#apertus IRC Channel Logs

morning

sebastian committed rBH63acdecaa179: updated PCB thickness and added power board PCB (authored by sebastian).

sebastian committed rBHd05d34bd0f7e: updated assembly (authored by sebastian).

sebastian committed rBH7db7fd8e3a1e: remove obsolete files (authored by sebastian).

sebastian committed rBHe84575cd00e1: removed outdated pcie connector (authored by sebastian).

hey yannick

hey

how you doing?

I just emailed you about updated beta enclosure files on github

I saw the bot message last night, will look at it right now

what is the empty space behind the chip?

the space for the image sensor socket

so I just left a bit of empty space

yes

but the common concern was that the receiver side has to be a rather large PC and that pcie cards are still expensive

se6astian: Where do questions get moved to?

http://lab.apertus.org/ponder/ ?

intracube: as always, you rock. How to move a query to ponder now?

intracube: Rock - 0.1

intracube: This must be fixed!!!1111!!!1111!!!111111!!!!1111

Hrm. Let's try.

right, and Ponder is... something else. forum or Q&A type?

No idea.

Ponder desperately needs to be the _most_ easily found entry point.

So that the lab can return to being lab-like and not a "me too" comment bombed dump?

danieel: how goes the camera?

tasks are supposed to be tasks

philippej: that means ponder is broken by design.

probably

we could morel ikely won't fix those tasks

philippej: maybe the lab landing page should be reworded?

Ideas

feature requests

you are right, now that we havem ore people using it, rewording and linking to ponder might help (might :-) )

yep

Ponder should be "Here is where to put things"

Comments on the microSD vs SD thread is proof enough.

hahaa

back

and the protocol?

DP even in DP++ does not do hdmi levels, you can not wire a hdmi transmitter to it directly

danieel: correct

thats how herbert explained it to me

Bertl_zZ: ^^ drop me a note on how you make hdmi to dp++

thanks se6astian

se6astian, just corected it :-)

troy_s: had to took some design jobs to cover the expenses, so the answer is - on hold

but unfortunately it is not possible to move a task to a question

philippej: there is another ponder point now below

plus "ask questions :"

well we have two douplets now

feel free to enhance it though :-)

It needs _much_ more prominence.

prominence yes, but people will not notice it just because those points are there twice now :)

se6astian: It needs to have life for certain. Some peeps certainly post to the tasks looking for discussion and chatter.

(I am skeptical that closing solves the core issue I guess. The "Ponder" region needs active participants to pull focus.)

(Culture is a big thing for me obviously, and frankly I am already rather sickened by the lack of gender diversity. It makes my participation desire wane certainly.)

troy_s: which thread?

intracube: It was a rather novel idea that had legs, and the toxicity already clammed it.

except where someone is discriminated because of it

intracube: discrimination and toxicity happens below the level of discourse. See the countless bits of interesting studies on gender for example, where graded tests results are different if the marker knows the gender of the participant.

you mentioned about lack of diversity being an issue

Yes. 100% men.

is often hard to tell gender/race/class from an IRC nick or mailing list name

intracube: It is, but then you are relying on suppression.

(Similar to Derrida's insistence that to break a duality / polemic model, one must not engage upon the negative or counter, for that reinforces the same taxonomy. Instead, the notion of a minimal third entity / axis was required.)

sebastian committed rBH45f63e4ab0db: DP connector import (authored by sebastian).

Interesting... The iPhone new patent shows a prismatic beam splitter to land on three independent sensors

troy_s: must be a very special prism

no doubt that the future sensor will be some spectroscopic solution, so the use of microprisms would make sense, but beam splitting with three sensors?

+of

Bertl: What I see in that is a

The only non duh obvious element is that magic beam splitter

I'm pretty sure lenses are already patented by some crazy company

Which gets them the ability to attack any and all implementations that implement the mirror, even though they will be missing the magic splitter

Been donr

Splitters just never work because of the tolerances. Even a CCD I am sure has issues as the light passes through.

I would like to state that Bertl and I have discussed a spectral sensor dated this date

(Used to escape the pre-baked spectral data issue surrounding tri-light sensors)

Bertl: Last I checked the most rudimentary spectral renderer needs to trade the tricolor three floats of RGB in for 40 floats.

Which implies a 40 unit sensor per photosite.

just think trading a 40 megapixel sensor (monochrome) against a 1 megapixel spectral sensor :)

Too

I'm pretty sure once quantum dots are a standard element for optical devices, such 3D stacks will become a commodity

Bertl: make no mistake, if there were a method to get a spectral image it would be a HUGE deal in cinema work

as every photon has a specific energy, converting and sampling that energy on a per event basis would eliminate the need for several sensitive elements

this is what is done in detectors for PET for example

higher frequency -> higher energy

counting events based on the energy (in some kind of bins) would result in spectral information

one of the major problems nowadays is that single photons are not that easy to detect, but technology advances and especially medical imaging requires new solutions

gnight!

nn

probably not