Current Server Time: 20:05 (Central Europe)

#apertus IRC Channel Logs

2019/07/08

Timezone: UTC


01:42
illwieckz
left the channel
01:52
illwieckz
joined the channel
03:22
Bertl_oO
off to bed now ... have a good one everyone!
03:22
Bertl_oO
changed nick to: Bertl_zZ
06:07
BAndiT1983|away
changed nick to: BAndiT1983
06:40
BAndiT1983
changed nick to: BAndiT1983|away
07:43
se6astian|away
changed nick to: se6astian
09:35
n0fx_[m]
joined the channel
09:35
n0fx_[m]
yo
09:39
se6astian
changed nick to: se6astian|away
09:54
Bertl_zZ
changed nick to: Bertl
09:54
Bertl
morning folks!
10:26
se6astian|away
changed nick to: se6astian
10:27
illwieckz
left the channel
10:45
Bertl
off for now ... bbl
10:46
Bertl
changed nick to: Bertl_oO
11:19
se6astian
changed nick to: se6astian|away
11:53
TheProgrammer
joined the channel
11:55
TheProgrammer
left the channel
13:36
se6astian|away
changed nick to: se6astian
15:38
Y_G
joined the channel
16:24
Nira|away
changed nick to: Nira
16:46
BAndiT1983|away
changed nick to: BAndiT1983
16:53
dev__
joined the channel
17:00
se6astian
good evening
17:00
se6astian
irc meeting time
17:03
se6astian
aSobhy: please start the reporting
17:03
aSobhy
Hello all, this week I have few time to work in GSOC project as the Graduation Project delivery and final discussion will be next Wednesday day
17:03
aSobhy
But here is the few tasks I made:
17:03
aSobhy
changed the Clock recovery for the machXO2 to a shaered clock, wrote a word alignment code -not tested yet- for the machXO2.
17:03
aSobhy
After Wednesday I will work more hours to remain on track.
17:05
aSobhy
that's all
17:05
se6astian
thanks!
17:05
se6astian
would you say you are within, ahead or behind the proposed schedule?
17:06
aSobhy
I'm behind
17:06
se6astian
do you see the opportunity to catch up after wednesday?
17:09
aSobhy
yeah all my time will be for GSOC after next Wednesday
17:09
aSobhy
and I'll work more to catch up
17:10
se6astian
great
17:10
se6astian
thanks for the report
17:10
se6astian
nira, your turn!
17:11
Nira
hi everyone
17:11
Nira
this week I have finished a code using interrupt on change and another one with timer interrupt. I didn't know how interrupts work, so I have been doing basic codes and testing
17:12
Nira
and I have been working on what finally the debouncing test would be and started implementing it
17:13
Nira
that would be what I have been doing
17:16
se6astian
great, thanks for the report
17:16
se6astian
considering the originally proposed schedule do you consider your progress within, behind or ahead of schedule?
17:19
Nira
behind
17:20
illwieckz
joined the channel
17:20
se6astian
right, any measures you can think of to catch up currently?
17:21
dev__
left the channel
17:21
dev__
joined the channel
17:26
se6astian
Nira?
17:26
Nira
the first month I had exams, and I also thought it would be easier for me to learn what I didn't know yet, but I can see now that the schedule was too ambitious for me. I will do as much as I can and I will be much more on the project, so I think I will do it better
17:26
se6astian
good, thanks!
17:26
se6astian
dev__: your turn
17:27
dev__
Yes se6astian
17:27
dev__
hello everyone !
17:27
supraraj
joined the channel
17:28
dev__
last week i was able to complete aviencode and also added static allocator to it. Now, OC can produce an avi file after processing (downscaling right now) from RAW frames and make a video out of it by combining the frame. Trying to add Fuse so we can implement different callbacks.
17:28
supraraj
Umm... Unreliable net connection :(
17:28
dev__
Next week, I will try to complete this implementation of fuse and if everything works fine then we can go for implementing Ui for frameserver.
17:29
dev__
Yes, That's it thanks
17:29
se6astian
very nice, can you share anything regarding the avi file encoding process in action already (screenshots, screenrecording, sample avi file, etc.)?
17:30
BAndiT1983
i would like to postpone UI, till the implementation is stable
17:30
supraraj
Honestly, we won't get to that stage for UI that quick
17:31
dev__
https://github.com/kakashi-Of-Saringan/opencine , Here are the commits
17:31
BAndiT1983
yes, commits are there, but for people without development background it's not saying anything
17:31
dev__
okay, I didn't know the possibilities
17:31
supraraj
Have you tried with and without the downscaled?
17:32
dev__
with downscaled
17:32
dev__
is there anything which i can use to upload my avi file
17:32
supraraj
downscaler... Without downscaling, it works good?
17:33
BAndiT1983
how is AVI generation realized?
17:33
dev__
This is screanshot https://paste.pics/36613d9083df65f43c08663fa6a8213a
17:33
supraraj
Just wanted to know if the aviencode knows downscaler parameters
17:34
dev__
AVI generation realized , meaning ??
17:34
se6astian
dev__: is there anything which i can use to upload my avi file <- how big is it?
17:35
BAndiT1983
as far as i seen there was an implementation, which wasn't reliable for fuse
17:35
se6astian
dev__: This is screanshot https://paste.pics/36613d9083df65f43c08663fa6a8213a <- I mean of the encoding process not the finished file
17:35
dev__
17.6 mb
17:35
se6astian
will pm you account details
17:35
dev__
I had reused the code from your prototype also , after that
17:36
BAndiT1983
have seen this image before, why is it flipped? are there flags in AVI to get it in right direction? this and many questions you should be ready to answer
17:36
BAndiT1983
dev__, we need extra meeting, as i need more details, which go in-depth and not that shallow
17:36
supraraj
dev__ could you also answer my query please?
17:37
dev__
Aviencode knows about Height, width , frameCount and frameSize and the frameData
17:37
dev__
Nothing else right now
17:37
supraraj
+1 BAndiT1983...
17:38
dev__
OKay BAndiT1983
17:38
supraraj
dev__ I would recommend as have always been... Please keep us posted through mails etc regarding how the development is headed
17:39
supraraj
GSoC is a collaborative effort and everyone needs to be in loop on what and hows :)
17:40
BAndiT1983
dev__, also more IRC present is desirable
17:40
dev__
Yes, I understand , actually whenever i complete something, I post it on trello
17:40
BAndiT1983
my job is taking a lot more time at the moment, nevertheless i try to stay on track about progresses at GSoC
17:40
dev__
understood BAndiT1983
17:40
BAndiT1983
i would suggest a meeting twice a week besides monday one for frame server
17:41
supraraj
Also make sure when you claim that you incorporated oc with aviencode, you make sure that the whole pipeline is good
17:41
supraraj
For example the Avi encoder knows about. The downscaler parameters
17:41
supraraj
Etc etc
17:42
dev__
Yes , I will be sending you a mail for extraa meeting time
17:43
supraraj
Good :)
17:43
BAndiT1983
sorry for interruption se6astian, please continue
17:44
dev__
Whenever I head for some task , I will mail regrading which process i am taking
17:44
dev__
Thanks ,
17:44
se6astian
feel free to do a more indepth chat meeting right after we finished the official part of this irc meeting
17:44
se6astian
considering you are all already here now :)
17:44
se6astian
Y_G please go ahead
17:44
se6astian
your turn
17:45
Y_G
Hi all,
17:45
Y_G
This week I worked on changing the the message packet structure. Message sent to daemon are split into 2 types ,one for uploading LUT_config files ,and one for sending other commands as before(Number of parameters have been increased from 2 to 4.)
17:46
Y_G
I am still in the process for uploading LUT files part (where I am getting error for message being too long.)
17:47
BAndiT1983
i hope it's a general data blob packet, as uploading specifically LUT file would be too specific
17:47
Y_G
In the upcoming week I will try to debug this ,then will add the functionality for i2c communication directly through DaemonCLI for the initial part.
17:47
Y_G
Yes it's a general blob packet
17:48
Y_G
Probably can use it for `mat4_conf` too. As this also takes 20 float parameters
17:49
BAndiT1983
it shoudn't matter what it sends to daemon, as long as it's dissectable there afterwards
17:49
BAndiT1983
have you understood why i've commented on your idea to send the path of the file to daemon?
17:50
Y_G
Yes ,Makes sense
17:51
se6astian
thanks for the report, anything else to add?
17:52
se6astian
otherwise please also stick around for technical discussion with mentor after the meeting concludes
17:52
Y_G
That would be it from my side ,Thanks
17:52
se6astian
thanks
17:52
se6astian
apurvanandan[m]: your turn
17:53
apurvanandan[m]
Hi everyone
17:54
apurvanandan[m]
This week I completed both side of gearwork for BER testing on single LVDS lane, although it had some problem in working on hardware.
17:57
apurvanandan[m]
Lattice Diamond doesn't offer simulator out of the box, this week most of my time got wasted in setting this up. Had to try Altera, Microsemi's IDE etc. Ubuntu packages dependencies was broken. Much time was wasted.
17:58
apurvanandan[m]
But anyways everything has been setup now and I am working on the simulation now.
17:58
dev__
left the channel
17:58
apurvanandan[m]
I will cover up this week.
17:59
se6astian
thanks for the update
17:59
se6astian
anything else?
17:59
apurvanandan[m]
Thats all I have to say.
17:59
apurvanandan[m]
Thanks for the time.
17:59
apurvanandan[m]
Nope
17:59
se6astian
Fares emailed a quick message:
17:59
se6astian
I’m sorry I won’t be able to be in the meeting today.I’m currently out of house as today is my graduation project discussion.
17:59
se6astian
As an update, I implemented AXI lite module for the AXI_hp[reader|writer] as well as some work related to the axi_hp. And I documented some of the modules I previously written. I know this is not much but I was busy in the graduation project and I will work more in the coming weeks.
18:00
se6astian
so that would conclude todays meeting, unless anyone has something else to share/report now
18:00
dev__
joined the channel
18:00
se6astian
not much to share from my side unfortunately
18:03
se6astian
meeting concluded then, many thanks everyone for attending
18:03
se6astian
stick around IRC :)
18:04
supraraj
Thanks se6astian
18:05
dev__
left the channel
18:05
BAndiT1983
Y_G, commented on trello, as it was the opposite of your description
18:05
BAndiT1983
if we would allow daemon to load files, when someone sends the path, it would open a lot of backdoors
18:06
BAndiT1983
proper way is to send the data and unpack it in the according handler
18:07
BAndiT1983
also the data can be stripped to necessary parts, e.g. if we don't need LUT header we could leave it out and only use the rest
18:07
supraraj
dev__ , BAndiT1983 : if the discussion is over regarding dev's project, I would be signing off
18:08
BAndiT1983
supraraj, let us make another meeting this week, tomorrow or wednesday
18:08
BAndiT1983
i want to get more infos, as it's important for stable and flexible base
18:09
supraraj
Whenever you feel comfortable BAndiT1983
18:09
supraraj
However I can't guarantee my connectivity this time... Such a remote place... (Beyond the point really)
18:10
BAndiT1983
argh, dev already signed off
18:10
BAndiT1983
alright, will notify him a bit later, so we can define a time
18:10
supraraj
Not sure if he reads the logs lol
18:10
BAndiT1983
am not sure that there are laready offset maps for AVI file for FUSE
18:10
supraraj
There are none as of now
18:11
supraraj
I searched last year... They were quite complex
18:11
supraraj
We can write ourselves.... But just would need it to be precise
18:11
BAndiT1983
?
18:11
BAndiT1983
what do you mean?
18:12
supraraj
Are offset maps the mapping from incoming byte location to file mapping - I meant this
18:13
BAndiT1983
yes, this is mandatory to be able to read correct locations
18:14
supraraj
That shouldn't be a big trouble
18:14
BAndiT1983
it maybe difficult at first, but the ground work is generating the map of locations of frames in the file, as i would keep the header same
18:14
supraraj
Yeah... Implementation has to be accurate and clean
18:15
supraraj
Should we wait for dev to mail us the next meeting's time and date?
18:16
supraraj
Or should we select one and notify him...
18:17
BAndiT1983
he should tell us if he can tomorrow or on wednesday, this is one ofl imitations, otherwise it would be postponed till next monday, i suppose
18:18
Y_G
Hi BAndiT1983 ,could you elaborate more about `send the data`.How do I send the file data through CLI and hence to daemon
18:19
supraraj
Fine by me :) good night BAndiT1983 Y_G se6astian Bertl_oO
18:19
supraraj
And all others in the room
18:19
supraraj
Signing off
18:19
BAndiT1983
good night supraraj
18:19
Y_G
Good Night supraraj
18:19
supraraj
left the channel
18:20
BAndiT1983
Y_G, you load the file in the CLI app, build a data blob packet with the content and send it to daemon
18:20
Y_G
for loading ,you do require the path ?
18:21
BAndiT1983
yes
18:21
BAndiT1983
but not on the daemon side
18:23
BAndiT1983
this mechanism could also be used to upload the data from some other computer to the camera, e.g. new LUT
18:24
BAndiT1983
so giving daemon a path would not work, the hazards of it were already mentioned on trello
18:25
Y_G
That's what I was doing right now . Send the path through DaemonCLI (ex ./DaemonCLI lut_conf set lutBuffer pathToFile ).Then build the blob in CLI App (Main.cpp of CLI).
18:26
Y_G
Am I missing something here ?
18:26
BAndiT1983
yes, read this and told you a couple of times that this is a bad approach, which violates also the security of things
18:27
BAndiT1983
when oyu upload something to a web server, you are not giving it a path to the file on your computer, but the browser builds the data packet and sends it, also the path is usually omitted,so the server does not know the structure of your machine
18:28
BAndiT1983
same should be done for daemon, things should be sent as data packet and verified on daemon side, splitting between user space and kernel, or whatever, should be prio 1, so there are no back doors or other malicious holes
18:29
illwieckz
left the channel
18:32
BAndiT1983
off for now, will read logs later
18:32
BAndiT1983
changed nick to: BAndiT1983|away
19:18
RexOrCine|away
changed nick to: RexOrCine
19:26
Y_G
left the channel
19:34
Nira
changed nick to: Nira|away
19:38
Nira|away
changed nick to: Nira
19:38
Nira
changed nick to: Nira|away
19:50
Umori
left the channel
19:53
Umori
joined the channel
20:00
BAndiT1983|away
changed nick to: BAndiT1983
20:09
Umori
left the channel
20:14
Umori
joined the channel
20:18
Y_G
joined the channel
20:35
Umori
left the channel
20:37
Umori
joined the channel
20:51
BAndiT1983
changed nick to: BAndiT1983|away
21:14
BAndiT1983|away
changed nick to: BAndiT1983
21:22
BAndiT1983
changed nick to: BAndiT1983|away
21:23
se6astian
off to bed, good night
21:24
se6astian
changed nick to: se6astian|away
21:44
Y_G
left the channel
23:52
illwieckz
joined the channel
00:05
leonardus
joined the channel
00:06
leonardus
Is anyone here familiar with DCP/KDMs https://en.wikipedia.org/wiki/Digital_Cinema_Package or knows of a better place to ask about those?
00:10
RexOrCine
leonardus: I'd be surprised if Bertl hasn't, if you wanna hang on for him to look away from whatever he's doing. Possibly save time by describing what you want to know specifically, though.
00:11
leonardus
I just want to know if it would be viable to write software that can play back DCPs and handle KDMs, or if the major movie distributors only target specific playback devices.
00:12
leonardus
I can't seem to find much official documentation, particularly regarding KDM
00:13
Spirit532
left the channel
00:13
Spirit532
joined the channel
00:16
RexOrCine
I take it you haven't contacted SMPTE yet.
00:16
leonardus
The reason I'm asking is: "Each playback system has its own unique public/private key pair. The private key is never shared and is buried in the playback systems within secure hardware meeting FIPS-140 security standards."
00:16
leonardus
I haven't, RexOrCine
00:18
RexOrCine
... as this may well fall outside of our experience.
00:18
RexOrCine
I see.
00:19
Bertl_oO
I'm pretty sure you need to sign a bunch of NDAs and buy a ton of standards to make heads or tails of it
00:19
RexOrCine
Are you new leonardus? (sorry, I don't check in here too often)
00:19
Bertl_oO
and probably pay some royalties too
00:20
leonardus
RexOrCine: Yeah this is the first I've been in this channel
00:21
RexOrCine
Welcome friend. If you pursue this and you find out would you come back and tell us?
00:21
leonardus
Absolutely
00:22
RexOrCine
I would speak with someone at SMPTE yeah.
00:24
leonardus
Pretty sure it would in fact be difficult to write software without signing a bunch of contracts. Otherwise it sounds like the time limits could be easily circumvented.
00:24
leonardus
e.g. just have a device that spoofs the time.
00:26
RexOrCine
Right.
00:36
felix_
on smpte standerds: at least some standards are also available as itu recommendations; in my experience the latter are the useful ones. not sure though if what you described is also an itu standard
00:38
felix_
from the smpte stuff i mostly looked at the different sdi standards and the smpte stuff lacked some essential information the itu recommendation had; also in general the itu rec is more useful when implementing a standard
00:40
RexOrCine
leonardus: felix_ is the man behind this https://wiki.apertus.org/index.php/SDI_Plugin_Module
00:43
leonardus
looks like the spec is behind a $120 paywall? https://ieeexplore.ieee.org/document/7290381
00:51
Bertl_oO
check out sci-hub
00:55
leonardus
Tried that earlier, it directed me to a PDF with the abstract, a link to purchase the spec, and about 12 pages full of white text that only appears if you select it. The text was just a bunch of random words - maybe to appear in search engine results?
00:57
Bertl_oO
the IEEE page says 'Superseded'
00:59
Bertl_oO
might be interesting to know by what :)