Current Server Time: 09:48 (Central Europe)

#apertus IRC Channel Logs

2020/05/20

Timezone: UTC


01:15
futarisIRCcloud
left the channel
01:24
aombk
left the channel
01:27
aombk
joined the channel
02:32
Spirit532
left the channel
02:33
Spirit532
joined the channel
03:22
Bertl
off to bed now ... have a good one everyone!
03:22
Bertl
changed nick to: Bertl_zZ
04:53
illwieckz
left the channel
04:57
illwieckz
joined the channel
05:16
BAndiT1983|away
changed nick to: BAndiT1983
07:21
se6ast1an
good morning!
07:26
futarisIRCcloud
joined the channel
07:29
BAndiT1983
hi
09:17
RexOrCine
Hi.
09:22
BAndiT1983
hi RexOrCine
09:32
se6ast1an
how are things going
10:00
RexOrCine
Here? Not ideal. Doesn't look like the film is happening. Just been sent screenshots of a conversation I was having with someone about the camera four years ago. The golden days "Just around the corner" etc... awww.
10:18
se6ast1an
what film?
12:13
mumptai
joined the channel
12:38
Bertl_zZ
changed nick to: Bertl
12:38
Bertl
morning folks!
12:41
se6ast1an
good day!
12:43
se6ast1an
alex pushed a few additions to raw2dng last night, I tested the pipe/parameter issue https://lab.apertus.org/T1201 and closed it after confirming the fix
12:43
se6ast1an
maxime and bertrand will create new darkframes for testing
12:58
RexOrCine
(11:18:02) se6ast1an: what film? < There was supposed to be another film happening. Following on from the last one. But because of everything it's not happening at the moment.
12:58
RexOrCine
If it had gone ahead, or if it does, it would be the answer to a lot of problems.
12:59
Bertl
is that so? please elaborate ...
13:11
RexOrCine
Who? Me?
13:11
RexOrCine
Or Seb...
13:25
Bertl
sorry, wrong channel ...
15:38
BAndiT1983
changed nick to: BAndiT1983|away
17:02
BAndiT1983|away
changed nick to: BAndiT1983
17:52
l_
joined the channel
17:53
se6ast1an
lennart?
17:54
l_
yes
17:55
l_
english is the language here right?
17:55
Bertl
yep, welcome!
17:55
l_
hi
17:56
Bertl
we can move to a private channel if you prefer German :)
17:56
l_
(english is fine) first question before we start would be, if I should keep the stock firmware on the axiom for debugging, or would it be better to do it with the latest nightly?
17:56
l_
(the issue appears on both)
17:58
Bertl
let's stick to the stock for now
17:58
l_
ok
17:59
Bertl
so when the camera has completely booted up, you get an image but with bad quality, correct?
18:01
l_
currently I get no image at all (with the shogun flame)
18:02
l_
although the _blue_ led on the axiom flashes (so hdmi is on, If I understand correctly)
18:02
Bertl
well, that mostly means that the sequencer is working
18:02
l_
ok
18:02
Bertl
i.e. frames are captured and an image is sent to the plugin
18:03
Bertl
doesn't say anything about the HDMI output but we can assume that it is on
18:03
l_
when I connect a computer display (instead of the shogun) I get an image
18:04
Bertl
okay, good, so we know HDMI is there, but probably with bad settings
18:04
l_
ok
18:04
Bertl
can you logon to the camera as root and try a few commands for us?
18:04
l_
yes
18:04
l_
already on the camera
18:05
l_
up to now I just booted and ran ps aux
18:05
Bertl
good, so let's do: i2cdetect -y -a -r 2
18:06
Bertl
please paste the output on some pastebin and the link here
18:08
l_
https://pastebin.pl/view/87bf7f99
18:08
l_
it seems empty...
18:09
Bertl
okay, do you know your power board version?
18:10
Bertl
if you look at the image here: https://wiki.apertus.org/images/7/76/ABCS_00-PB-002-_ABDK_Power_Board_V0.30R1.0_Show_V2_1150.jpg
18:10
Bertl
the version is right besides the SD slot in the rounded square
18:10
Bertl
it should be relatively easy to see from the side
18:13
l_
exactly as on the picture, v0.30 R1.0
18:14
Bertl
perfect, so that is normal then
18:15
Bertl
let's try the following command:
18:15
Bertl
./rf_sel.py A
18:15
l_
[root@beta ~]# ./rf_sel.py A
18:15
l_
assuming power board version 0.30
18:15
l_
selecting RFW [bus A] ...
18:16
Bertl
and now the i2cdetect line again
18:16
l_
it is unchanged
18:16
Bertl
that is a problem
18:16
Bertl
let's see what we get on the other side with:
18:16
Bertl
./rf_sel.py B
18:17
Bertl
and then the i2cdetect line
18:17
l_
the table is also unchanged
18:18
Bertl
okay, so it looks like your secondary I2C bus is not working as expected
18:18
l_
ok
18:19
Bertl
let me think what we could test to see what's the problem there
18:20
Bertl
did the camera auto start when you powered it on?
18:20
Bertl
or did you run anything manually?
18:22
l_
it autostarted
18:22
Bertl
okay, let's run:
18:22
Bertl
systemctl disable cmv12k.service
18:23
Bertl
this should stop the auto start next time you power on
18:23
Bertl
but you will still be able to logon as usual
18:23
l_
shall I reboot?
18:23
Bertl
power down the beta (shutdown -h now)
18:24
Bertl
then remove and reapply power
18:24
l_
the blue led is still flashing, that is ok?
18:24
Bertl
yep, that's fine
18:24
l_
ok
18:25
Bertl
can you easily losen up the side where the microzed is connected?
18:25
Bertl
should be four nuts, two on the plugin holder, two on the microzed itself
18:25
l_
right, yes I could do that.
18:26
l_
should I do it while the beta is running?
18:26
Bertl
no, we do a test now, and if that fails, power down and losen them up and briefly disconnect and reconnect the microzed
18:27
Bertl
for the test, you run:
18:27
l_
ok
18:27
Bertl
./prep_icsp.sh
18:27
Bertl
(again pastebin the output)
18:27
Bertl
and then:
18:28
Bertl
./rf_sel.py A
18:28
Bertl
i2cdetect -y -a -r 2
18:28
Bertl
we are only interested in this output if it isn't all zeros
18:28
l_
https://pastebin.pl/view/3578fe4b
18:29
Bertl
then after that:
18:29
Bertl
./icsp_id.py A
18:29
l_
https://pastebin.pl/view/bce8e86e (first time i2cdetect shows something)
18:30
Bertl
okay, that is looking good
18:30
l_
https://pastebin.pl/view/bbdfbc5f
18:30
Bertl
that too
18:30
Bertl
let's see if the other side works as well
18:30
Bertl
./rf_sel.py B
18:30
Bertl
i2cdetect -y -a -r 2
18:30
Bertl
./icsp_id.py B
18:32
l_
https://pastebin.pl/view/a049beaa
18:32
l_
I put all the three command responses into one pastebin
18:33
Bertl
perfect
18:33
Bertl
so the hardware seems to be fine here
18:34
Bertl
let me double check the prep output ...
18:35
Bertl
yep, all looking good
18:36
Bertl
let's run:
18:36
Bertl
./kick_manual.sh
18:36
Bertl
and again, pastebin the output
18:36
l_
does it matter whether a computer display or the shogun is connected?
18:36
Bertl
no, doesn't matter
18:36
l_
ok
18:37
l_
https://pastebin.pl/view/beff8e3a
18:38
Bertl
okay, I presume that:
18:38
Bertl
./rf_sel.py A
18:39
Bertl
i2cdetect -y -a -r 2
18:39
Bertl
now shows all zeroes again, yes?
18:39
l_
correct
18:39
Bertl
okay, please upload the kick_manual.sh somewhere
18:41
Bertl
then, while I check it, please run the following:
18:41
Bertl
./halt_manual.sh
18:41
Bertl
./prep_icsp.sh
18:41
l_
https://gist.github.com/lennart/8a4d2ae0a61c3d69239259e5396cfff8
18:41
Bertl
i2cdetect -y -a -r 2
18:42
l_
(still zeroes)
18:43
Bertl
hmm, that's interesting
18:44
Bertl
I do not have the 'i2c0_bit_set 0x22 0x15 7' in my kick_manual.sh here
18:44
Bertl
let's try the following:
18:44
Bertl
./gpio.py init
18:45
Bertl
./prep_icsp.sh
18:45
Bertl
./rf_sel.py A
18:45
Bertl
i2cdetect -y -a -r 2
18:45
Bertl
do we get addresses?
18:45
l_
yes
18:46
Bertl
let's comment out line 35 in your kick_manual.sh
18:46
Bertl
i.e. just put a '#' in front of the command
18:46
Bertl
you know how to edit this file?
18:46
l_
ok, done
18:46
Bertl
perfect
18:47
Bertl
now let's run the ./kick_manual.sh again
18:47
Bertl
only interesting part is the last few lines
18:48
Bertl
s/is/are/
18:48
l_
https://pastebin.pl/view/5a3ac84e
18:49
l_
no error, enough lines in the pastebin?
18:49
Bertl
that looks good now
18:49
Bertl
perfect
18:49
l_
I do get an image on the computer display but none on the shogun
18:49
Bertl
check the HDMI image for quality
18:49
Bertl
if I remember correctly, you had intermittent image on the computer display as well
18:50
Bertl
the shogun might require a specific setup to work
18:50
l_
yes, I had, but recently these issues did not appear. It seems as if the image now runs at higher framerate (it feels snappier)
18:50
l_
also, exposure seems correct now (before it was quite overexposed)
18:51
Bertl
interesting ...
18:51
Bertl
well, I still wonder where this one line came from
18:51
Bertl
anyway, we found that and fixed that, so we can move on to the other problems :)
18:52
Bertl
what resolutions does the shogun you use support?
18:54
l_
it is a 4k variant, I try to understand which resolutions it should support (reading: https://downloads.atomos.com/shogun-flame/Shogun_Flame_User_Manual.pdf right now)
18:55
Bertl
does your computer screen report what it receives?
18:55
Bertl
should be 1080p60 by default, but that might have been adjusted by you
18:57
l_
I get resolution and hz (1920x1080 54.1khz 48hz) not sure if this is helpful
18:57
Bertl
48Hz that is interesting
18:57
Bertl
let's do a few checks on the config then
18:57
Bertl
(from the commandline :)
18:58
l_
ok
18:58
Bertl
grep ^./gen_init setup.sh
18:58
Bertl
ll cmv_hdmi3.bit
18:58
l_
ah that's interesting: ./gen_init.sh SHOGUNp24
18:59
l_
I remember talking about the SHOGUNp24 option
18:59
l_
lrwxrwxrwx 1 root root 26 Oct 11 2018 cmv_hdmi3.bit -> /opt/BITS/cmv_hdmi3_60.bit
19:00
Bertl
okay, let's edit the setup.sh and replace the SHOGUNp24 with 1080p60
19:00
Bertl
you can copy the entire line, comment one and change the other
19:00
Bertl
(maybe the 1080p60 is already there as well and commented out)
19:01
Bertl
then run ./halt_manual.sh
19:01
Bertl
and once that finishes, ./kick_manual.sh again
19:02
l_
I get signal on the SHOGUN but it flickers
19:02
l_
shall I try the SHOGUN option?
19:03
l_
(it is a different flickering than the one I experienced before!)
19:03
Bertl
yes, give that a try
19:04
l_
works
19:04
Bertl
great! so everything fine now?
19:05
l_
I think so. last question: is the thin white border around the image an overlay? (I see the axiom logo and this border)
19:05
se6ast1an
yes
19:05
l_
ok, fine
19:05
Bertl
you can change or remove it easily
19:05
se6ast1an
https://wiki.apertus.org/index.php/AXIOM_Beta/Manual#Image_Overlays
19:06
se6ast1an
clear overlay:
19:06
se6ast1an
./mimg -a -o -P 0
19:06
l_
thank you
19:06
l_
ok, this also works
19:06
se6ast1an
you can also add that line to the startup script if you want to make it permanent
19:06
se6ast1an
in kick_manual.sh
19:06
l_
ok, regarding startup: shall I reenable the cmv service we disabled earlier?
19:06
Bertl
yes, but best comment out/replace the existing one instead of just adding
19:07
Bertl
(the overlay)
19:07
Bertl
regarding the service, yes, you can enable it with
19:07
Bertl
systemctl enable cmv12k.service
19:08
l_
ok. so in order to check if everything comes up again with working image, can I reboot the system, or is it better to shutdown, and power cycle?
19:08
Bertl
shutdown -r now is fine
19:09
l_
ok great, video came up again after reboot
19:10
l_
sometimes the image "jumps" slightly though this doesn;t interrupt recording via shogun. I'll observe how this behaves...but for now this is already a vast improvement
19:12
Bertl
the shoguns are somewhat picky on the input signal
19:12
Bertl
might be that your model (which we haven't tested yet) requires some tuning
19:13
l_
ok, is there some kind of documentation on how to tune? or is this kind of trial and error?
19:14
Bertl
as the shogun is a proprietary device, we have no feedback from that end except for the test results
19:14
se6ast1an
note that your current exposure time setting might be too long to reach the full FPS of the HDMI stream
19:15
se6ast1an
you can try if you see a difference:
19:15
se6ast1an
./cmv_snap3 -z -e 5ms
19:15
l_
ok
19:19
l_
it _seems_ more stable with 5ms
19:19
l_
and yes I can notice a difference
19:20
se6ast1an
I think the "border" is around 13ms IIRC
19:20
se6ast1an
you should be able to see it easily
19:20
Bertl
11.8
19:21
se6ast1an
ah, perfect
19:21
Bertl
11.7 should be still fine
19:23
l_
ok. thanks for the support. I'll test the different values.
19:24
Bertl
you're welcome!
19:24
se6ast1an
You are welcome!
19:24
se6ast1an
:)
19:41
l_
left the channel
19:45
l
joined the channel
19:46
l
changed nick to: Guest44100
19:47
Guest44100
left the channel
22:26
megora
joined the channel
22:39
BAndiT1983
changed nick to: BAndiT1983|away
23:05
megora
left the channel
23:31
mumptai
left the channel