AXIOM Gamma Camera Concept Update
Submitted by Sebastian on Mon, 01/06/2014 - 20:45
Romance with the Big Camera
Looking at this "behind the scenes" image from Romance on the High Seas, one wonders how people managed to create entire feature films whilst working under the limitations imposed by professional cinema technology in the 1940's and 1950's. Today we are used to carrying around greater processing power in our pockets than the fastest supercomputer (the Cray 1 from 1979) was capable of more than 30 years ago. Times have changed dramatically!
Camera Size Matters?
In our Build your dream camera survey, when asked about optimal camera enclosure size, 69% of responders have so far stated that camera size does not really matter, 27% prefer a small camera and 4% prefer a big camera. Our motto has always been "as small as possible, as big as necessary" and now we're in the process of evaluating different options to judge exactly "how big is necessary". Space for electronics is obviously one of the most crucial factors, however miniaturization poses a difficult problem for designers/developers due to the expenses involved with fitting more parts into the same space (requiring higher layer counts on printed circuit boards) and the added complexity that is connected to working with large numbers of small components. Another consideration that must be taken into account is the ease of repair-ability. Obviously, it is important that a camera is designed from the outset to withstand heavy usage in a variety of scenarios. However, accidents do happen, and when something goes wrong or the hardware is damaged, it is equally important to us that an open source camera can be repaired easily and with minimal costs in expenses.
On top of all of this, we've also started adapting our open modules concept to allow for more modularity on both the inside of the camera as well as on the front (making it easy to change lens mount/filters/image sensor). In doing so, we've started evaluating different enclosure sizes. The original concept drafted the camera head with only 8cm (width) x 8cm (height) dimensions. After testing a series of variations, we've come across the 10cm (width) x 11cm (height) form factor as a good balance, providing both compactness and the necessary space for the AXIOM's internal components. We want to stress that this size is not set in stone, and that we're continuing to evaluate these dimensions in greater detail.
and the AXIOM project is certified Open Hardware by
.
15 Comments
You guys should put a
You guys should put a "somewhat useful" option on the survey
A pocket camera with a big
A pocket camera with a big lens.
Why not have best of both worlds, a case within a case, the camera module stays small but can be mountable in a bigger case with extra functionality in it. thus 8x8cm POV module inside a 9x12 case carrying extra functions like GPS, storage etc, even optional battery, android and a rotatable screen mountable on the back, for portable use, couple little handle/control bars and you get a BMCC or Scarlet alternative.
Thanks.
Wayne.
Couldn't agree more. The
Couldn't agree more. The minimal core should not do much more than holding lens and sensor, taking sensor/lens control input and outputting the RAW video signal when connected to power source. Other manufacturers can concentrate on usability and embedding the core into various scenarios.
Much more important (and probably much harder) is providing a simple yet useful hardware and software interface to interconnect with add-ons like storage, display, power supply and so forth.
I also agree - recently, I
I also agree - recently, I shot my film "Noctua" on a Blackmagic Pocket Cinema Camera for two reasons:
1) The entire film was POV-based, so I wanted to make sure I can mount the camera on the actor's head, and
2) The GoPro didn't give me a cinematic image quality, lens controls or any of the other things I wanted - hence I bought the BMPCC together with a friend and now it's my main camera.
So, if there is one mini version of the Axiom that then fits LEGO-style into the larger "Standard" housing with i.e. 4 hexkey screws as a mounting option, then you can use the small module for anything POV related (and have the best smallest cinema camera in the industry), and use the combined regular body (small module plus standard housing around it) for all other shoots.
Toby
Way to go, amazing work !
Way to go, amazing work !
As long as it's sized and weights not too much and is rugged enough so that it can be easily put into a backpack, I don't think it size will matter that much. For pocket-sized, smartphone and consumer cameras are fine.
Cost of PCB layers isn't high compared to the camera sensor and other components, but if you have a bit extra-room for shocks-absobtion, and better mounting, including better compatibility in size with e.g. storage devices (e.g. SSDs), then it will make the stackability easier. Also if reparaibality is easier, it will help too. So I would say small and leight but not miniature.
I would focus on fastest path to first full prototype on your Pro target. Then it's always easier to make a second one, smaller, lighter and cheaper. Wanting to accomodate every use and person will not lead to something nice. Keep focus and the good work. 2014 is year of 4K, and meeting that year with a Pro open-source product is more important than anything else (like size or other concerns that can be postponed to your 2nd generation camera).
E.g. Apple's 1st-generation products are usually quite poor in function and features, yet it allows them to gain invaluable feedbacks from the field, and gains them time to improve further 2nd generation products, which are usually much better.
So, really, do the design, choose a case that fits it, and move on to solve next show-stopper tasks :-)
Get the show going :-)
Although you probably make a
Although you probably make a real good point, I think it would be best to keep the essentials to a bare minimum and put it in the smallest box possible. (no non-sense, no bells, no whistles).
I think people need to start to accept this idea of starting off with one DAMN GOOD basic building block, a strong foundation, to build on. The same principle with this block phone thing. https://phonebloks.com/en/goals
Making a great launch in the 4K year will depend on their market: don't expect that the current professional industry will jump on this camera like it is the new big thing. They have invested in their Alexa stuff (and the likes) and they'll be more than happy to invest in the new 4K Arri (which they'll have to release somewhere this year anyway).
Apertus' market are the people who need quality but don't have large budgets and who rather see most of their budget go to the actual "art"-proces in stead of the "technical"-process: like insane camera rental prices for instance. (I think this is the basic philosophy of Apertus in the first place. Which does not mean I don't think they deserve a big fat paycheck by the end of the year haha.)
Apertus will have to keep the price at a bare minimum so everyone looking for the best quality for the lowest price is going to find what they need. These people don't need bells and whistles, they couldn't care for less. Most of these folks own their own gear (bells and whistles) already anyway.
After the release of the basic module, Apertus could release -for instance- a studio update: a second module (time sync and stuff like that) that connects to the basic module like a piece of lego. (I think this was their initial plan in the first place anyway)
The reason I say this is because the release of their first product has to be BANG on. So people who bought the first camera won't be disappointed after 5 months when their hear about the release of the new, better, updated version - at a better price! (or whatever)
Offer the people all the necessary hardware in one small box, no bullshit. All the hardware they are going to need for the next (3-5 years or so) to be safe. You can always feed them software updates that make the image even better after that. You can always release new modules that extend the capabilities of the camera after that.
Probably the best solution is
Probably the best solution is to keep it really as small as you possibly can. The smaller, the better! (There are already tons of very nice heavyweight shoulder rigs to transform the feeling of lightweight cams to that of a big 35mm film cam.)
That way it would be:
1. more affordable for budget filmmakers to buy the essential kit (super 35 + global shutter + RAW + DR all in one small box)
2. a lot more interesting in terms of filming in tight locations (cars, very small houses, closets, ...)
3. very flexible for filmmakers who already own a lot of camera rig mount stuff (rods, brackets, clamps, external monitor, viewfinder, shoulder rig, ...)
P.S.: I'm really looking forward to the performance of the colour rendition of natural daylight, tungsten and fire of the sensor you guys are using. If it is even half decent I think you have a fantastic, industry renewing product here! Best of luck!
I couldn't agree more with
I couldn't agree more with you Wouter.
You can't remove weight (and volume) from a RED One or Alexa, but you can make a Blackmagic Pocket Cinema or any DSLR as bulky as you want using rigs.
More weight means more expensive tripod too.
What's important it's that the camera is rig-friendly. If it does well with sales, I'm sure TILTA, Gini, Letus, RedRock Micro and others will start designing custom cages for it.
Agree with you, Wouter.
Agree with you, Wouter.
KISS (Keep It Simple and Stupid) while keeping quality will allow for easier and faster development, getting out in 2014 (The Year of the 4K), and keep the price as affordable as possible (while it still shouldn't be the #1 reason to sell or buy open-source solutions!).
A plan should be layed out for a release in 2014 (maybe even with a kickstarter or indiegogo crowd-funding campaign to make things going fast with a full-time team).
I love the bigger size
I love the bigger size cameras this is why. Where in a time where we can get raw processing, 4k imaging, and 444 and 422 color processing on 35mm chips. That's Amazing. BUT, The SMALL size of these cameras LACKS the traditional WEIGHT and SPEED of shots in films from the past 100 years. When someone moves a tripod with a little dslr on it, you know that camera looks reasonable nice, or relatively nice for what we have access to, but you know it weighs little. Someone can just so quick whip pans, and tilts, and lift the camera at some unnatural speed because its so LIGHT. The Bigger cameras like the Alexa and Red MX these cameras look better also because the shots are REQUIRED to be more stable, because any real stabilization on a professional level is designed to handle much more weight. Outside of color space, and resolution, and dynamic range, weight is all we have left as a defining factor to what mirrors film in the digital world, in my opinion. I LOVE putting my fluid heads to high levels of resistance, I love to use counterweights on stabilizers, and the JIB is such an amazing help because it provides heavy weight to lighter cameras. Add all the WEIGHT you can, in the smallest form you can, but remember, Having a camera with its OWN natural weight in every shot will make everyone's shots looks more balanced and stable, because they cant just throw it around they are forced to move it with elegance and calculated speed. The Weight of a camera is important to keep up with the Jones's but it is also vital in making sure that every shot has weight to it which looks like a human moving around not some light featherweight object that we know moves so fast it could never be perceived as us or a person(most of the time). Just an opinion. But Hope it helps. More weight for me is good, in moderation because it means me and my crew can only move so fast and it gives the whole entire film a more stable and smooth look to itself.
You can always add a heavy
You can always add a heavy-weight module to a small camera for that stabilization-purpose. The reverse is not true. ;-)
As posting feedback replies
As posting feedback replies on https://groups.google.com/forum/#!topic/apertus/CG_Rdp1Svw4 seems not setup, I'm pasting my reply I tried to do there here:
Hi,
This is a personal opinion (as an open-source software and hardware
developer): Seems a very reasonable and sound approach. Changing sensors is
not something that you do more often than when you buy a better sensor.
Also, most people have a set of lenses with a given lens mount.
Choose the simplest options for you for now to get this open-source camera
going so you can start shipping soon. Real-use feedbacks will be hugely
different (and more valuable) than feedbacks on drawings imho. Also, once
shipped, the open-source community contributions back will be giving new
ideas and solutions you don't even dream of yet. So, don't worry too much
for best decisions now. Next to best which allows it to go out faster to
users and contributors is much better than best solution ready only in a
year's time (and this is not even taking in account the speed of technology
improvements).
I can't stress enough how important it is to release (in your case ship) as
early and often as possible in an open-source environment (same being quite
true for proprietary products too).
So, compared to intrinsic quality and awesomeness of the concept, camera and the huge value of being open-source, the design details of how things are screwed together or sized in details has only minor importance. What's important is to make decisions quickly and move ahead, so that decisions can be changed quickly too if you hit road-blocks because of a wrong decision. Wrong decisions are not a shame if fixed quickly. Open-source projects can refactor and release more often.
You don't want version 1.0 (or 0.1) of your camera to be the perfect one, but you want it out asap, so that you can improve in version 2.0, and aim for a "perfect camera" (in today's view) for version 10.0+. ;-)
Best Regards, and keep the fantastic work!
Beat
The newsletter google group
The newsletter google group is not intended for discussion but rather as distribution list of news.
Thanks for all the feedback.
Any size remotely near what
Any size remotely near what you are shooting for is fine. Having said that, design with 8-layer PCBs, which are so cheap relative to the sensor and integrated circuits is just silly. Wow, that is pretty extreme fixed pattern noise (in your example). Is that due to the high clock speed? Is it less when you shoot at lower frame rates? Yes, I know it can be fixed in software... up to a point.
Overall, the camera looks great. Now I'm going to look for design and technical specs to compare with cameras I've designed. Best wishes!
All of the Polecam Operators
All of the Polecam Operators would highly appreciate the 8x8 version. The smaller the better. And having 4k on a Polecam soon would be awesome. Don't know about the Polecam? Have a look here http://www.poleview.com
Add new comment