#apertus IRC Channel Logs

is there a way i can view telegram without having to sign up

ItsMeLenny I was under the impression that if you were to click on that link you'd be able to browse chat history... on mobile at least.

its a pretty sketchy program imo, compared to most

If there's something better with similar features I'm unaware of it.

even this "Cryptography experts have furthermore criticized Telegram's use of a custom-designed encryption protocol that has not been proven reliable and secure."

Discord I've yet to be convinced about. But I did try it.

havent seen any in irc either

even uses the same sensor doesnt it

it says its "powered by linux", uses an intel nux7i8dnbe processing board, with an i7 8650um , and either a CMV20000 or Sony IMX253 sensor

Not sure what sensors the camera manufacturer uses. They make industrial cameras. Many variaties in all shapes and sizes.

i just said the 2 sensors it uses

pardon?

the camera manufacturer isnt the people behind the camera?

whats the original camera called

Ximea.

ximea is the sensor

No. Ximea is the camera manufacturer.

ximea are like usb3 cameras and stuff ye

Anyways, despite what everyone is saying, it's not open source by any means. Not open hardware either.

If it was open source they'd say it was open source. They call it an 'open platform' instead. The cinegear news website call it 'open source', probably because they don't know the difference and the word 'open' was used in its marketing materials.

Hope I didn't sound snobbish there. What I mean is, we're a community that's building a camera from the ground up, those guys are building an enclosure of some sort.

I have a MQ022CG-CM from them - the camera is wonderful, built like a tank, tiny, and the API is also super easy to use

The sensors they're using are definitely of the machine vision kind, with only a few baaaaaaaarely scraping the cinema world. The dynamic range is lower than the already low DR of the CMV12000 on most of them

Because the nuc won't be powerful enough to do anything besides dumping CDNG onto the SSD or doing ugly low-bitrate compression(OBS or something)

But I believe even the cheapest Ximea cameras are over $500

se6ast1an mentioned last week they might not be able to attend because of the maker faire

RexOrMatrix[m]: ?

I guess that would be best, Bertl_zZ is also still on the road, as is BAndiT1983|away

Great, no problem!

at the same time?

in 23.5h basically :)

I will also write an email when I get back

Meeting in half an hour.

Is everyone present for the meeting at this point?

I'm present, but will it start now or in 30 minutes?

I'm also here, no problem for me

OK Fares do you want to go first... For the reader - Fares is working on JPEG Lossless Encoder Core. His tidy work can be found here - https://wiki.apertus.org/index.php/JPEG_1992_lossless_encoder_core

hi everyone

I implemented a partial reset for the core that will happen after every frame, so it now can handle stream of frames

I also continue working on the report adding two sections, performance analysis and explaining the core configurability

thank you and hello everyone

I have continued with the debouncing code, so I have added some cases I hadn't considered before

and I have been sorting what I want to have on the documentation, and been writing about it

that's it from me, thank you

tell us*

That's excellent thank you.

Hi everybody

Now I have perfectly working 40 bits gearwork work from the Virtex to Plugin. I tested the BER of all five connections together and it is lower thatn 10^-12

Things which are left to do are: 1) Increase the Bandwidth from 2.4 Gbps to 3.0 Gbps, 2) Write the code that sends the data from AXI FIFOs to the encoder instance on Zynq, 3) Debug the issue in data transfer due to high CPU usage. Currently I am debugging the code on Remote Beta alongwith creating documentation and report.

Thats all I wanted to say, thanks to all

Apparently, I wasn't able to detect any error till now ( I tested till 10^12 bits), but yes ECC and CRC will be good to implement

I have done completely unidirectional data flow without any feedback signals to the Zynq board

So the FT601 interface needs to be worked on more, but the 6 LVDS lanes are tested and performing well enough.

sorry for interrupting the meeting.

Thanks

Anything else you wanna mention Apurva?

Thanks for your time :)

Hi All,

Made changes to the CLI .Added functions to help segregate different types of request as the number of parameters were different for each type of request. Yet to test it in Beta as it is not available currently.

This week will mostly focus on completing the Unit Testing part .

aSobhy: You about?

Thanks.

When will the mentors return from Hannover?

In the next hours

Great!

It's one of the two end points to control the camera parameters, the other one being the WebUI

apurvanandan / Nira / Fares / Y_G: Is there anything any of you need to help with your work?

Nothing here as well, thank you

XD

ok, that sound interesting :))

Otherwise thanks everyone for the updates and let's assume that the next meeting is the usual time next week. Have a great week everyone and well done on the progress since the last one.

You mean we are getting?

Yipppee \o/

We call it "the goodie bag" :)

hi Y_G, will check unit tests and trello comments tomorrow, seen them today but had not much time yet

:O

https://twitter.com/ApertusOSCinema/status/1163105395000467461?s=19

The students want a picture of your rotton sandwiches. If you didn't take a photo of yourself stuffing your face earlier please take one of the next sandwich. No one will know the difference.

In fact, I think everyone's more interested in your eating habits than they are in Maker Fair... and that's the truth.

this would also be a big step in the direction of controlling the camera from outside

https://lab.apertus.org/T939 <- Went through these.

Will look at it in some time

great