Opened 17 years ago
Closed 17 years ago
#2802 closed defect (invalid)
Scanning in Mythtv-setup on dvb-s with rotor does not rotate rotor.
Reported by: | Owned by: | danielk | |
---|---|---|---|
Priority: | minor | Milestone: | unknown |
Component: | channelscan | Version: | head |
Severity: | low | Keywords: | rotor dvb-s channelscan |
Cc: | Ticket locked: | no |
Description
Revision 12256 (patch 2793-v4 applied)
Having a dvb-s capture card wit a diseqc rotor (USALS) and one lnb to the rotor.
Adding an input for the dvb-s card through the "4. input connections" menu options and trying to do a channel scan in the "DTV device Configuration" screen it tries to do a scan directly without rotating the rotor.
This also applies for the channel scan button in "5. Channel Editor" menu option.
As far as I could see in the sources there is no part in there that handles the rotating of the rotor before doing the channel scan.
Attachments (1)
Change History (12)
comment:1 Changed 17 years ago by
Version: | 0.20 → head |
---|
comment:2 Changed 17 years ago by
comment:3 Changed 17 years ago by
Reporter: | changed from Michel@… to michel@… |
---|
comment:4 Changed 17 years ago by
Sorry daniel the email address is still wrong: (should be michel@…)
comment:5 Changed 17 years ago by
Reporter: | changed from michel@… to michel@… |
---|
comment:6 Changed 17 years ago by
Severity: | medium → low |
---|
I asked Yeasah about this and here's what he said:
Well, it rotates the dish for me -- not sure what's going on there.
The scanner doesn't know about the dish rotating status though, and since you run the scanner from mythtv-setup, it's generally the first tuning request made, which means it assumes the worst in terms of previous rotor position, and it won't allow tuning until that is complete (on the order of a minute) -- so the first time you do a scan with a rotor it generally fails (or at least takes a very long time to start).
It would be better if the scanner saw that the dish was rotating, didn't time out when this was happening, and put up a bargraph that showed how long until the rotation was complete. It doesn't seem like it should be hard to add that to the scanner; it's already got the bargraph display, it just needs to know to put up a "dish rotating" message onto it, and update it as the rotation completes.
comment:7 Changed 17 years ago by
I'm wondering if this is an issue with sending diseqc commands too soon after LNB power is applied -- the amount of time required can vary depending on hardware, and it's currently 500ms. Maybe give this a try: in the file libs/libmythtv/diseqc.cpp, change the line
#define DISEQC_POWER_ON_WAIT (500 * 1000)
to a larger timeout, say
#define DISEQC_POWER_ON_WAIT (3000 * 1000)
which would be 3sec.
The above comment about the scanner not knowing about dish rotation and possibly timing out the first time still applies of course -- I'd like to fix that, but there's a lot of other minor annoyances I'd like to fix too, and the list doesn't seem to be getting any shorter lately.
Changed 17 years ago by
Attachment: | mythtv-scan-rotor.patch added |
---|
patch to improve scanner when dish rotor is present
comment:8 Changed 17 years ago by
This patch 1) adds a rotor position status bar to the scan progress dialog, and 2) causes the scanner to always wait for rotor completion instead of timing out if it takes longer than its configured timeout.
It works for me, but I can't confirm that it works properly for systems which don't have a rotor.
comment:9 Changed 17 years ago by
Component: | dvb → channelscan |
---|
comment:10 Changed 17 years ago by
comment:11 Changed 17 years ago by
Resolution: | → invalid |
---|---|
Status: | new → closed |
The original reporter had his signal timeout set too short. But I kept the ticket open because Yeasah had added a rotor GUI patch which I wanted to use in MythTV, and which ignores the signal timeout in the channel scanner until the rotor is in position. Anyway the patch has now been applied to the channel scanning branch so I'm closing the ticket.
Just added the ticket with the wrong email address. Could someone change this to the right one.