Slideshows don't work in Slovenia and Slovakia #21
Comments
|
BTW in welle.io the (Slovene) slides work as well. |
|
Hi Andreas
I'll look into it, You know of course that welle.io uses the full qt-dab
dab engine (It basically provides a different interface and I noticed that
the sound out is different. The dab decoding uses the qt-dab files) I had a
quick look and he uses the pad and mot handling from dablin. (welle.io does
not handle dab data, only program associated date).
I ran a quick test on the file you sent me a couple of weeks ago, and I
found this. I know there are problems under Windows, that
has to do with the Qt engines.
Anyway, if you can provide me with some large files with raw data where
qt-dab cannot extract slides while the dablin software can, it will be
really helpful to address this issue
best
jan
2017-03-21 13:07 GMT+01:00 andimik <notifications@github.com>:
… BTW in welle.io the (Slovene) slides work as well.
—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
<#21 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AITzwEyxVPlkfEvJNIMugN77lTEsRH0Tks5rn717gaJpZM4MUjwk>
.
--
Jan van Katwijk
+31 (0)15 3698980
+31 (0) 628260355
|
|
Please see AlbrechtL/welle.io#30 (comment) In this raw file to be found in http://matkaboska.sytes.net/s/tvascXIUIsC8UBe you should see slideshows on two services: on SId 0x2248, 'RTE GOLD' (subchannel 9) and on SId 0x2202, 'RTE 2FM' (subchannel 2) They only appear in dablin_gtk (after I have exported the ETI frames with eti-frontend) and in XpadXpert. Welle.io, qt-dab and eti-backend do not show them ... |
|
Hi Andreas
That is correct, these data are embodied in plain DAB files and - up to now
- I did not feel the urge to add the pad handling to plain DAB (I always
figured that plain DAB was dead). That welle and the current eti-backend do
not show the pictures is obviously since that that is all the same MP2
code. If you insist I'll look into it next week (seems you are fond on
giving homework)😀
best
jan
2017-03-24 18:46 GMT+01:00 andimik <notifications@github.com>:
… Please see AlbrechtL/welle.io#30 (comment)
<AlbrechtL/welle.io#30 (comment)>
In this raw file you should see slideshow on 2 services:
on SId 0x2248, 'RTE GOLD' (subchannel 9)
[image: mot_body_7915_bytes bin]
<https://cloud.githubusercontent.com/assets/24510556/24306608/6e496622-10c1-11e7-8097-53f8c3c646cf.jpg>
and on SId 0x2202, 'RTE 2FM' (subchannel 2)
[image: slide2509]
<https://cloud.githubusercontent.com/assets/24510556/24306683/b3fb831c-10c1-11e7-96f6-66bb42002391.jpg>
They only appear in dablin_gtk (after I have exported the ETI frames with
eti-frontend) and in XpadXpert.
Welle.io, qt-dab and eti-backend do not show them ...
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#21 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AITzwIQNZWxO73kyC9Xdk9ZvtAA5a8Ozks5rpAF_gaJpZM4MUjwk>
.
--
Jan van Katwijk
+31 (0)15 3698980
+31 (0) 628260355
|
|
Now it's a bit better. Some stations now show slideshows. Playing the Irish raw file I got it for 2FM:
But still no slideshow on RTE GOLD:
Same in Slovenia, Slideshow not shown (see screenshot above, this is the same file I investigated) here:
but success here:
|
|
I know it is not perfect yet. There is one case where the incomping
packages fail on the crc test. If I skip the crc test you see most of the
picture, it is the very last segment that cripples the data.
To be continued this weekend
best
jan
2017-03-30 0:15 GMT+02:00 andimik <notifications@github.com>:
… Now it's a bit better. Some stations now show slideshows. Playing the
Irish raw file I got it for 2FM:
we found for RTE 2FM serviceId 2202
we have now DAB
going to show picture slide2509.jpg
But still no slideshow on RTE GOLD:
we found for RTE GOLD serviceId 2248
we have now DAB
End of file, restarting
End of file, restarting
End of file, restarting
Same in Slovenia, Slideshow not shown (see screenshot above, this is the
same file I investigated) here:
we found for Slo Int Radio SI serviceId 933A
we have now DAB+
End of file, restarting
but success here:
we found for Radio Ognjisce serviceId 942F
we have now DAB+
going to show picture
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#21 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AITzwOrKC1EnoJ2v75XnpoSXGM0fXUrSks5rqtgdgaJpZM4MUjwk>
.
--
Jan van Katwijk
+31 (0)15 3698980
+31 (0) 628260355
|
|
Regarding RTE GOLD it looks like that between the MOT Header (with correct CRC) and a MOT Body repetition (with correct CRC) there are further transmissions of the same MOT Headers/Bodys, but all with wrong CRC due to e.g. missing AUs. Maybe this is the reason here. |
|
Wrt RTE GOLD, reading the file (and continuing after a failing crc), this
happens
we have now DAB
crc failed grouptype 4, segmentNumber 0, lastFlag 1
crc failed grouptype 3, segmentNumber 0, lastFlag 1
crc failed grouptype 4, segmentNumber 0, lastFlag 1
going to show picture slide11.jpg�
Corrupt JPEG data: premature end of data segment
cannot write file slide11.jpg�
crc failed grouptype 3, segmentNumber 0, lastFlag 1
crc failed grouptype 4, segmentNumber 0, lastFlag 1
crc success grouptype 3, segmentNumber 0, lastFlag 1
ALSA lib pcm.c:7963:(snd_pcm_recover) underrun occurred
End of file, restarting
As can be seen from the partial picture, the last part of the picture is
corrupted. Note that the picture is encoded in a
single msc group (so the complete picture and description is in a single
group3 group4 combination), It is not shown here,
but the size of the container for the group 4 block is OK, slightly larger
than the size of the block.
2017-03-31 23:50 GMT+02:00 Stefan Pöschel <notifications@github.com>:
… Regarding RTE GOLD it looks like that between the MOT Header (with correct
CRC) and a MOT Body repetition (with correct CRC) there are further
transmissions of the same MOT Headers/Bodys, but all with wrong CRC due to
e.g. missing AUs. Maybe this is the reason here.
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#21 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AITzwHA6p6-coFBJRLFRAj0JwChJv7UYks5rrXUlgaJpZM4MUjwk>
.
--
Jan van Katwijk
+31 (0)15 3698980
+31 (0) 628260355
|
|
Also when the CRC is checked and failing DGs are discarded, (exactly) one picture should be displayed. I just took a look into the file and it happens that here the MOT body is transmitted (uncorrupted) before the MOT header: |
|
Newer versions show them, but it's still not running smoothly. Anyway I'll close it but maybe basicmaster could help you in this matter. |
Closed
|
Jan, could you try to improve the code for MOT handling according to @basicmaster 's proposals. The MOT icon is very often "blinking" and slideshows are not shown or it takes very, very long. All my other tools show slideshows when the file has been received. Maybe you could use code from welle.io? |
|
I'll look into it, right now I am trying to figure out how a hackrf is to
be interfaced
So, it might be next week
2018-05-15 21:10 GMT+02:00 andimik <notifications@github.com>:
… Jan, could you try to improve the code for MOT handling according to
@basicmaster <https://github.com/basicmaster> 's proposals.
#21 (comment)
<#21 (comment)>
#21 (comment)
<#21 (comment)>
The MOT icon is very often "blinking" and slideshows are not shown or it
takes very, very long. All my other tools show slideshows when the file has
been received.
Maybe you could use code from welle.io?
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#21 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AITzwKnFSmxWCQ0M2BqS0b3wsyyD0DIkks5tyyhBgaJpZM4MUjwk>
.
--
Jan van Katwijk
+31 (0)15 3698980
+31 (0) 628260355
|
|
With 1874dd9 it's the same, still not all slides are shown. I've tried it several times with the same rawfile (in my case one of the radio programs has 2 alternating slides), sometimes the slides appear, but in lots of the cases they still don't and MOT is blinking like before. |
|
can you send me a link to that file then I can have a look.
The blinking is because each time a new message starts I switch it off, and
only switch it
on when everything is/seems OK
wrt the naming: I think it is OK now
2018-05-26 13:42 GMT+02:00 andimik <notifications@github.com>:
… With 1874dd9
<1874dd9>
it's the same, still not all slides are shown.
I've tried it several times with the same rawfile (in my case one of the
radio programs has 2 alternating slides), sometimes the slides appear, but
in lots of the cases they still don't and MOT is blinking like before.
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#21 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AITzwD28Xq8l_YwsN-uwJEL2YrYW-NPbks5t2T-4gaJpZM4MUjwk>
.
--
Jan van Katwijk
+31 (0)15 3698980
+31 (0) 628260355
|
|
If you still have mine from Slovenia, then use |
|
Hi, Lastest version of qt-dab don't display slides from ODR tools, other receivers working fine. Regards |
|
Is the picture written onto a file?
2018-08-17 8:40 GMT+02:00 zygmund2000 <notifications@github.com>:
… Hi,
Lastest version of qt-dab don't display slides from ODR tools, other
receivers working fine.
Regards
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#21 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AITzwBB2obnWfFjZ1T8W8n5roR5ko3i8ks5uRmV7gaJpZM4MUjwk>
.
--
Jan van Katwijk
+31 (0)15 3698980
+31 (0) 628260355
|
|
Which OS? |
|
It is Linux
I just ran a quick test and notices that a version from may showed a
picture and the current one does not
The interesting this is that the saved versions of the pictures, generated
with the two instances, are exactly the same (so differences when comparing)
The code that should display the picture is also the same in the may and
current version, so I have to puzzle a little
2018-08-17 10:59 GMT+02:00 andimik <notifications@github.com>:
… Which OS?
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#21 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AITzwLoOsGQs1agZfZGPHZmX4fRhFZMfks5uRoXUgaJpZM4MUjwk>
.
--
Jan van Katwijk
+31 (0)15 3698980
+31 (0) 628260355
|
|
I did not look at windows yet, my windows system is a mess
For Linux:
if/when you use the Qt-DAB appImage then the pictures do not show, if you
compile your own version they do
The AppImage is made in an old Ubuntu environment. Why? it does contain
most of the libraries needed, however
it depends on some generally available libraries and to prevent that the
AppImage has to rely on older "general"
libraries, it is built on a pretty old system.
2018-08-17 11:42 GMT+02:00 jan van katwijk <j.vankatwijk@gmail.com>:
… It is Linux
I just ran a quick test and notices that a version from may showed a
picture and the current one does not
The interesting this is that the saved versions of the pictures, generated
with the two instances, are exactly the same (so differences when comparing)
The code that should display the picture is also the same in the may and
current version, so I have to puzzle a little
2018-08-17 10:59 GMT+02:00 andimik ***@***.***>:
> Which OS?
>
> —
> You are receiving this because you commented.
> Reply to this email directly, view it on GitHub
> <#21 (comment)>,
> or mute the thread
> <https://github.com/notifications/unsubscribe-auth/AITzwLoOsGQs1agZfZGPHZmX4fRhFZMfks5uRoXUgaJpZM4MUjwk>
> .
>
--
Jan van Katwijk
+31 (0)15 3698980
+31 (0) 628260355
--
Jan van Katwijk
+31 (0)15 3698980
+31 (0) 628260355
|
|
Could you check if jpg and png behavior the same? |
|
I don's see pictures in specified directory picturesPath=/home/user |
|
The default is /tmp/qt-pictures
Btw, I assume you use a Linux version
2018-08-17 16:32 GMT+02:00 zygmund2000 <notifications@github.com>:
… I don's see pictures in specified directory picturesPath=/home/user
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#21 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AITzwEul5rwVTPdiEvLOH_Q3-P1kA1-Pks5uRtQGgaJpZM4MUjwk>
.
--
Jan van Katwijk
+31 (0)15 3698980
+31 (0) 628260355
|
|
Yes Linux version (Manjaro), I changed default to /home/user but /tmp/qt-pictures also no images. |
|
did you compile yourself or did you use the appimage
is manjaro ubuntu based?
2018-08-17 22:24 GMT+02:00 zygmund2000 <notifications@github.com>:
… Yes Linux version (Manjaro), I changed default to /home/user but
/tmp/qt-pictures also no images.
[image: mot]
<https://user-images.githubusercontent.com/14185298/44287213-499d1880-a26c-11e8-8e5e-5bc481dbce90.png>
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#21 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AITzwICInwYz4HTovYxXs0TQnyfW6FOiks5uRyaJgaJpZM4MUjwk>
.
--
Jan van Katwijk
+31 (0)15 3698980
+31 (0) 628260355
|
|
I compiled from git, manjaro is arch based. |
|
Under kubuntu (compiled with qmake and make) they both work: png and jpg |
|
What is on your system the behavious of you use an AppImage (see the
releases section of the github rep)
Op vr 17 aug. 2018 om 22:37 schreef zygmund2000 <notifications@github.com>:
… I compiled from git, manjaro is arch based.
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#21 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AITzwEL_q2_gAmp12X470bIOH9Pmv_-Mks5uRymQgaJpZM4MUjwk>
.
--
Jan van Katwijk
+31 (0)15 3698980
+31 (0) 628260355
|
|
Also no slides, please check it with DAB generated by ODR tools, maybe bug is not in qt-dab. |
|
Could you share an eti file? |
|
I can't confirm that it should be an issue with ODR. I tried various eti streams from satellite and made a raw and opened it in qt-dab. They all work. But, you are right, the initial bug is still unsolved. The MOT is blinking and slides are shown rarely, sometimes after several replays. Basicmaster gave a hint, so Jan could check the code again. |
|
This issue is difficult to track because slides work fine on eg. LG Stylus and welle and dablin but not with qt-dab today, some older versions of qt-dab sometimes show images after long time. |
|
With the recording I have (10D_Slovenia from 2017-12-26) there are no
differences in output
with different qt-dab versions (from april until right now)
It would really be helpful to have some recording that in your context
shows the
difference in behaviour.
Op za 18 aug. 2018 om 17:15 schreef zygmund2000 <notifications@github.com>:
… This issue is difficult to track because slides work fine on eg. LG Stylus
and welle and dablin but not with qt-dab today, some older versions of
qt-dab sometimes show images after long time.
IMHO ODR-padenc has something different than commercial encoders because
this muxes which use it have always images on every my receiver including
qt-dab also.
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#21 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AITzwAOPC9R2S-eFxdiAf2Nnq-k_AZTlks5uSC-MgaJpZM4MUjwk>
.
--
Jan van Katwijk
+31 (0)15 3698980
+31 (0) 628260355
|
|
Well actually this is not completely true, as mentioned earlier, the
AddImage, created on an ubuntu 14 version, does not show some pictures
although they are stored into files.
Op zo 19 aug. 2018 om 21:17 schreef jan van katwijk <j.vankatwijk@gmail.com
…:
With the recording I have (10D_Slovenia from 2017-12-26) there are no
differences in output
with different qt-dab versions (from april until right now)
It would really be helpful to have some recording that in your context
shows the
difference in behaviour.
Op za 18 aug. 2018 om 17:15 schreef zygmund2000 ***@***.***
>:
> This issue is difficult to track because slides work fine on eg. LG
> Stylus and welle and dablin but not with qt-dab today, some older versions
> of qt-dab sometimes show images after long time.
> IMHO ODR-padenc has something different than commercial encoders because
> this muxes which use it have always images on every my receiver including
> qt-dab also.
>
> —
> You are receiving this because you commented.
> Reply to this email directly, view it on GitHub
> <#21 (comment)>,
> or mute the thread
> <https://github.com/notifications/unsubscribe-auth/AITzwAOPC9R2S-eFxdiAf2Nnq-k_AZTlks5uSC-MgaJpZM4MUjwk>
> .
>
--
Jan van Katwijk
+31 (0)15 3698980
+31 (0) 628260355
--
Jan van Katwijk
+31 (0)15 3698980
+31 (0) 628260355
|
|
I'll try to save eti file or I can share as live stream if you wish |
|
For me easiest would be to get a pointer to a raw input file, then I can
trace all steps and see where things go wrong
Op zo 19 aug. 2018 om 22:27 schreef zygmund2000 <notifications@github.com>:
… I'll try to save eti file or I can share as live stream if you wish
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#21 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AITzwDiR7ZEHSH-r5IVsbcJoO6aQiPhLks5uScpOgaJpZM4MUjwk>
.
--
Jan van Katwijk
+31 (0)15 3698980
+31 (0) 628260355
|
|
raw IQ from air? |
|
Yes, just a raw input file (either 8 bits raw or made by qt-dab, a wav file)
In the mean time I found one thing that might have caused a problem, it is
corrected and put in the repository,
Op ma 20 aug. 2018 om 15:24 schreef zygmund2000 <notifications@github.com>:
… raw IQ from air?
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#21 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AITzwN_oVljIEvR5U6LgZY0YJNF2H3cTks5uSrimgaJpZM4MUjwk>
.
--
Jan van Katwijk
+31 (0)15 3698980
+31 (0) 628260355
|
|
Still nothing, I sent IQ to email. |
|
Thanks,
that is really helpful
I think I have addressed the issue, it is on github. Now I will do some
further checking that
no other issues are introduced
best
jan
Op ma 20 aug. 2018 om 17:33 schreef zygmund2000 <notifications@github.com>:
… Still nothing, I sent IQ to email.
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#21 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AITzwMpw5WrJxYkQK8GUIIGbaNwZwJJqks5uStbOgaJpZM4MUjwk>
.
--
Jan van Katwijk
+31 (0)15 3698980
+31 (0) 628260355
|
|
Now I see slides, can you explain why others muxes work before this change? |
|
That is pretty complex. It has to do with detection of erroneous frames. It
turned out that in some cases an msc datagroup was erroneously
rejected. That followed from the handling of an error that occurred with an
English DAB service, I then strengthened the test to
check whether or not a datagroup was (supposed to be) legal
Op ma 20 aug. 2018 om 21:38 schreef zygmund2000 <notifications@github.com>:
… Now I see slides, can you explain why others muxes work before this change?
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#21 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AITzwHYG9wsEJ1yvc9TXwnWNgMC6xvudks5uSxA0gaJpZM4MUjwk>
.
--
Jan van Katwijk
+31 (0)15 3698980
+31 (0) 628260355
|
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
I found out that the slideshows in qt-dab and eti-stuff don't work at all in Slovenija and in Slovakia, whereas dablin_gtk or XPADxpert (and the DABPlayer from Andreas Gsinn under Windows) show them.
BTW: In Vienna it worked under qt-dab well on the same Laptop, see screenshot: .
According to XPADxpert it's in the same subchannel.
See the (zipped) eti-file http://andimik.bplaced.net/temp/10d_lang2.eti.zip , please load in eti-backend, tune to the English speaking "Slo Int Radio SI" (other programs didn't have slideshows at that time and the sample is too short).
The text was updated successfully, but these errors were encountered: