#apertus IRC Channel Logs

Bertl: wie lautet deine emailadresse?

Also, the GAMA modular design. Is the Beta essentially the "base" of this. I realize that the beta requires an external recorder but in the future if you guys design like a SSD module will it be functional with the beta?

the beta plattform is for testing and developing

the main purpose of the beta and crowdfunding is to speed up and ensure high quality for the gamma

The Super35 is 2300 after the campaign. How much of that cost is the sensor?

err 1300

Wescotte: no, Beta is not complete, the crowdfunding is for the development of AXIOM Beta

https://forums.adobe.com/message/6748854#6748854

good morning

added "as seen on:" block to the campaign page: https://www.indiegogo.com/projects/axiom-beta-the-first-open-digital-cinema-camera/x/5022798

I also created some axiom button designs: https://docs.google.com/document/d/1A6ElumHb5MQNAouJSprBH6KXH8vp-5OJmLfCYGb6aTE/pub

some things like jean connery and van damme I find funny but nobody else... :)

the buttons are nice

on the as seen on blogs

having angry comments is a natural side effect of big public exposure

give me a link

http://www.reddit.com/r/linux/comments/2fsvjh/please_share_and_help_pave_the_way_for_the/

we should maybe post to reddit again

for example

great

please review as well, here: https://cloud.gerade.org/index.php/apps/gallery/ajax/image.php?file=88644302ffde4f6e5659627a3c8bbf32%2FAxiom%2Freview.jpg

yes thats nice.

just send it to me in pm

youtube url

like "Ugly thing obviously designed by someone who doent know a word about desing." :)

the fun continues

ok reauploading to youtube

good "morning" ;)

people were not so skeptical even with digital bolex when it was proved that there was no prototype and the sample footage were actually shot by another camera that was using the same sensor bolex would use

true, but on the other hand DB raised 200.000USD and took full price camera preorders, we take development donations only and have a prototype already, and dont offer a full featured camera yet but a development platform

maybe we should have done this, but on the other hand... maybe the people wouldnt have spent that amount of money then...

so i think this was a very clever step, as the money donated is purely for developing and not for hardware

and then, boom, sunshine blue sky and out of notheing ice from the sky... weather is so creazy right now in north east of austria...

se6astian: I've attempted to start a conversation about the camera on Reddit http://www.reddit.com/r/Filmmakers/comments/2gy5bo/do_you_want_an_open_source_hardwaresoftware/ and http://www.reddit.com/r/Cameras/comments/2gy5z4/do_you_want_an_open_source_hardwaresoftware/

lets see how it develops :)

Kickstarter vs IndieGoGo might have also played a more significant factor too..

:)

hello

as I noticed it was still present in the ungraded .mxf video

intracube you download a fixed pattern noise correction tiff and you subtract it

is fixed noise consistent across every sensor of the same model or is it something each user will have to generate a tiff for on their own under controlled lighting conditions?

se6astian?

aombk: yep, somewhere on the apertus site or fileserver there was an example FPN .tif for the alpha cam

you tried it with the ungraded mxf file right?

yes i think the mcf file has fpn alread subtracted

maybe the pattern noise varies according to parameters

like gain f stop etc

not fstop i mean apperture speed

it can be done internally and autamatically

and even if this not happens at realease i think it will happen some days after by someone of the team or the community

https://www.youtube.com/watch?v=ylF4Y0t59Xw

but the profile is not 100% perfect

the pattern is unique per sensor and changes with the said parameters

in the FPGA processing pipeline we can already compensate FPN entirely, the problem is measuring the profile properly and adapting it when the parameters change

yeah, it sounds tricky :/

especially with some insight into how Canon does it :)

heh

hey guys, i think we should switch strategy and promote the gamma version instead of the beta version. instead of "selling" beta cameras, collect money for the developmentprocess (which is the beta in fact!) so we could advertise modular design, resell older parts and buy new ones, 4k raw, xlr hq sound module and so forth... so people see the PATH and not just the one product (the beta) which isnt that great over all.... i guess

we dont know when gamma will be ready and we dont know what it will be able to do

as we austrians say: i ll be back

the beta is just a tile on the path to the gamma, a very very important one, but i wouldnt sell it as a standalone product, as thats not what it is, but the whole crowdfuning campaign tries to sell the beta version instead of the development of the gamma...

anyhow, i ve to leave, cu later

tomorrow morning should be fun.. holiday trip to canada starts.

evening folks!

hi Bertl :)

seku: link?

just a small mention of that he is doing it. no real explanation, or how he does it

too bad they didnt get into dolby vision postprocessing, even tho it was asked as a question to the panel

*back* had to restart

seku: sounds like they are running two synced cameras at difference exposures with a beam splitter and then combining the image in post.

well, that is the only correct way to get HDR, where the motion blur is same for both channels

*exposures

Isn't motion blur created by the shutter speed/angle? So changing the aperture wouldn't affect motion blur?

you cant change aperture on a mirror rig, the objects would be differently sharp (you change DOF by aperture)

it is very visible in the alpha footage that it was not shot by a pro - there are scenes where the short exposure is ruining the look

that might also be a mighty interesting diy way with the axiom

Does this happen in analog with the raw voltages on the sensor or later in digital space?

Bertl: ah ok. I don't know about this area at all :)

would the FPN processing happen before or after the profile adjustment in the digital chain?

maybe analog amplification is the question?

that's what I was wondering. otherwise any changes to the profile would make subtraction a lot more difficult

as Bertl says, that's done in digital space

you can read up on all the different (HDR) modes and how to configure them there

heh

fun nevertheless

*shrugs* i didnt even understand that then

but it wouldn't hurt to discuss features like the HDR modes here on the channel

piecewise linear is not log, never will be

just you say. thats something major :)

linearising the linear bits, i guess...

afaik they dont even probe the HDR capabilities of the sensor?

the thought that kinda blowed my mind was something you wrote about in the campaign about. ... configurable dynamic range. like with a knob. (and somewhere i imagined that one day, that process could be made automatic, so the camera would configure the sensor to use the dynamic range it needs per scene)

seku: auto ettr does it ;)

yeah, I'll rewrite the algorithm to make use of the latest sensor findings

btw, have a link showing that blackmagic and cion use the same sensor?

i guess it were threads like these : http://forum.blackmagicdesign.com/viewtopic.php?f=2&t=21206

even if its the cmv12000, might be another version

after holidays :)

<quote></quote> tags don't seem to work beyond the comment preview

click your account->posting and you can change the mode.

just switched to classic. comment post mode is set to HTML

thanks, I'll give that a try

tried to make it at 23:XX

Right now it's not very clear that the main purpose/end result of supporting this development phase is something like the Gamma.

Maybe we should start using a slogan like "Help us to develop the camera of your dreams." Or "Support the development of the ultimate camera and purchase a development kit."

The oculus rift was very clear that their beta products were for development only, but people still got really excited about purchasing to test and develop with them.

But only if the upgrade path is a little more clear for potential buyers.

nice quote about the oculus

if to many buy the beta this could even backfire

its a diy starter package :)

yep

Also the reuse of the sensor is mentioned

or so

so you have some 300-500 early adopters for ever to test the products on and with them, but for that they pay just material cost. they pay to testing but getting everything erlier than everyone else

but I have trouble to promise things for the distant future

"hey I got the 300€ perk 50 years ago young man" :)

well, one has to think it trough, but this could be kind of a "licence" which can be sold

as they act as beta testers all the time

usually such people get paid

no more erlkönig anymore

I think that the second collection of beta 35s may have been added just a little too soon.

mhhh,... jo

i would like to hack this for the axiom :D

yeah

weellll.... :D

that would at least make it very scarce :)

it doesn't look like the folks want to go with the KAC anyway, I'm pretty sure some of those 9 Units were mistakes as well

*community

and its cheaper and got better dr and... well...

global shutter

mhhh well its hard to decide now anyways...

before you have fo order the parts for the first time

ok time for bed

"<Bertl> not sure why the truesense is treated so badly"

and the shallower DoF possible

jop

well, that too ;)

its all about marketing

promote a 4k raw 300 fps modular open hardware & software camera for around 2000€

front/top/side views

if there would be an DOABLE upgrade path for the gamma to 8k, i would even promote that

i would say, standard edition 4k, upgrade option later on to 8k

Bertl: ok

why not?

first, we have no idea how the gamma will look like and what it will be able to do

derWalter: is 8K -really- needed now? only a few productions are going for 4K currently!

to answer that in short, no, but you remember the megapixel hype?

at least

then we are fixing and developing finishing the beta

going into gamma developing phase

another year

another year till they get shipped to the customers

you would need to change the board and the sensor for that, mostlikely, because of that its called an upgradepath :) which will be available by then :)!!

yeah, just for video cameras :)

based on the same sensor tech that's in the new Sony A7s

that would give awesome low light performance

nothing like the gamma will be, and everyone (but the geeks and nerds who are really into this!) are interessted in the beta

derWalter: I don't think that the majority of backers are geeks and nerds

nope, i dont think that as well

but as the thole campaign is set out for selling beta pices for retailcost...

all you would need is 100 000 ppl donating 1€ for the development of the gamma

you would need around 300 people who are interessted in buying a beta version

what would they develop for? where would they get such a money? for what?

people who want to try crazy stuff

those are the people for the beta

you would need some kind of powersupply which they would have to construct, some kind of external recorder and screen

they got 4k raw

a bmcc 4k is 2765€, with 4k raw internal recording and touchscreen attached. so people who are buying the beta are not doing this for the device, they must have other reasons, but what are those reasons i ask?

I don't think so, indiegogo is very limited

s/want /talk

scientists... good point, just emailed a researcher who teaches "Cameras as Computing Systems" and asks his students to program CHDK cameras as homework

well if it fails we know how to do it better next time :)

i am dreaming for years of this: Single-Photon Sensing

hey alexML_! how's going?

not really, it just demonstrates the concepts with existing frameworks