Tvheadend

Ron L wrote:

Yes, that was it. Only sends gotox when switching to a channel on a different sat now. The grace period seems to be calculating very realistic wait times now too. I think you have a winner

Great.

BTW: Another topic about the rate configs:

For USALS: the rate is in "Rate (millis/deg)" which is implemented 1/1000 sec resolution per one degree.
For GOTOX: the rate is in "Rate (millis/click)" which is implemented as 1/100 sec resolution per one degree.

My idea is to use only one resolution 1/1000 sec but I'm not sure what's the standard...

Can I ask please for some setting details for the page attached?

I'm using 3.9.2252~g1145484, but now I'm not sure if the sat card works anymore, it will not even scan at all.
How can you guys monitor if the rotor turns the dish? Should that appear in the debug window?

Okay, seems I've been too impatient. Scanning started, but I still don't see any sign of rotor movement in the log. The fact that it started scanning another satellite might show this can you confirm?
My dish is out of alignment, so I need to set that, it finds 13° Hotbird services at 28.2° Astra, but that's not an issue. I need to understand how TVheadend handles this, as the current dish setup was working with my Samsung smart TV USALS setup.

2014-12-15 13:15:04.525 subscription: 009D: "scan" subscribing to mux, weight: 4, adapter: "Cine S2 : DVB-S #0 1", network: "28.2E:Astra", mux: "11785H", hostname: "", username: "", client: ""
2014-12-15 13:15:26.000 mpegts: 11785H in 28.2E:Astra - scan no data, failed
2014-12-15 13:15:26.000 subscription: 009D: "scan" unsubscribing
2014-12-15 13:16:31.000 mpegts: 10910V in 13.0E:Hotbird - tuning on Cine S2 : DVB-S #0 1
2014-12-15 13:16:32.891 opentv-ausat: registering mux 10910V in 13.0E:Hotbird
2014-12-15 13:16:32.893 subscription: 009E: "scan" subscribing to mux, weight: 4, adapter: "Cine S2 : DVB-S #0 1", network: "13.0E:Hotbird", mux: "10910V", hostname: "", username: "", client: ""
2014-12-15 13:16:47.000 mpegts: 10910V in 13.0E:Hotbird - scan no data, failed
2014-12-15 13:16:47.000 subscription: 009E: "scan" unsubscribing
2014-12-15 13:16:47.000 mpegts: 11075V in 13.0E:Hotbird - tuning on Cine S2 : DVB-S #0 1
2014-12-15 13:16:48.835 opentv-ausat: registering mux 11075V in 13.0E:Hotbird

Sorry to all, but I reshuffled code again to make logic more clear. I think that I found another issue which may solve Levante's problem. Also, I moved USALS setup log from trace to debug to see something in log event without --enable-trace.

v3.9-2259-gcfc78ab

Note that the site coordinates and motor rate (1/1000sec) are shared for all positions now to keep configuration more easy (and logical).

Still looks good. Motor rate should be good. Might want to consider changing the Max Rotor Movement (seconds): to something along the line of Initial Rotor Setup Time (sec): to make it easier to figure out what that does. This value can be small but the grace time will still calculate a higher value if the sats are far apart. If you remember what sat you left the dish parked on and start with a channel on that sat this value only needs to be a second or two. Otherwise if you set it high you wait a long time for tuning initially.

My V-Box should be here this week. I suspect this code is going to work just fine.

Hello,
commit 2248 works perfectly for me with USALS, but in 2263 it seems broken again.
The dish doesn't move to the correct position. Will try to delete networks and rescan.

Best regards,
Phil.

Ooops- forgot to add- should 'max rotor movement' be the time taken for the rotor to move 160 degrees (thats from end to end in my case).

Philip Reynolds wrote:

Hello,
commit 2248 works perfectly for me with USALS, but in 2263 it seems broken again.
The dish doesn't move to the correct position. Will try to delete networks and rescan.

I don't see any error in the code. Note that "Latitude Direction South" and "Longtitude Direction West" should be probably unchecked for you. Provide some logs to check (diseqc).

Ooops- forgot to add- should 'max rotor movement' be the time taken for the rotor to move 160 degrees (thats from end to end in my case).

Probably yes, it's time to wait for the initial movement (the tvheadend does not know the motor position after start, so this time is used).

Hi Jaroslav,
checking the logs from diseqc, for 0.8W version 2248 will move the rotor 27.7 counter-clockwise, but 2263 will move it to 26 counter-clockwise.

Version 2248-
Dec 15 20:11:00 JungleServer tvheadend¹⁷³⁹⁸: mpegts: 10872H in Thor 0.8W - tuning on TurboSight TBS 6923 DVBS/S2 frontend : DVB-S #0
Dec 15 20:11:00 JungleServer tvheadend¹⁷³⁹⁸: diseqc: initial tone off
Dec 15 20:11:00 JungleServer tvheadend¹⁷³⁹⁸: diseqc: rotor angle azimuth 208.9241 elevation 18.0122
Dec 15 20:11:00 JungleServer tvheadend¹⁷³⁹⁸: diseqc: rotor USALS goto 0.8W (motor 27.7 counter-clockwise)
Dec 15 20:11:00 JungleServer tvheadend¹⁷³⁹⁸: diseqc: sending diseqc (len 5) E0 31 6E D1 BB
Dec 15 20:11:00 JungleServer tvheadend¹⁷³⁹⁸: diseqc: waiting 4 seconds to finish setup for USALS

Version 2263-
Dec 15 20:19:12 JungleServer tvheadend¹⁸⁸⁸²: mpegts: 10872H in Thor 0.8W - tuning on TurboSight TBS 6923 DVBS/S2 frontend : DVB-S #0
Dec 15 20:19:12 JungleServer tvheadend¹⁸⁸⁸²: diseqc: rotor angle azimuth 207.1830 elevation 18.3989
Dec 15 20:19:12 JungleServer tvheadend¹⁸⁸⁸²: diseqc: rotor USALS goto 0.8W (motor 26.0 counter-clockwise)
Dec 15 20:19:12 JungleServer tvheadend¹⁸⁸⁸²: diseqc: sending diseqc (len 5) E0 31 6E D1 A0
Dec 15 20:19:13 JungleServer tvheadend¹⁸⁸⁸²: diseqc: waiting 4 seconds to finish setup for USALS

My long and lat is the same. Any ideas?

Best regards,
Phil.

Philip Reynolds wrote:

Dec 15 20:11:00 JungleServer tvheadend¹⁷³⁹⁸: diseqc: rotor angle azimuth 208.9241 elevation 18.0122
Dec 15 20:19:12 JungleServer tvheadend¹⁸⁸⁸²: diseqc: rotor angle azimuth 207.1830 elevation 18.3989

The input values must be different. The azimuth and elevation is computed from:

Site Latitude, Site Longtitude, Site Altitude and Satellite Longtitude values.

Very odd,
The values entered are exactly the same. I'll try one more time.

Br,
Phil.

Edit- just the same with identical values, position off by 1.7. Bugger.

just updated to 3.9.2270~gf39f7b8

stream not playing in vlc unless i select universal lnb mode instead of usals, the usals code is moving the dish i believe, the signal and s/noise alive at 75% and 72% my end, just not playing!

I added ROTOR_TEST code, use it as follows:

$ gcc -DROTOR_TEST -lm -o rotor_test ./src/input/mpegts/linuxdvb/linuxdvb_rotor.c

$ ./rotor_test 
Usage: <site_latitude> <site_lontitude> <site_altitude> <sat_longtitude>

$ ./rotor_test 48.9742 14.4723 430 4.8
Input values:
         Site Latidude: 48.9742
       Site Longtitude: 14.4723
         Site Altitude: 430.0000
  Satellite Longtitude: 4.8000

Result:
               Azimuth: 192.8230
             Elevation: 33.0293
                 Angle: 10.8 counter-clockwise

The azimuth/elevation values match http://www.dishpointer.com/ for my site. I also added the input values to tvh trace log now.

Trevor Tate wrote:

just updated to 3.9.2270~gf39f7b8

stream not playing in vlc unless i select universal lnb mode instead of usals, the usals code is moving the dish i believe, the signal and s/noise alive at 75% and 72% my end, just not playing!

I don't understand. In advanced settings, you can configure universal lnb type, too... Which is identical to simplified universal LNB config in the SatConfig checkbox...

yes, in advanced settings

I select universal lnb and vlc opens up ok with picture, I then reselect USALS , picture does not load, the signal strength status is ok (not changed, same as in universal lnb) , so dish must be in correct position. for some reason vlc does not open the picture,

Trevor Tate wrote:

yes, in advanced settings

I select universal lnb and vlc opens up ok with picture, I then reselect USALS , picture does not load, the signal strength status is ok (not changed, same as in universal lnb) , so dish must be in correct position. for some reason vlc does not open the picture,

Provide some logs - --trace diseqc to see what's wrong for you. Note that tvh waits for the max/init rotor time seconds for the first tune to ensure that rotor finished the movement. This value is 120 seconds by default (2 minutes).

Jaroslav Kysela wrote:

Trevor Tate wrote:

yes, in advanced settings

I select universal lnb and vlc opens up ok with picture, I then reselect USALS , picture does not load, the signal strength status is ok (not changed, same as in universal lnb) , so dish must be in correct position. for some reason vlc does not open the picture,

Provide some logs - --trace diseqc to see what's wrong for you. Note that tvh waits for the max/init rotor time seconds for the first tune to ensure that rotor finished the movement. This value is 120 seconds by default (2 minutes).

This is exactly the same problem which I described in my previous post. I do have the same problem. Added a log file.

Suppose he has the same tuning problem I mentioned.
I noticed this also in previous v3.9.1567. But now usals/rotor positioning is working fine was hoping this would be solved using never release.
Added a log file in which I first tune at fixed dish 23.5E tv gelderland which works and second tune with rotor at 28.2E bbc scotland which doesn't give image, though there is a lock and signal. In de log is repeatedly printed:
http: /comet/poll?boxid=35401848648b012eb340c6caa14fa03573fb7c21&immediate=0{{Host=192.168.1.50:9981,Connection=keep-alive,Content-Length=58,Origin=http://192.168.1.50:9981,X-Requested-With=XMLHttpRequest,User-Agent=Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Ubuntu Chromium/37.0.2062.120 Chrome/37.0.2062.120 Safari/537.36,Content-Type=application/x-www-form-urlencoded,Accept=*/*,Referer=http://192.168.1.50:9981/extjs.html?,Accept-Encoding=gzip,deflate,Accept-Language=nl-NL,nl;q=0.8,en-US;q=0.6,en;q=0.4,de;q=0.2,Cookie=ys-api/mpegts/service=(removed the rest -> see log)
access: 192.168.1.46:<no-user> [SATWRE*], conn=0, chmin=0, chmax=0, profile=ANY, dvr=ANY, tag=ANY
Instead of restarting tvheadend can do a mux scan at this frequency and then it works.....for a while....and have to do the same at other frequencies.

just tried to compile your test code
gcc -DROTOR_TEST -lm -o rotor_test ./src/input/mpegts/linuxdvb/linuxdvb_rotor.c

/tmp/ccn83abI.o: In function `to_rev':
linuxdvb_rotor.c:(.text+0x93): undefined reference to `floor'
/tmp/ccn83abI.o: In function `sat_azimuth_and_elevation':
linuxdvb_rotor.c:(.text+0x1b6): undefined reference to `sin'
linuxdvb_rotor.c:(.text+0x207): undefined reference to `cos'
linuxdvb_rotor.c:(.text+0x251): undefined reference to `sqrt'
linuxdvb_rotor.c:(.text+0x2ec): undefined reference to `cos'
linuxdvb_rotor.c:(.text+0x336): undefined reference to `sin'
linuxdvb_rotor.c:(.text+0x3c8): undefined reference to `sqrt'
linuxdvb_rotor.c:(.text+0x3da): undefined reference to `atan'
linuxdvb_rotor.c:(.text+0x4cd): undefined reference to `atan'
linuxdvb_rotor.c:(.text+0x518): undefined reference to `atan'
linuxdvb_rotor.c:(.text+0x5c0): undefined reference to `cos'
linuxdvb_rotor.c:(.text+0x601): undefined reference to `sin'
/tmp/ccn83abI.o: In function `sat_angle':
linuxdvb_rotor.c:(.text+0x824): undefined reference to `cos'
linuxdvb_rotor.c:(.text+0x868): undefined reference to `sin'
linuxdvb_rotor.c:(.text+0x88d): undefined reference to `sin'
linuxdvb_rotor.c:(.text+0x8ad): undefined reference to `cos'
linuxdvb_rotor.c:(.text+0x8d9): undefined reference to `sin'
linuxdvb_rotor.c:(.text+0x8fe): undefined reference to `cos'
linuxdvb_rotor.c:(.text+0x92a): undefined reference to `atan'
linuxdvb_rotor.c:(.text+0xa16): undefined reference to `round'
linuxdvb_rotor.c:(.text+0xa64): undefined reference to `round'
linuxdvb_rotor.c:(.text+0xaa3): undefined reference to `round'
collect2: error: ld returned 1 exit status

Richard Bolster wrote:

Instead of restarting tvheadend can do a mux scan at this frequency and then it works.....for a while....and have to do the same at other frequencies.

You may try to enable "Full DiseqC" switch, if it changes something..

Trevor Tate wrote:

just tried to compile your test code
gcc -DROTOR_TEST -lm -o rotor_test ./src/input/mpegts/linuxdvb/linuxdvb_rotor.c

/tmp/ccn83abI.o: In function `to_rev':
linuxdvb_rotor.c:(.text+0x93): undefined reference to `floor'

This symbol is from the math library (libm), but it's requested to be linked with -lm . Do you have this library in another directory ?

$ ls -la /lib/libm.so.6 
lrwxrwxrwx 1 root root 12 Oct  2 12:43 /lib/libm.so.6 -> libm-2.18.so

Can you get someone to go outside and watch the dish to see if it turns when you flip between usals and not usals? If the dish doesn't move between config changes and it works on Universal but not usals only thing I can think of is 22Khz. Might not be selecting the band on the lnbf.

Not sure how accurate the signal strength indicator is in TVH. I don't think may of the kernel drivers give accurate signal strength info.

Hello Jaroslav,
my libm.so.6 is in /lib/x86_64-linux-gnu/ I'm getting the same error as Trevor.
Is it possible to change it somehow?

Thanks,
Phil.

Philip Reynolds wrote:

Hello Jaroslav,
my libm.so.6 is in /lib/x86_64-linux-gnu/ I'm getting the same error as Trevor.
Is it possible to change it somehow?

No idea, -lm should work when gcc and libc is configured properly.

http://www.linuxquestions.org/questions/programming-9/what-exactly-does-the-lm-option-of-gcc-787841/

EDIT: You may try to add the /lib/x86_64-linux-gnu path to library search patch: -L/lib/x86_64-linux-gnu

gcc -DROTOR_TEST -L/lib/x86_64-linux-gnu -lm -o rotor_test ./src/input/mpegts/linuxdvb/linuxdvb_rotor.c

Hello again,
well, whatever I try it wont compile. Ubuntu is using libm-2.19.so if that makes a difference.
If you have time, could you test these values-
./rotor_test 60.64 24.87 115 -0.8 (if -0.8 is correct for west?)?

These are the exact values used in 2248 and 2263.

Thanks for your patience,
Phil.

Philip Reynolds wrote:

Hello again,
well, whatever I try it wont compile. Ubuntu is using libm-2.19.so if that makes a difference.
If you have time, could you test these values-
./rotor_test 60.64 24.87 115 -0.8 (if -0.8 is correct for west?)?

It gives wrong values, but I found the culprit so fix is in v3.9-2274-g932ade3 .

You are the man

That was it, thank you very much.

Best regards,
Phil.