Jump to content

Last chance to order your early 4K raw AXIOM Beta - a Linux based open source camera


Andrew Reid

Recommended Posts

yep i agree, but there is another possibility of end user; people that want to make or modify their own camera and would love all the info   :)

 

My experience is the number of people who will buy the camera for that reason is very low.  And I couldn't agree more with Tim Naylor on the high end.  

 

When I got involved with ML I ended up trying to make the EOS-M a killer RAW camera (and it is, at 720p).  My video tutorial has 6,589 plays all-time.

Link to post
Share on other sites

great idea.  But as with all of these types of things the body design cries out 'software developer' rather than industrial designer and this is the first nail in the coffin.  copy Arri or Ikonoskop's mechanical design philosophy.  Dont go the black magic direction!  why not a sub PL mount or E-Mount with 18mm ffd to allow all lenses to be adapted?  doesnt make sense.   

 

Unfortunately even if they get off the ground and deliver a product, I can't see any of these actually being used for more than camera tests in the first week.  I'd have been more excited if black magic and the a7s weren't available for such low prices.  And i imagine black magic will have yielded the best they can from this now ageing s35 sensor - of which I haven;t yet seen many rave reviews or exciting stuff captured with it.  The pocket delivers a nicer image than the 4k bm camera doesnt it?  Low light capability is pretty much the number 1 criterion in todays list of low budget film maker requirements since lighting is the last thing most people in this price range think about.  

Link to post
Share on other sites

And i imagine black magic will have yielded the best they can from this now ageing s35 sensor - of which I haven;t yet seen many rave reviews or exciting stuff captured with it.  The pocket delivers a nicer image than the 4k bm camera doesnt it?  Low light capability is pretty much the number 1 criterion in todays list of low budget film maker requirements since lighting is the last thing most people in this price range think about.  

Stop reading specs and shoot. Sure the pocket cam has more DR but it doesn't compare  in overall IQ as the BMC4k. The detail and color space of the 4k is in a different league. Different tools for different uses. Just finished an NJOY campaign in which we used F55's and A7s. The A7s got us shots that would've been impossible (interviews at night in dark tinted cars with no lights for example). But in no way would I say it comes close to "slower" cameras such as the BMC 4k in terms of overall IQ.

Link to post
Share on other sites

 So let's say Tim Naylor was on a shoot and that happened when he was converting to preview.  How much would the client love him then?  NOT.  Of course, the RAW footage would have been fine, but he still would have looked incompetent.

 

I can almost guarantee that these kinds of problems will occur to AXIOM users if they use anything before it has been battle tested for months.

 

....or would I rather buy a BM or GH4 or A7S with my $150+/hr and actually go out and SHOOT a few things?  Similarly, if I'm a DP, it's not like I don't have a zillion-hours worth of projects I'd like to get done with the equipment I have!

 

 The market of high-end DIYers is minuscule.

For all these reasons and more, I'll pass on the Axiom. But because someone vetted, guaranteed and extensively tested the software for Sony, I took an A7s on commercial campaign and it was an absolute hero. Less than a week after I took delivery. It may have embarrassed the F55 we had in some respects. A hobbyist with loads of free time and money will toy with the Axiom. Unfortunately for many of us in the camerawork trade, framing, lighting and a dramatic eye doesn't always translate to computer skills. Also, I just don't see proper R&D happening at their current funding level. At 150/hr....... Unless of course you speak Mandarin.

Link to post
Share on other sites

Stop reading specs and shoot. Sure the pocket cam has more DR but it doesn't compare  in overall IQ as the BMC4k. The detail and color space of the 4k is in a different league. Different tools for different uses. Just finished an NJOY campaign in which we used F55's and A7s. The A7s got us shots that would've been impossible (interviews at night in dark tinted cars with no lights for example). But in no way would I say it comes close to "slower" cameras such as the BMC 4k in terms of overall IQ.

 

Let's not turn this into a which 'camera is better' argument.  My comment related to low light performance and in which case if your image is dark and grainy it doesnt matter how much detail or colour information you got.  As you know, the bm4k may have the res you needed, but lets not for one moment even suggest it could have taken the place of the f55.  if it had, you'd have needed a lot more lighting.  Maybe I'm wrong, but I can't see anyone using this camera on anything important enough to require the increased colour depth or resolution.  And by the time backers get their chance to pay the balance $1500 and get their camera there will be something available which just 'works' that outperforms everything this camera can do.  lets assume sony release 2 more a7s successors before this thing is ready to ship...  it's logical.  

Link to post
Share on other sites

 

There cannot be TOO MANY CAMERAS for me! :) :)  However...

 

o. The cost puts it out of the reach of most amateurs

o. It's trying to do too much, too fast and reeks of "vanity project" for all.

o. Commercial users will require commercial support and ML has a poor record of delivering that kind of reliability.

o. If a 3rd party offers commercial support can they survive on just one platform, against other better capitalized providers?

o. There are very, very few people who can/want to developer software--PERIOD.  I cannot stress this enough.  So what's the benefit to them?  In fact, it's a negative because no one wants to feel responsible for something they're uncomfortable with.  In short, VERY, VERY FEW PROFESSIONALS would use anything but a rock-solid, time-tested platform on a real shoot.

 

o. What happens if a serious flaw prevents the first batch of cameras from working?  Who will pay? 

o. Didn't Ricoh try interchangeable sensors?  Some ideas are good in theory but NEVER work in practice

o. Why can these guys do what BM can't?  Because they'll work for free?  Because they're smarter?  Because BM is dysfunctional?   There needs to be a GOOD reason for it to be sustainable.

o. I don't hate the camera makers (even after Panasonic wouldn't fix my GM1  :( ).  Open Source, and other social virtues can get awfully expensive and I'm not sure people will be forgiving when their $2,000 camera keeps resetting because of some shitty capacitor buried in a circuit-board.

 

My opinion, they should take the money and try to create the lowest-price open-source camera, start small, and build up.  Plan for it to take 3-5 years.

 

 

I understand what you're saying but using Magic Lantern as an example of what could go wrong with an open source project is probably misplaced. The first thing is that ML does have many problems, but ML is not the problem. The problem is the nature of the project - basically ML is a Hack, not really a development project. But why is that? Because the software/Hardware that they are deloping for is closed and Canon just will not/have not given them an ounce of help. Therefore, they have to reverse engineer everything. Most of what the ML team does is guess work and trial and error to figure out how the software/hardware works. Even so, ML has been wildly successful. So much so that AXIOM had to resort to ML's thriving community to reach their funding goals. They said that much themselves. That's what I call success. Rental houses, pro shooters as well as enthusiasts - if you own a Canon body you probably have ML running on it (Who shoots h.264 anymore?).

 

If you want to use an open source project as an example, then try referencing projects such as Mysql, Mozilla and there are many more!

 

It seems to me that you missed the point completely - the reason for an open source cinema camera is so that you take the guess work out of programming! . . . No guess work, no faulty software. If the hardware fails every now and then, well, Black Magic will probably understand (if you know what I mean) - that's why we have AXIOM!

 

You are right about not many amateurs being able to afford it, but then again if ML was not around you couldn't afford a real cinema camera that shoots 14-bit RAW anyways. BTW, who's to say they will not make a sub-$1,000 camera - in fact I'd be willing to bet that's in the works.

Link to post
Share on other sites

Let's not turn this into a which 'camera is better' argument.  My comment related to low light performance and in which case if your image is dark and grainy it doesnt matter how much detail or colour information you got.  As you know, the bm4k may have the res you needed, but lets not for one moment even suggest it could have taken the place of the f55.  if it had, you'd have needed a lot more lighting.  Maybe I'm wrong, but I can't see anyone using this camera on anything important enough to require the increased colour depth or resolution.  And by the time backers get their chance to pay the balance $1500 and get their camera there will be something available which just 'works' that outperforms everything this camera can do.  lets assume sony release 2 more a7s successors before this thing is ready to ship...  it's logical.  

Actually, you implied the pocket has a better image. As someone who has used both, it doesn't. But it has its uses nonetheless. I think you may have misread my post (or I expressed my self wrongly). Also, I don't think I suggested "for a moment" the BMC 4k compares to the F55.  I was pointing out how the A7s performed wonderfully alongside the F55 on a recent job.

 

Misunderstandings aside, I totally agree with you that by the time Axiom is rolling it'll be dated. What I'd really like to see from any camera maker is a Full Frame, well priced, proper video camera, let's say a full frame Sony F7 with a crop option.

Link to post
Share on other sites

I understand what you're saying but using Magic Lantern as an example of what could go wrong with an open source project is probably misplaced. The first thing is that ML does have many problems, but ML is not the problem, the problem is the nature of the project - basically ML is a Hack, not really a development project. But why is that? Because the software/Hardware that they are deloping for is closed and Canon just will not/have not given them an ounce of help, therefore they have to reverse engineer everything. Most of what the ML team does is guess work and trial and error to figure out how the software/hardware works. Even so, ML has been wildly successful. So much so that AXIOM had to resort to their thriving community to reach their funding goals. That's what I call success. Rental houses, pro shooters, enthusiasts - if you own a canon body you probably have ML running on it (Who shoots h.264 anymore?).

 

If you want to use an open source project as an example then try referencing projects such as Mysql, Mozilla and there are many more!

 

It seems to me that you missed the point completely - the reason for an open source cinema camera is so that you take the guess work out of programming! . . . No guess work, no faulty software. If the hardware fails every now and then, well, Black Magic will probably understand (if you know what I mean) - that's why we have AXIOM!

 

You are right about not many amateurs being able to afford it, but then again if ML was not around you couldn't afford a real cinema camera that shoots 14-bit RAW anyways. BTW, who's to say they will not make a sub-$1,000 camera - in fact I'd be willing to bet that's in the works.

Software aside, I 'm surprised how little is mentioned about Axiom's hardware. That to me is the weakest link. A slow chip, no EVF, HDMI outs, no internal ND, no smart Canon adaptor, no record module, etc. All the software in the world won't fix this.

Link to post
Share on other sites

Hi all,

 

While I agree with almost all of the posts here - both positive and negative, the thing to bear in mind is this camera is a Beta, not a production model.

 

I'm the person who's been described in other posts as a mythical creature - an amateur videographer who can afford the price and time to invest in this.  And I see that the potential this camera gives, especially with ML on-board, would be an opportunity wasted if I didn't commit to it. 

 

I've a very long relationship with Linux and have tinkered with the ML code so for me a quick fiddle with the source code and a re-compile is no big deal.

 

I know I'm in for a bumpy ride, but for me who has the luxury of not having to worry about screwing up paid gigs, and benefiting from a bleeding edge systems, is really exciting.

 

So for all of you nay-sayer, sit back and just let this project pan out.  It may crash and burn, but it also may be the best thing ever (well, for raw video cameras that is).

 

Just for the record I'm from a background similar to maxotics - been using ML for a few years, absolutely stoked when they developed Raw video, bought a 50D and really enjoyed the quality of the results - became frustrated somewhat with the core ML devs (but still an avid supported for what they've given me) jumped ship to BM and so enjoying my BMPCC because 'It just works' and enjoyed the rollercoaster of a ride with their firmware updates so much that I bought a seconds BMPCC (bugger still hasn't arrived yet though - effing BM and their Swiss distributers!)

 

And while I've been reading EOSHD for years, this is the first post that I feel so passionate about that I've created an account to post with.

Link to post
Share on other sites

i am a bit disappointed by eoshd for the timing of the report of the apertus axiom beta crowd funding campaign. at least you finally did report it.

magic lantern is open source but here is a though:
maybe ML has just a few active developers because not many of the people that can/want to get into development want to supercharge a proprietary product and boost canons sales?

so maybe ml is not the best open source paradigm and maybe thats the reason ML developers are really interested in this axiom beta project.

here is a few other open source projects out there that can be used as paradigms:

gnu linux, firefox, android aosp/cyanogen, apache, blender, mysql, libre/openoffice, inkscape, gimp, wordpress, x264, wikipedia and many many more i am sure i forgot to mention.

also take a look at this just for fun:
http://mylinuxbook.com/awesome-devices-powered-by-linux/

the hardware part is a bit more tricky. there are already quite a few successful projects out there and axiom is trying to be one of them. and as mentioned before, this project has been planed for years. it didnt all start with a crowd funding campaign a month ago.

the difference between apertus and blackmagic design or other camera manufacturers is not that they work for free(they dont) or that they are smarter(maybe they are) or things like that.
the main difference is that open source is a totally different business model. and that is a great advantage in my opinion.
for example its a great thing, having the datasheets of a sensor publicly available without having to sign NDAs with the manufacturer and hundreads of people already are coming up with ideas and tricks to drive the sensor to its limits, before even getting the camera itself.

the cost of this camera is relatively higher than other cameras because fair labor is a prerequisite for the apertus team and the open source culture. this camera wont be produced in sweatshops like many other cameras do and thats actually a good thing.

to use this camera, as any other open source project, you dont have to be a programmer.

its not the same sensor with the AJA Cion and the BM 4k. those companies keep the sensor they use a sealed secret and thats the actual problem. you are holding a black box in your hands and you have no idea whats inside and if you can use it more efficiently, let alone modify it and fine tune it for your needs. and dont forget, this is what DPs (they are professionals) were doing and loved doing in the past.

there are people that want to know how the things they use work and be able to customize or alter their function and there are people that dont. why call the former hobbyists and the latter professional? why not vice versa? anyway i dont think thats the criterion for professionalism

this is a beta camera. the focus is on building a system that can support and drive a sensor as capable as the cmv12000 for example. everything learned from this will be used in the gamma, the sensor as well. it is the first time so many will be able to get this close to a sensors inner workings (if they wish of course) dont you think this will produce amazing results?

 

its ok not everybody has to get one of these cameras and be on board if they feel uncomfortable with this whole idea. this project is not trying to make you feel bad about not joining. after all there are risks involved.

just wait and see the effect this project will have on the proprietary culture in general

 

also visit the apertus website and just give it a look. there are many interesting things in there.

https://www.apertus.org/

Link to post
Share on other sites

The great thing about open source is that anyone capable can code and sell their self made additions to Axiom camera. Think it like this.

For an example let's a imagine a guy called Andrew R. who is a hardcore coder. Say he wants some nifty effect or feature like HDR Video ?? 

Now Andrew spends 3 months in his basement and voila he has something very special. Now Andrew starts selling the feature for 49.99 and earns some x amount of thousands from it.. 

 

How awesome would that be. everyone stands to benefit.. coders can earn nice moneys and us hobby/ pro film-makers can get awesome features for the cheap. 

 

Naturally this tech will have it's challenges as would any thing new. 

Link to post
Share on other sites

FYI

For a limited time you can still

get your AXIOM Beta voucher

(offer

ends Oct. 23rd

)

This campaign raised €174,520 by its October 08, 2014 deadline,

and is continuing to accept contributions.

Perk

€500EUR
AXIOM Beta Super35 Change
Pay 500€ now for development and est. € 2.300 due at Ship of AXIOM Beta Super35 (using CMV12000 image sensor) 5th Batch - Limited time offer!
Estimated delivery: August 2015
11 claimed
Link to post
Share on other sites

Some elements of this are definitely "the future" so to speak.

 

The ability to write third-party apps, modules and so on...

 

I've twice cancelled BlackMagic orders when things ran late, or (numerous) unfavourable bugs were discovered. I tend to use the lower-spec Canons because they have fewer quirks and are stable, and I can replace them on another continent in a few hours. Because of these two things I'm no longer willing to early-adopt.

 

It even happened with A7s... I held out and then the blue-channel blow-out bug surfaced.

 

I really hope it does well as I have a huge respect for what they're doing. I shall wait and see though!

Link to post
Share on other sites

Of course the biggest bonus here is the track record of the Magic Lantern developers.

 

I was only able to use a 550D/600D for so long because of their ingenious free firmware hack.

 

It's surreally brilliant, and if they can do the same in a commercial context they'll have it made! 

 

I shall follow this very closely...

Link to post
Share on other sites

Sorry to be the fun-suckerer again, but I don't see the difference in Axiom's open source hardware and Canon's closed-source camera when it came to ML commercial development.  On the ML forum a photographer offered $300 to fix the boot-flag on the 5DIII and you'd think he was a Christian Right Fundamentalist asking a question at a LGBT forum.  Not only did he fail in getting a dev to look into fixing the problem for him, which he said should be shared, others implied he was trying to ruin ML with a "money" attitude.  The failure of ML gaining widespread commercial success was not due to Canon's proprietary cameras, it was due to the dominant culture of ML

 

IN NO WAY am I saying the ML devs and community are wrong about this.  It is THEIR TIME and THEIR CHOICE.  It is a true democracy, for good and bad.  

 

What the devs work on is what THEY feel an interesting challenge and for those who ask it that they are inspired by.  How will it be any different with Axiom?  Again, I'm not saying the devs won't create GREAT software for the Axiom.  Past experience is already PROOF they will.  

 

The question is who will take responsibility for the system out in the real world where the $5,000 camera (which can be re-sold) is dwarfed by, say, HOURLY $1,000 OUT-THE-WINDOW, ONE-TIME, on-set costs (talent, crew, time to schedule, location scouting--everything that goes into those precious hours)?  

 

Like JGHarding, I'm rooting for all involved!   The sad thing for me is I WANTED the ML devs to get more recognition for the work they did which I SO ENJOYED.  In the end, I accepted that I'd rather have new stuff to play with than a reliable camera.  

 

So what I'm finally saying is that, unless they tackle this problem straight on, Axiom will have to rely on some form of non-profit funding to continue past this first phase of funding.  Perhaps that's a good thing.

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...