents had?
Aren’t these two just adorkable? I desperately want them to succeed.
But after watching the well-financed and well-intentioned Red Digital Cinema company struggle to deliver on their promises, and abandon the “3K for $3K” design/price-point of their Scarlet camera, one has to wonder if these flmmakers-turned-cameramakers have any idea of the challenges they’re about to face.
I wish no ill on anyone. I really want these guys to win. But there are a few things that give me pause as I consider whether to send these kids my heard-earned money.
- “Raw” is just a word, it’s not an acronym. So you don’t put it in all caps. Don’t you want to buy a camera made by someone who knows that? No? Just me?
- The spec sheet lists the color depth of the files as “12 bit — 4:4:4”. But the accepted understanding of a bayer-pattern sensor is that it does not deliver 4:4:4 color at its native resolution.
- That’s important. Take a nice, sharp, raw photo with your DSLR, and process it with your favorite software, such as the awesome new Lightroom 4. Now crop a 1:1 1920x1080 window out of the middle of it. Does the image quality thrill you? There’s more to Red’s 4K and 5K specs than simple pixel-count boasting and dreams of 4K presentation. Oversampling is a good thing. The 3K scarlet was as low as Red was willing to go, I imagine in part because 3K bayer downsamples to a nice, clean 1920x1080. The Canon C300 uses a 4K sensor to provide full 4:4:4 color sampling at HD resolution, even if it offers no way to get that color sampling fidelity out of the casing.
- Raw is big. The D16 is said to shoot uncompressed Cinema DNG frames of 2–3 MB each (which seems small to me) to dual CF cards. If each of those is 32 GB, that’s about 18 minutes of recording time per reload, on $130 worth of media at today’s prices.
- Raw requires transcoding. When you get home with your CF cards full of footage, you’ll be processing it with some as-yet-unknown software for a good long time before you can work with it, or even review it. I suppose you could think of this as reminiscent of sending your 16mm rolls to the lab. Except you’re the lab.
- All this adds up to a huge data footprint for images that might not be as mythologically “film like” as one would hope. For me, personally—and you are welcome to disagree, as this is a subjective as it gets—uncompressed 2K bayer is precisely the “sour spot” of digital cinema; a data-heavy, workflow-intensive image that won’t survive much pixel-peeping.
- Monitoring and focus could be an issue. The cute little viewfinder is planned to be only 320x240. Video out is also SD. HD-SDI is said to be offered “in [a] separate unit.”
- We’ve had a fun ride together over the past few years working out for ourselves what a “film like” look means to us and to an audience. Is it uncompressed 2K frames? Or is it soft, rolled-off highlights? Maybe it’s 35mm depth of field characteristics. Or maybe it’s just where you put the camera.
- Or maybe part of the new “film look” is “Look, I’m editing my film on my MacBook Air on the ride home from the shoot, because I shot to a mildly-compressed codec that’s compatible with my editing software.”
- But don’t discount 16mm sensor size. Two years ago, a film shot on Super 16 won the Best Picture Oscar—and was nominated for Cinematography, where it lost to a movie shot on 2/3” digital cameras.
- This camera almost already exists in the form of the A-Cam dll from Ikonoskop, a Swedish 16mm camera manufacturer. You can order one now for €7,700 ($10,000 US). It’s interesting that the Digital Bolex team, in celebrating the legacy of one European 16mm camera maker, is effectively claiming the ability to beat another at the digital game.
I Worry Because I Care
I’ve funded a few Kickstarter campaigns, and it’s never lost on me that It’s an odd (but kind of exciting) thing to send a stranger your money in the hopes that they’ll send you a thing sometime in the future. Joe and Elle remind me of another breakout Kickstarter project, the PID-Controlled Espresso Machine. Two adorable, scruffy guys promise something that seems too good to be true—a commercial-quality, temperature- and pressure-regulated espresso machine for a $200 pledge.
For comparison, my dual-boiler, temperature-controlled espresso machine costs $2,000 today. I recently had the PID unit replaced, and the part cost alone was over $300.
These guys were hoping for $20,000. They funded successfully—at $369,569.
They win! Right?
Well another way of looking at that is they have $370,000 of other people’s money, and less than a year to deliver 1,300 commercial-quality espresso machines to eager backers, at a price-point that no one in the coffee industry has ever touched. Have they promised the impossible?
What do they do when they hit a snag? When something costs more than they thought? When life gets in the way? When they realize they just volunteered to work their asses off for a year for a combined salary of whatever profit they can scrape from that investment that thousands of people have made?
If responsible Italians (or in my case, Spaniards) need to charge upwards of $1,500 for an burly espresso machine, and trusty Swedes need to charge $10,000 for a digital 16mm camera, one has to ask: Do scruffy hipsters know something that responsible Italians and trusty Swedes do not?
What a Grumpy Gus I am (did I mention it’s raining?). Why can’t I just be happy for these guys? Look at their adorable glasses! Every industry needs a shake-up now and then, someone to come in with a fresh perspective and no entrenched interests. Remember “Here’s to the crazy ones?”
I worry because I care. I want it to work out, I really do. Count me in for a dollar.
Check out the Digital Bolex Kickstarter page. Follow @digitalbolex on Twitter. And don’t miss the follow-up.
Stu
A few updates:
- Folks have gently pointed out that Joe has a professional background in sensors and software.
- The Digital Bolex FAQ lists the frame size as 3.5 MB, which is more in keeping with sample frames from the 1920x1080, 12-bit Ikonoskop camera. So that’s 13 minutes on two 32 GB cards.
- Elle is a sport.
- Swedes.
Reader Comments (26)
I think you've hit the nail squarely on the head.
Phillip Bloom uses all caps as well.
I understand the healthy skepticism, and count me among those not dying for 1080 from a 2/3" Bayer.
But I wanted to point out that one reason people are able to produce things so inexpensively on Kickstarter is the near absence of overhead. The Swedes need to charge $10,000 because they have offices, employees, marketing, taxes (careful, Kickstarters—you do have to pay taxes!), etc.
It's true that entrepreneurs often overreach, underestimate difficulty, misunderstand the market, or end up eating ramen and working 20 hours a day. But if they didn't, we wouldn't have laptops or cameras. And it's all the more reason to back them early and get the espresso machine cheap (I sure did!) before they realize they need to be charging $1500!
I say this as someone who Kickstarted a manufacturing project and got in over his head. But it really did jumpstart us, and now we're working on bigger and better products... Step by inept, expensive step. Not every successfully funded Kickstarter will be successful as a business, but those that are couldn't have done it any other way.
Thanks for that wonderful comment Ben!
Hi Stu,
That is a thoughtful blog about the D-Bolex. Thank you for putting forth the effort.
In short order you addressed most every concern I have about the project- or any such camera creation project. Thank you for also throwing in the comments about Kickstarter in general. Again, I think a lot of people get caught up in excitement and perhaps aren't thinking of all the serious technological, financial and temporal hoops that go into product development.
That's not in any way being a Grumpy Gus, it's just being a realist.
Building a digital cinema camera is not for the faint of heart. My hat's off to anyone who succsessfully pulls it off.
I.
Illya Friedman
President
Hot Rod Cameras
www.hotrodcameras.com
I don't think your post was too grumpy. Even though I sent this to your inbox and figured a bunch of others probably did, I'm only in for $5 as they are funded and I'm both skeptical and cheering them to success.
Even though I'm not sure I even want one, a success for them will certainly change the marketplace and price points and everybody wins there. So let's hope they hit the home run or even a solid double despite the large obstacles.
And as others pointed out, if people don't continue to keep trying hard stuff like this, we might as well all pack it in and go home.
PS. Weirdly enough, this just announced: http://provideocoalition.com/index.php/news/story/marquise_technologies_to_announce_real-time_raw_debayering_with_mosaict/
Am I the only person that thinks the footage looks quite videoy?
Especially in the trailer: http://vimeo.com/38286282
Or maybe the shutter speed is just too high?
Great blog Stu.
The interview with Philip Bloom seems to have answered a few questions. Apparently a smart company can do a lot if they aren't trying to control the ecosystem. Red couldn't do it since the risk/reward didn't scale for them. These guys are using off the shelf tech and aren't really implementing a vertically integrated ecosystem strategy, hence $3k camera.
They also seem to have an altruistic streak since they acknowledge they could charge a lot more but choose not to.
So their motives aren't what one would expect. Their hippie aesthetic is their competitive advantage the way I see it.
Technology sure is disruptive these days...
One of my concerns is the decision to base the camera on a CCD produced by a company that's currently going through bankruptcy. They're betting the future of the product on a part that may not be available or have a wildly different cost when Kodak emerges from bankruptcy or is broken up.
I don't know if the espresso guys will succeed, either, but their Kickstarter page goes a lot further in laying out their case. It's very well done. It might help if the Digital Bolex Kickstarter page had a little more information to alleviate concerns.
Also, you can buy a Rancilio Silvia for around $600 and hack a PID on for about $300. You do the work yourself, but you get a PID controlled espresso machine for ~$1000 now. Take away the stainless steel exterior and put all the money into the internals and $400 is within the realm of possibility. For instance, Gaggia used to sell the Carezza (essentially the Classic internals in a cheap plastic exterior) for around $300. (I had two, they both broke down after about a year, buyer beware).
Excellent Blog post Stu,
RED was supposed to be the hippie outfit that democratized digital film making for all. Be it the monumental task or a shift in priorities they left all of the lower base out of the picture when they abandoned the Scarlet 3k for $3k. They decided to be a pro camera company and not a hobbyist and amateur shop. They lit a spark but they dropped the ball on the wider market which may be a good decision based on margins and sustainability.
These hippyish d Bolex folks could be the Apple Garage of the cinema world for the low end base. It is only a matter of time for the RAW cameras to get lower in price and for storage to get so ridiculously large and inexpensive to to be able to be functional in every handheld low cost camera.
In the end we all win, a couple of full release films were made with the 5D and 7D, RAW video isn't everything but it sure is the logical progression of the technology.
I wish them much success in failure or victory.
Let me throw some optimism to the table...
* the sensor in this thing costs $350 - they're just storing the info coming out of it, so very little processing power may be necessary - in fact, they actually built a camera that's working and that can be sold at a profit at their price point - all they need to figure out is how to make *many* cameras - not trivial (ask RED), but not impossible
* I'm not in for either S16 or 2K bayer, but I'm very happy this has come to life, precisely because it will help iron out all the raw workflow issues you mentioned: if lots of people are using Adobe's cinemaDNG on low-end systems and standard software, all these issues will get ironed out - by the time I go for raw video (hopefully, in a year or two; certainly, with an APS-C or bigger sensor), all these crazy explorers will have paved the way, and I'll be able to bring my raw footage to the editor just as I do now with files from my DSLR (and those crazy explorers will troubleshoot the workflow out of joy and excitement, happy to be riding the wave; it's not for everybody, but if you are one of them, you know you are; I used to be, not so sure anymore)
I want these guys to be the next RED: don't listen to the naysayers, bring a new revolution to the market, and slash prices by an order of magnitude, providing what the customer wants to buy instead of what the maker wants to make
if it's not them, it will be somebody else: I think we are, again, at a time where manufacturers are holding technology back - in this new world, that situation can't last long
Phew thought it was just me but no I've found another Grumpy Gus in the wild.
Theres so many red flags here you could stitch them together to give the bedtime story girl a wonderful circus tent to read herself to sleep in. Though of course I too would love to see them all furled up and put back in the dusty retro flag box for the next couple of dreamers.
First of the red flags that have this camera parked in curiosity pit lane is the reliance on the retro Bolex heritage, name and design cues. Any model Bolex has bags of nostalgia and coffee table appeal for your next 50's cocktail party but that doesn't make it the ergonomic basis for a 21st century indy camera. Unless your a fixie bike soccer mom.
Bolex's were form-before-function in the thirties so i can't see the point beyond retro cool to perpetuate the Star Trek phaser, Hasselblad design cues today. It was the box brownie of the moving image. something to keep in the picnic bag on the luggage rack of your MG or in your purse till you needed it or the iPhone was invented. Its an ergonomic nightmare for any shot longer than a short grab of juniors first ride in the radio flyer.
Look to the Ikonoskop for a lesson in how to make something very small but very ergonomic, Im willing to bet they started with a shape that suited its job best then built the electronics around it rather than a retro cool camera based on the shape of a 100ft roll of pulped horses hooves.
I'll leave the 'RAW', 4:4:4, '2k iPad' comments and numerous other technical red flags to such highly decorated scrutineers as Stu but in my opinion as an owner of a few expensive raw cameras, its overrated. You can get a great, grade able image from many many non raw cameras and holding onto the raw dream at the expense of many lost rendering & processing hours is a rapidly fading one. Ill trade debayering, copying and exporting time for creative editing time any day.
Look theres a lot to spark the interest here, mostly their passion and sense of fun and yes theres always room for someone new to reinvent a category and shake up an industry but just ask the Ikonoskop boys how many deadlines they missed, prototypes they threw away and price points they didn't make before they got to the A-Cam Dll they aren't selling many of.
Best of luck though really.
Jas
I shot on the Ikonkoskop and loved it. We had budget so we had a post house to do the transcode for offlines and then went back to raw for a grade. Looking at using it again but with less budget so may have to do the transcode ourselves...bit of a pain but once you ave the workflow its just time adn a fast computer, right? We shot a 2nd camera which was a 5d. They intercut well - we purposefully graded some shots with popping the yellow - most of them are 5D shots.
The bolex looks fun but is it pretty enough as an object to get executive hipsters to buy on a whim and is it practical enough for anyone else to buy it to actually use?
For my ikonoskop project please see here
http://vimeo.com/32780383
I am convinced many of the DSLR shooters who have been clamouring for "RAW" for the last few years don't actually know what it means.
I think for many they simply think that raw is uncompressed, which is also the impression the D-Bolex page give - but that's certainly not the case. RED's R3D format is heavily compressed but still raw.
In most cases if the operator can white balance and expose correction then a raw image doesn't offer a whole lot more. It's potentially useful in grading, but for the most part that ability still isn't widely utilised.
Working in post production I'd much rather see cameras shooting high-quality (compressed, I don't need uncompressed) edit-ready formats like DNxHD than yet another unprocessed raw format that needs either pre-processing or loads my system with a heavy real-time processing load.
As for the format itself, it's interesting. Although I suspect many people forget that 16mm is pretty close to the "video look" of the 2/3" chips we're used to from broadcast cameras. That said I know some people with some great Super16 PL glass who'd probably love to have one of these kicking around.
Does the handcrank actually work as you might expect a handcrank would? Because that would be pretty cool :)
Why does 2-3MB for a 2K Raw frame seem low to you?
Scarlet's throughput is 50MB/s, which means about 2MB per 4K frame.
Also, lots of people write RAW. Yes I know it's wrong, no I don't know why they do it, but it's pretty common.
Stu summarizes most of what I feel as well. Certainly competition is good for the industry. Last time I checked, cinema quality super 16 lenses cost a LOT more than any non cinema DSLR glass. So this camera is affordable? In what way? This upcoming NAB should be quite interesting...
I think they're writing RAW because most people expect to see it written like that
the official Canon 5D3 spec list also says RAW
http://usa.canon.com/cusa/consumer/products/cameras/slr_cameras/eos_5d_mark_iii#Specifications
also, 2048x1152x12 is exactly 3.375 MB per frame, which matches the FAQ number but not the product description number
sorry guys but there is no dbolex prototype ...
their promo video was shoot on prosilica gx2300 machine vision camera...
http://www.alliedvisiontec.com/uploads/pics/Prosilica-GX.jpg
http://www.alliedvisiontec.com/emea/products/cameras/gigabit-ethernet/prosilica-gx/gx2300.html
download their BTS and look for a camera...
https://vimeo.com/38252059
the dvxuser thread is burning hot...
what they're using in the BTS footage seems to be an industrial camera (prosilica GX2300C)
my (still optimistic) take is that the d-bolex is a hacked version of that industrial camera (GX2300C), to which they added:
* a film-making friendly casing
* an audio recorder
* an interface to easily control the industrial camera
* a simple way to record the RAW signal that the industrial camera outputs through its ethernet port
*a small screen that displays the analog image-out from the industrial camera
as long as that GX2300C can be bought for well under $2K (which I'm not sure is the case), I'm calling GENIUS!!
http://www.1stvision.com/cameras/AVT/Prosilica-GX2300-GX2300C.html
http://www.1stvision.com/cameras/AVT/dataman/Prosilica_GX_DataSheet_2300.pdf
http://www.1stvision.com/cameras/AVT/dataman/70-0067_GX_TechMan.pdf
(there you go: simultaneous post, half empty, half full)
http://www.graftek.com/pages/GX2300.htm
$5660
Isn't the GX2300 a monochrome camera?
Well... of course, the hacked gh2 is out for $700 now, which achieves everything we ever wanted for those who know how to color. Kind of makes the 5dmk3 & Digibolex irrelevant, but I still dig it.
it has been funded, they're well over their target
not it has to happen, which I guess is the point on Stu's blog post
it seems they're not putting a GX2300 inside the DBolex, but instead trying to do something very similar to what that camera does, and using the same sensor too, but in a more film-maker-friendly way
They did an update and said the following:
...
We're really excited to start producing the first 100 cameras, and of course any funding that exceeds our $250,000 goal will go towards designing new mounts and accessories and making the first 100 models as strong as possible. Our Kickstarter camera backers are our ambassadors to the rest of the camera world, so rest assured that we're going to work personally with each one of you to make sure your camera works as promised.
Our manufacturing partners, based in Canada, Iensco (http://digitalcameradesign.com/), are putting together a video update for all of you that we hope will be online by this evening. That should answer some of the questions about everyone's tech background, and will show some of the electronics they're working on for the camera. Check out their website to learn more about the design work they do on really cool niche cameras.
Joe and I offer our eternal gratitude to each and every one of you. Keep on asking questions, and we'll do our best to answer as soon as we can!
Wouldn't using a film camera be better to get the film look. I hope the D Bolex succeeds, but I still maintain if you want the film look, get a film camera
I hope they move away from the animations in demonstrations and go to a place where there is plenty of colour, like a garden, or check out the local pond and see the texture on a duck's feathers. The demos so far have been dark and painful on the eye.
However, with luck and talent they will get there. I would go the ikonoskop at the moment though, or, just get a film camera, a real Bolex, or maybe Arriflex SR3, very nice, and bloody expensive to produce on compared to digital
Such is life