214 replies to this topic

[WanWizard]

Is it possible to add youtube tv plugin same as openatv ?

 

If this is about the VU+ Chromium plugin, the answer is no, it crashes and leaves the framebuffer detached. Which is why VU+ hasn't enabled it in the BSP. VU+ has to fix this.

[Erik Slagter]

 

Hi Team, setting up 7.1 dated 20190916 with a brand new VuDuo4K gives a reduced cropped video image in the top left corner, about 1/9 of the screen dimension, right after you open any menu, and it does not come back to normal. Is this something known of? Any workaround available ? 
 
Cheers

Resolution problem. Get back to 1080P if you don’t have a 4K tv
If you have a 4K tv the resolution must be 2160P

 

VU+ does not support any other frame rate than 25 (interlaced) or 50 (progressive) Hz. Normally you would not need anything else.

[cmatte]

Hi Team, setting up 7.1 dated 20190916 with a brand new VuDuo4K gives a reduced cropped video image in the top left corner, about 1/9 of the screen dimension, right after you open any menu, and it does not come back to normal. Is this something known of? Any workaround available ?

Cheers

Resolution problem. Get back to 1080P if you don’t have a 4K tv
If you have a 4K tv the resolution must be 2160P

VU+ does not support any other frame rate than 25 (interlaced) or 50 (progressive) Hz. Normally you would not need anything else.

I imagine 2160p30 shall not be available then, not sure why it is?

I can also confirm it was not there in the original image, but I was tempted to choose it on OpenPLI and that generated the issue as unsupported.

Sent from my GM1913 using Tapatalk

[WanWizard]

If it is there the drivers advertise it as a valid choice. Chances are the VU+ Image has a hardcoded workaround (instead of fixing the drivers).

[Erik Slagter]

The hardware can do it. The drivers advertise it as a valid choice but the it's not fully implemented, so it doesn't work.

 

I can't see a useful goal for using a 25 or 30 fps output, especially for UHD, so for that, I can go with the reasoning of VU+. The only situation where you'd want to use it, is when you connect a UHD device that does not support HDMI >= 2.0. But in that case, you're f*d anyway, you really don't want to watch UHD content at half the frame rate, it will be jerky.

[dudule22]

Opening .ts files.

 

It works fine with the mediaplayer on file m3u with ts lines inside but it doesn't work on channels list.

 

In the channels list, i have the good format.

 

#DESCRIPTION : Toto
#SERVICE 1:0:1:0:0:0:0:0:0:0:http%3a//toto.com%3a80//576:Toto
 

[WanWizard]

http%3a//toto.com%3a80//576:Toto = http://toto.com:80//576

 

Apart from the double slash that shouldn't be there, that URL doesn't work here at all, it just opens a japanese website?

[dudule22]

toto is just an example, it is not an url.

 

But what is the difference between the mediaplayer and the player in the channels list.

 

The mediaplayer it is better to play ts files

[WanWizard]

In terms of processing, there is no difference.

 

The problem you may have is that playing them with a "SERVICE 1:0:1:0..." definition means the box will process the input in hardware. That expects realtime data (as it is made for broadcasts, tuner input), there is absolutely no buffering, so any hickup in the delivery of data (for example due to internet latency, packet drops, etc) will cause the image to pixelate or freeze.

 

There are some buffering options available, and it may also help to switch to gstreamer (4097 service), but that is not my area of expertise...

[dudule22]

Thanks for your response, it is better with the service 4097, more buffering probably and I can have EPG, but i lost the picons.

 

If someone knowns a solution, it will be nice.

[WanWizard]

What's the point of having a picon for a TS file? Since you talk about EPG as well, my guess is this isn't a TS file at all?

[dudule22]

Not quite

 

I install ServiceApp and i use 5002 service.

I fix picons, it was a channel name problem.

[cmatte]

Hey, it seems the vu duo 4k is randomly switching on, you could find it tuned to a channel at random times. Troubleshooting both crossepg and epgrefresh, there seems to be nothing wrong with them. Instead, the logs show OOM-killer runs as it is out of memory and it regularly decides to kill enigma2. That's surprising as the box specs 2gb of ram, even more surprising, though, to see in WebIF 225176 KB free / 567744 KB total...how comes there is only 554MB total memory reported? Is this normal or is this an hardware issue maybe? Or is there an issue for the box to manage 100Mb+ of EPG?

Hope you can shed some light

Edited by cmatte, 1 October 2019 - 21:56.

[betacentauri]

I think it's "normal" as drivers, framebuffer, transcoding,... eats much RAM.

More than 100MB EPG data is quite much. All EPG data is stored in the RAM. Try to reduce EPG data and see whether it solves the problem.

[cmatte]

Thanks, in fact the mem usage is layed down over the boot procedure and it seems it is split like this:
6M kernel
0.3M rwdata
2M rodata
0.3M init
0.2M bss
1494M reserved by bmem (drivers?)
16M reserved by cma
...turns out to be the ~550M available out of 2048. To be fair this is quite disappointing as you believe you've got at least half the mem spec ed available to you, in general...a fourth of it is quite uncommon! Is this definitely the expected use of it?

Anyways, reducing epg is the way to go as it's not really needed, but still, could be far from fixing it. Another solution is swapping some, but I guess it'd be detrimental unless you use an extremely performing USB key. Any feedback on this route?

Edited by cmatte, 1 October 2019 - 22:18.

[betacentauri]

Yes, I was also surprised that the box uses to much RAM. But we can’t change it and we also don’t know whether so much RAM for the drivers is needed. Only VU knows it and can fix it.
Just try to add swap. Don’t think you need a very fast USB key as hopefully not so many memory is used. I guess 500 MB swap should be more than sufficient. But it’s also only a workaround like using less EPG.

[cmatte]

Thanks a lot for the input! EPG sources for CrossEPG and channels to refresh for EPGRefresh shall be a decent number, now There are a few doubts still floating around.

1. First, just wondering why OOM happens at all: free RAM is around 300M without loading EPG. The current epg.dat is around 20M, CrossEPG reports the whole DB to be below 50M, when loaded the free RAM drops to 180M and that's fine. Then, why would the RAM be randomly used so much to then just finish up? It's true that EPGRefresh and CrossEPG are refreshing in the background, but you would not expect any of that to triple its usage and get to an OOM as they both just, literally, refresh something already there, mainly! However, that could be wrong, that brings up point 2.
2. Wondering other causes, is it correct to have the same epg.dat for CrossEPG, EPGRefresh, and set in the enigma2 settings? If not, shall they all be separate? I'm not sure how this works overall, are they all adding/refreshing data there or not really, and then adding or refreshing the in-memory EPG for enigma2?
3. A swapfile works when activated by the swapon cmd, live, however it fails to be reactivated after a reboot, regardless attempting to add in /etc/fstab any of the following (tried separately). The system tools plugin also does not work, saying it can't find USB, possibly because the mount was made all capitals in the first place by the system, and has been set like that a bit everywhere! Is there anything else that could be done to ensure it's re-loaded at boot?

Cheers

/media/USB/swapfile	none	swap	sw	0	0

/media/USB/swapfile	swap	swap	defaults	0	0

Edited by cmatte, 2 October 2019 - 00:11.

[cmatte]

1. [...]
2. [...]
3. A swapfile works when activated by the swapon cmd, live, however it fails to be reactivated after a reboot, regardless attempting to add in /etc/fstab any of the following (tried separately). The system tools plugin also does not work, saying it can't find USB, possibly because the mount was made all capitals in the first place by the system, and has been set like that a bit everywhere! Is there anything else that could be done to ensure it's re-loaded at boot?

Cheers

/media/USB/swapfile	none	swap	sw	0	0

/media/USB/swapfile	swap	swap	defaults	0	0

 

It seems everything is fine from a configuration point of view, however a further swapon command is needed to activate the swap.

Troubleshooting this one, the /etc/rcS.d/S03mountall.sh startup script recalls /etc/init.d/mountall.sh which in turn attempts to "swapon -a" by the following code. However, it seems for some reason this is not after the usb is mounted (it is a fat fs, set as auto in /etc/fstab), so it fails to work...or at least this the best guess. Any other take on this?

#!/bin/sh

### BEGIN INIT INFO

# Provides:          mountall

# Required-Start:    mountvirtfs

# Required-Stop: 

# Default-Start:     S

# Default-Stop:

# Short-Description: Mount all filesystems.

# Description:

### END INIT INFO

 

. /etc/default/rcS

 

#

# Mount local filesystems in /etc/fstab. For some reason, people

# might want to mount "proc" several times, and mount -v complains

# about this. So we mount "proc" filesystems without -v.

#

test "$VERBOSE" != no && echo "Mounting local filesystems..."

mount -at nonfs,nosmbfs,noncpfs 2>/dev/null

 

#

# We might have mounted something over /dev, see if /dev/initctl is there.

#

if test ! -p /dev/initctl

then

rm -f /dev/initctl

mknod -m 600 /dev/initctl p

fi

kill -USR1 1

 

#

# Execute swapon command again, in case we want to swap to

# a file on a now mounted filesystem.

#

[ -x /sbin/swapon ] && swapon -a

 

: exit 0

 

In case this helps anyone else, in the meantime creating a further S99mountswap.sh -> ../init.d/mountswap.sh startup script that launches a delayed swapon -a works:

#!/bin/sh

### BEGIN INIT INFO

# Provides:          mountswap

# Required-Start:    mountall

# Required-Stop: 

# Default-Start:     S

# Default-Stop:

# Short-Description: Delayed mount swap

### END INIT INFO

 

. /etc/default/rcS

 

#

# Execute swapon command again, in case we want to swap to

# a file on a now mounted filesystem.

#

[ -x /sbin/swapon ] && sleep 2 && swapon -a

 

: exit 0

Edited by cmatte, 2 October 2019 - 02:53.

[twol]

Deleting all the unused languages also saves a reasonable amount of MB's if you are that short

[serdeliuk]

Somebody should change the thread title because -»- truncate the link on the notification email and this is annoying