This is the first Team Talk filmed with the AXIOM Beta and shot in experimental 4K RAW, thus it covers the topic of color calibration and raw processing. This Team Talk was actually shot quite some time ago but because the RAW post processing was being developed at the same time the Team Talk was filmed, we had to reprocess it several times due to continuous software improvements and results that kept on getting better and better. This Team Talk was shot with the first generation experimental 4K RAW mode whereas the April Fools video was shot with the second generation experimental 4K RAW mode. The RAW processing is quite resource intensive and currently takes around 0.5 to 1 second per frame depending on the computer's performance (especially fast SSDs to read the files result in a significant speed boost). There is still room for optimization and for now this software only runs single thread. We already created a script that converts an entire Edit Decision List (EDL) so the RAW processing can be done in multiple threads once the edit is complete. More details will follow in a separate article soon.
In the first part of this video we talk about why the AXIOM breaks the system of how we compare products to each other. As the AXIOM Beta in our case is constantly evolving it's like comparing a kit of LEGO pieces to a toy helicopter - which ones is better?
In the second part of this video we talk about the process of calibrating cameras for accurate colors and then the deviation from this for achieving "pleasing colors" (e.g. for skin tones) as not just consumer camcorders are nowadays expected to shift colors towards the "pleasing" areas of the spectrum anymore. This is a very much creative and individual process and the AXIOM will feature a wide variety of options and variations here.
GUYS!!!!
You are so great, and I think Im one off your biggest fans (old red fan ;) ) but come on.......Delivery dates ??? 4K is almost history, NAB is around the conner for 8K ;)
Dates Dates Dates....Speed up, this summer, or next year ;-)
Love to the TEAM :-D
Yeah. The development is very very slow. With the current speed the final Beta will wait a few more years: ( Where only take so much money to develop?)
well to be fair it's a big task todo and for the beta there where not many developers on the hardware side so the progress is slow. the good part is every progress helps later i mean if you need later a bigger sensor it's much a smaller task than to build a complete camera with software.
I think the idea is to keep the system very flexible: not every situation is suited for RAW recording. It's expensive compared to recording compressed with baked in picture profiles.
Thanks for the honesty! Hard work and keep up the demystification.
Might I suggest for the next team talk you up the production values with a more subtle/3d lighting set up. It is a little disconcerting to watch a light panel next to the camera, full frontal, over exposed face talking about the subtleties of 'colour science'!
I know you are now more focused on technical stuff. But we're all filmmakers and are trained to have a critical look on a lighting setup. Here I have just a quick tip for your lighting guy how he could improve the team talk videos. This is Eve Hazelton's breakdown of a pleasing lighting setup: https://www.youtube.com/watch?v=uMzJUjMVPwM
I agree the lighting setup for this team talk (and the next 2 episodes still to be released from this shoot) is not the most sophisticated one and we will try to improve it in the future. But we also need to be careful not to over-engineer these development insight videos - they are produced with a tiny team and pretty spontaneously. The more crew/gear/personnel we require to do them the longer it will take to organize a shooting and the less often we will be able to produce them. This last team talk should already have been release for months so we need to find a good balance between spontaneity and sophistication in the future.
Sony a6300 is overrated IMO. The only good thing about it is that it is the first DSLR that doesn't do the pixel binning thing. Apart from that it still has rolling shutter; a limiting codec (although a very good one for its size) and only 8bit recording.
Have some patience.. I'm pretty sure it will pay off in the end! This camera is going to change everything.
I love you guys.. all of you! Thank you so much for putting in the hard work and still keeping it open and thank you for not conforming to the marketing games and assuring full modularity etc. I'm looking forward to the future! :)
It seem to me that you need to increase the gain quite a bit because judging from that reflection on the testchart there doesn't seems to be a lot of highlight latitude.. It would be nice to aim for an equal under- and overexposure latitude of at least 7 stops if the super35 sensor can capture differences up to 15 stops.
I would like to hear more about the Experimental 4K recording and post processes. What recorder are you using? What cables and connectors? How is the Experimental data stream transcoded? - and what format is it transcoded to? What are you using to edit the footage? You know - Everything! LOL.
There's also some blocking going on that flickers in the shadow areas below the desk, but that might just be Youtube.
Well... flashy animation, wish-list of features and promises...Sorry to remind you that, but there is not a huge difference with the Apertus project...
We have seen the Alpha Prototype, and footage from the Alpha. Now we've seen the Beta and footage from the Beta. Also, Beta Units have been shipped to the earliest of adopters. So I would say that Apertus are several steps ahead of the Craft Camera folks.
The one area where I feel Apertus has fallen short (and I argued this when features for the Beta were being discussed) is that it should have had the capability to record directly to internal storage; either SSD or some type of Flash Card. As it is now we'll have to depend on external recorders, and not many of those can do 150 F.P.S., or 8 F.P.S. or any other odd ball frame rate that the Beta is capable of. So the true advantages of the camera will be hidden behind a crippled recording system.
Oh well. Maybe a third party add on will be created to allow internal recording.
Thanks for the feedback, its much appreciated! Agreed, the recording media issue is indeed a problem in the long run but for us it was important to provide a working solution that allows people to record externally until we found a proper solution to internal recording. It would be a pitty if nobody currently could use the camera just because everything would be stalled until internal recording is posible. We actually already have a plan for solving the internal recording issue, but since we first have to test some things (PCBs already arrived) to verify if this way of doing it really works in reality before we want to make any promises so we did not mention it in broad public yet. If it works though it could become a very elegant solution for capturing even high speed footage in a cost efficient, compact and modular way. Stay tuned :)
26 Comments
GUYS!!!!
GUYS!!!!
You are so great, and I think Im one off your biggest fans (old red fan ;) ) but come on.......Delivery dates ??? 4K is almost history, NAB is around the conner for 8K ;)
Dates Dates Dates....Speed up, this summer, or next year ;-)
Love to the TEAM :-D
Keep rocking - but time goes fast - rock on!!!
Yeah. The development is very
Yeah. The development is very very slow. With the current speed the final Beta will wait a few more years: ( Where only take so much money to develop?)
well to be fair it's a big
well to be fair it's a big task todo and for the beta there where not many developers on the hardware side so the progress is slow. the good part is every progress helps later i mean if you need later a bigger sensor it's much a smaller task than to build a complete camera with software.
But how does it go using
But how does it go using resolve (a profile in that)? Which is a leading commercial package with a free version.
GP-GPU processing is 10x faster.
I think the idea is to keep
I think the idea is to keep the system very flexible: not every situation is suited for RAW recording. It's expensive compared to recording compressed with baked in picture profiles.
I'm all for setting a desired
I'm all for setting a desired look of shooting at filming and adjusting as needed afterwarxs.
Hey LUV
Hey LUV
I hope you know that the beta is 8k-Ready, all you need is to plug in a 8K-Sensor and keep shooting, yo?
BTW. who needs more than 1080p? have not heard of a movie which won an oscar for having the greatest
resolution ;)
Thanks for the honesty! Hard
Thanks for the honesty! Hard work and keep up the demystification.
Might I suggest for the next team talk you up the production values with a more subtle/3d lighting set up. It is a little disconcerting to watch a light panel next to the camera, full frontal, over exposed face talking about the subtleties of 'colour science'!
Thanks for the informative
Thanks for the informative video.
I know you are now more focused on technical stuff. But we're all filmmakers and are trained to have a critical look on a lighting setup. Here I have just a quick tip for your lighting guy how he could improve the team talk videos. This is Eve Hazelton's breakdown of a pleasing lighting setup: https://www.youtube.com/watch?v=uMzJUjMVPwM
I agree the lighting setup
I agree the lighting setup for this team talk (and the next 2 episodes still to be released from this shoot) is not the most sophisticated one and we will try to improve it in the future. But we also need to be careful not to over-engineer these development insight videos - they are produced with a tiny team and pretty spontaneously. The more crew/gear/personnel we require to do them the longer it will take to organize a shooting and the less often we will be able to produce them. This last team talk should already have been release for months so we need to find a good balance between spontaneity and sophistication in the future.
Yes, I can understand that.
Yes, I can understand that. Thanks for your work anyway.
Great work guys but as LUV
Great work guys but as LUV mentioned above we supported you from the beginning and beta should have been delivered by April last year.
Currently 4k capabilities are available everywhere. Sony a6300 is cheaper than the first payment batch.
Please deliver - We love the project but somehow frustrated!
Keep up
Sony a6300 is overrated IMO.
Sony a6300 is overrated IMO. The only good thing about it is that it is the first DSLR that doesn't do the pixel binning thing. Apart from that it still has rolling shutter; a limiting codec (although a very good one for its size) and only 8bit recording.
Have some patience.. I'm pretty sure it will pay off in the end! This camera is going to change everything.
I love you guys.. all of you!
I love you guys.. all of you! Thank you so much for putting in the hard work and still keeping it open and thank you for not conforming to the marketing games and assuring full modularity etc. I'm looking forward to the future! :)
THANK YOU!
It seem to me that you need
It seem to me that you need to increase the gain quite a bit because judging from that reflection on the testchart there doesn't seems to be a lot of highlight latitude.. It would be nice to aim for an equal under- and overexposure latitude of at least 7 stops if the super35 sensor can capture differences up to 15 stops.
I would like to hear more
I would like to hear more about the Experimental 4K recording and post processes. What recorder are you using? What cables and connectors? How is the Experimental data stream transcoded? - and what format is it transcoded to? What are you using to edit the footage? You know - Everything! LOL.
There's also some blocking going on that flickers in the shadow areas below the desk, but that might just be Youtube.
Thanks.
We will publish a full
We will publish a full article about the 4k raw mode soon (with software tutorials, demo clips and sample footage)!
guys from Atlanta do a
guys from Atlanta do a similar project http://www.craftcamera.com. Currently there are no real images, but promise delivery in December.
Don't hold your breath -- all
Don't hold your breath -- all they have, it seems, is some flashy CGI animation and a wish-list of features at an almost comical price point.
Animations, wish-list and
Animations, wish-list and promises. Sounds like another Apertus project?
Apertus have shown actual
Apertus have shown actual hardware. I bought in based on the Alpha.
Well... flashy animation,
Well... flashy animation, wish-list of features and promises...Sorry to remind you that, but there is not a huge difference with the Apertus project...
You may think there is not
You may think there is not huge difference, I happened to think there is.
We have seen the Alpha
We have seen the Alpha Prototype, and footage from the Alpha. Now we've seen the Beta and footage from the Beta. Also, Beta Units have been shipped to the earliest of adopters. So I would say that Apertus are several steps ahead of the Craft Camera folks.
The one area where I feel Apertus has fallen short (and I argued this when features for the Beta were being discussed) is that it should have had the capability to record directly to internal storage; either SSD or some type of Flash Card. As it is now we'll have to depend on external recorders, and not many of those can do 150 F.P.S., or 8 F.P.S. or any other odd ball frame rate that the Beta is capable of. So the true advantages of the camera will be hidden behind a crippled recording system.
Oh well. Maybe a third party add on will be created to allow internal recording.
Thanks for the feedback, its
Thanks for the feedback, its much appreciated! Agreed, the recording media issue is indeed a problem in the long run but for us it was important to provide a working solution that allows people to record externally until we found a proper solution to internal recording. It would be a pitty if nobody currently could use the camera just because everything would be stalled until internal recording is posible. We actually already have a plan for solving the internal recording issue, but since we first have to test some things (PCBs already arrived) to verify if this way of doing it really works in reality before we want to make any promises so we did not mention it in broad public yet. If it works though it could become a very elegant solution for capturing even high speed footage in a cost efficient, compact and modular way. Stay tuned :)
Sounds promising Sebastian.
Sounds promising Sebastian. Hope everything works out.
Add new comment