Toppy.org.uk ~ View topic

Author

Message

< TAP and patch development ~ [Ck] ClockFix

Page 1 of 1

R2-D2

Posted: Thu Nov 20, 2008 7:49 pm

Frequent contributor Joined: 18 Dec 2006 Posts: 12149

A fluke uncovered an issue with the time conversion routine used primarily in handling the EIT data. [Ck] ClockFix ought to fix it, and at the same time optimises the routine (in a proper way!) for GMT and BST.

The problem can manifest itself on the tick over of every single minute! And is probably the cause of weird times in Series Link timers under 5.14.X/5.15.X. Technical description ended up here. Smile

_________________
Troubleshooting -- User Manual -- Dark Side of the Matrix: Firmwares and Patches

Livy

Posted: Fri Nov 21, 2008 4:30 am

Frequent contributor Joined: 14 Jul 2006 Posts: 584 Location: Fife, Scotland

Patched and been running since about 11pm.

no problems. Very Happy

_________________

TF5800PVR Black Panther 250GB - 5.13.65 Patched to the hilt - aXeL 1.01
Autostart - MyStuff v6.4, Extend v1.7, EPG2MEI v0.96
Tap Commander 1.34, TF5000 Display 1.53a, SecCache (UK) v0.4, EIT Sub v0.6
Connected via Buffalo Linkstation Live 2TB

R2-D2

Posted: Fri Nov 21, 2008 8:02 am

Frequent contributor Joined: 18 Dec 2006 Posts: 12149

I forgot to mention I've got a TAP that shows the issue, so maybe I should tart it up a bit so people can see for themselves...

_________________
Troubleshooting -- User Manual -- Dark Side of the Matrix: Firmwares and Patches

R2-D2

Posted: Fri Nov 21, 2008 8:42 am

Frequent contributor Joined: 18 Dec 2006 Posts: 12149

ClkTest -- a simple TAP that will run for a maximum of 10 minutes. It constantly checks the time converted by the UTC2LocalTime routine to look for discrepancies in the returned minutes and shows an error message when one is discovered. You can abort the check by pressing the Exit button, and it will also prompt when the 10 minute sample is complete (with no errors).

It does seem that the error is less common during certain activities, like playback -- watching live TV is bad, though.

_________________
Troubleshooting -- User Manual -- Dark Side of the Matrix: Firmwares and Patches

harrismail

Posted: Fri Nov 21, 2008 10:01 am

Frequent contributor Joined: 14 Aug 2006 Posts: 391 Location: Warwickshire

Clktest tested and errors abound - ClockFix patched straight away!

Will this result in even more reliable EIT Collection on 1365 as well?

Steve

R2-D2

Posted: Fri Nov 21, 2008 10:16 am

Frequent contributor Joined: 18 Dec 2006 Posts: 12149

harrismail wrote:

Will this result in even more reliable EIT Collection on 1365 as well?

It should only affect the UTC times that happen to be (roughly) now (according to the system's idea of UTC now). This is probably most significant for Extend (under 5.13.X), I guess. But there's a possibility that the system's idea of UTC "now" is worse than just a few seconds adrift of the real now.

_________________
Troubleshooting -- User Manual -- Dark Side of the Matrix: Firmwares and Patches

chris_s

Posted: Fri Nov 21, 2008 10:40 am

Frequent contributor Joined: 23 May 2008 Posts: 328 Location: Coventry

R2-D2 wrote:

And is probably the cause of weird times in Series Link timers under 5.14.X/5.15.X.

Sorry to be a bit thick, R2, but will this improve 'Native' Series Link timers under 5.14.X/5.15.X?

_________________
TF5810, F/W: MS6 Recommended F/W 12/9/2009
TAPs: QuickJump 1.72; MPDisplayLITE V1.2; QuickBlank 0.2; SecCache (UK) v0.4; EIT Sub (Game) v0.6; MyInfo B5.6; EPG2MEI v0.96; MyStuff 6.4; Font Manager 1.0d; Extend v1.7; MHEG On/Off A3; (Start Channel LCN);
Sig generated by MyInfo on 14/3/12,
TV=Panasonic TX-32DT2 IDTV (32" CRT via CVBS-SCART), Tx: Sutton Coldfield. "My other PVR is a TiVo".

R2-D2

Posted: Fri Nov 21, 2008 11:08 am

Frequent contributor Joined: 18 Dec 2006 Posts: 12149

chris_s wrote:

will this improve 'Native' Series Link timers under 5.14.X/5.15.X?

Maybe a bit -- especially if you've seen them gain a minute regularly. Although there are probably a bucketful of more bugs in there.

_________________
Troubleshooting -- User Manual -- Dark Side of the Matrix: Firmwares and Patches

R2-D2

Posted: Fri Nov 21, 2008 10:09 pm

Frequent contributor Joined: 18 Dec 2006 Posts: 12149

Update ClkTest to work in different time zones (including BST!).

I've also discovered that the problem can be worse than I'd thought: the UTC now time is not updated (in normal circumstances) when playing back a recording. This means it will stay as the time when you started playback, so any conversion of that UTC time will yield a wildly wrong result. Since that is in the past it's probably not a big issue, but it serves to show how rubbish the "optimisation" that causes this problem was.

_________________
Troubleshooting -- User Manual -- Dark Side of the Matrix: Firmwares and Patches

ryclark

Posted: Sat Nov 22, 2008 2:24 pm

Frequent contributor Joined: 10 Jul 2005 Posts: 1450 Location: Shropshire

Could this be part of the reason that Now and Next info seems to be quite often incorrect when you go back to live after playing back a recording?

_________________
Toppy 5800 silver, Hazler Hill (The Wrekin). TAPs - OZ Surfer v0.11d, EPG Navigator 6, QuickJump 1.72, UK Timers 1.2c, UK AutoScheduler 0.73, TAP Commander 1.31, Extend 1.7 Firmware 13.65 TPatched

Valiant

Posted: Sat Nov 22, 2008 3:56 pm

Frequent contributor Joined: 19 May 2008 Posts: 254 Location: UK

Also the non-firing of a timer if a playback has been paused?

_________________
TF5810, F/W: MS6 Recommended F/W 12/9/2009 -Fm+BmFsIPsUUuUxZ
TAPs: SecCache (UK) v0.4; EPG2MEI v0.96; Extend v1.7; MyStuff 6.4; MyInfo B5.6; (EIT Sub v0.6 SnG); (CorrectMsFp X1); (FastScanGUI v0.6d);
Sig generated by MyInfo on 18/7/11

R2-D2

Posted: Sat Nov 22, 2008 9:45 pm

Frequent contributor Joined: 18 Dec 2006 Posts: 12149

ryclark wrote:

Could this be part of the reason that Now and Next info seems to be quite often incorrect when you go back to live after playing back a recording?

More likely it's the fact that EPG data is not collected at all during playback (on an unpatched firmware).

Valiant wrote:

Also the non-firing of a timer if a playback has been paused?

On 5.14.X/5.15.X? No, I expect that's more to do with the CRID data (or the EPG data being collected at all).

_________________
Troubleshooting -- User Manual -- Dark Side of the Matrix: Firmwares and Patches

Valiant

Posted: Mon Nov 24, 2008 10:28 am

Frequent contributor Joined: 19 May 2008 Posts: 254 Location: UK

Quote:

On 5.14.X/5.15.X? No, I expect that's more to do with the CRID data (or the EPG data being collected at all).

You may be right there. I did lose a whole half-hour recording whilst a playback was paused some time ago. I've just tried the same scenario a couple of times and haven't missed anything.

ccs

Posted: Mon Nov 24, 2008 10:34 am

Frequent contributor Joined: 30 Oct 2007 Posts: 1434 Location: E Yorkshire Belmont Transmitter

I've also been deliberately pausing playback this weekend and SL recordings have been ok. I've had quite a few (unwanted) fails in the past, but have never been able to reproduce them.

_________________
TF5810, F/W: MS6 Recommended F/W 12/9/2009 -FmXl+CtEzIScVdZ
TAPs: EIT Sub v0.6; EPG2MEI v0.96; MPDisplayLITE V1.2; MyInfo B5.6; SecCache (UK) v0.4; Extend v1.7; MyStuff 6.4;
Sig generated by MyInfo on 15/2/13

All times are GMT

View next topic View previous topic Page 1 of 1 Toppy.org.uk Forum Index ~ TAP and patch development

You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum