2012-09-16, 03:49 AM
This may be related to the other recording cut short threads here.
Given the following scenario:
3 tuners installed,
2 programs ending the same time,
1 program starting the same time that the other 2 are ending,
pre and postpads being used,
The programs are scheduled as one time recordings.
the result:
One of the earlier programs will be cut short by several minutes, and the cut short recording will end about 5 seconds before the minute mark. The NTFS last modified times on all files will be correct (having included the postpad), implying that the files were all closed at the correct time, and that one tuner stopped sending data several minutes too soon.
I haven't determined if the pre and postpad times affect the result. I use a prepad of 3 minutes and a postpad of 8 minutes. The cut short recordings are about 11 minutes short.
I have 3 nPVR boxes. Here are some notes I've compiled:
nPVR Box 1 (nwlp), Lenovo Thinkpad T61:
XP Pro 32bit, Core2Duo 2.0Ghz, 2GB RAM, 500GB SATA2 NCQ working, nPVR 2.4.3,
Box 1, tuner experiment A (this is the setup I'm currently running):
(2) Hauppauge HVR-950Q USB
notes: no problems
Box 1, tuner experiment B:
(3) Hauppauge HVR-950Q USB (via a D-Link DUB-H7 powered USB hub)
notes: recording cut short issue exists
Box 1, tuner experiment C:
(2) Hauppauge HVR-950Q USB
(1) Silicondust HDHomeRun (dual tuner)(via gig-e switch or dedicated USB LAN adapter)
notes: recording cut short issue exists, HDHomerun has ~50 bad frames per hour dedicated LAN or not and I tried it on 2 other machines too and with their command line recorder... probably the use of UDP at fault
-----
nPVR Box 2 (bsod), Sony VAIO PCV-7752:
XP Home 32bit, P4 2GHz, 512MB RAM, 200GB IDE, 300GB IDE recording drive, nPVR 2.4.3,
amazingly this ancient box is sufficient to record 3 streams, I named it BSOD because when I rebuilt it, all it would do was BSOD... hehe.. finally determined the RAM bus drivers were weak and it would only run stable with 1 stick..
Box 2, tuner experiment A:
(2) Hauppauge HVR-1150 PCI
notes: no problems
Box 2, tuner experiment B:
(2) Hauppauge HVR-1150
(1) Dvico Fusion5 HDTV PCI
notes: recording cut short issue exists, Fusion5 has ~50 bad frames per hour, also causes BSODs every few days (this is nothing new.. the other boxes this card has been in since 2006 have suffered BSODs caused by the Fusion5)
Box 2, tuner experiment C:
(2) Hauppauge HVR-1150 PCI
(1) Sabrent TV-PCIDG PCI
notes: recording cut short issue exists, Sabrent stream quality is as good as the Hauppauges and is half the price (wish I'd known that a few months ago...)
Box 2, tuner experiment D (this is the setup I'm currently running, I try to schedule such that it's only using 2 tuners max):
(3) Hauppauge HVR-1150 PCI
notes: recording cut short issue exists
-----
nPVR Box 3 (hdtv), HP dc7800pC8:
XP Pro 32bit, Core2Duo 2.33Ghz, 2GB RAM, 80GB SATA1, 500GB SATA2 with NCQ working recording drive, nPVR 2.5.5
Box 3, tuner experiment A (this is the setup I'm currently running):
(3) Hauppauge HVR-1150 PCI
notes: recording cut short issue exists, I have (3) HVR-1250 and (2) HVR-950Q I'd like to install for 8 tuners total after the cut short issue is resolved, I built this box this month and it's eventually going to be my main recording box but for now it's to help resolve this issue
-----
I'm receiving ATSC via a 15 ft Channel Master antenna (yeah a little overkill, but I used to live 70 miles away, plus I like the outdoor eye candy of a big antenna..hehe..). No amp as I'm only 11 miles from the TX farm. RG6 coax, Antronix splitters, PPC compression connectors.
The OS installs are all new and the boxes are only used for nPVR. I don't run antivirus on them. They are headless and I only use them to record, controlling them via NEWA from my main machine. I use UltraVNC if I need to get to the desktops. I image with Ghost 2003 after any significant changes and use it to revert to previously known good states if anything bizarre happens (and when removing tuners so no remnants of previous tuners remain). The XP installs have all Windows updates (including DirectX) and the only codec installed is FFDSHOW. I use Schedules Direct for program info.
I use a post processing app written with Delphi 4 to generate a batch file that renames via the file's alt. data stream (with a lookup table for the virtual chan num) and moves the file to my file server. I usually run the batch file via the Windows Task Scheduler only when nothing is recording (to reduce the chance of bad frames).
The only thing I haven't ruled out is the Hauppauge drivers. In all my experiments, there has always been at least 2 Hauppauge tuners present. I do have 2 more of the Sabrent cards on the way to experiment with a Hauppauge-less box.
Here are the logs for one of the cut short recordings. These are from Box 3.
[ATTACH=CONFIG]34308[/ATTACH]
Here are 3 programs I recorded, the NTFS last modified times, and the recording lengths (via Videoredo or MPC-HC):
2012-09-03 1600 13.1 WTVT DT Judge Judy.Anime Action Figures Missing.hdtv.ts
NTFS last modified time: 9/3/2012 16:38:01
Recording length: 00:40:57
2012-09-03 1600 3.1 WEDU HD Martha Speaks.Martha Runs the Store.hdtv.ts
NTFS last modified time: 9/3/2012 16:38:01
Recording length: 00:29:55 (this is short by about 11 minutes)
2012-09-03 1630 16.3 CREATE The American Woodshop.Sectional Plantation Table.hdtv.ts
NTFS last modified time: 9/3/2012 17:08:01
Recording length: 00:40:57
I'm thinking the cut short recording ending around the minute mark may be a decent clue as to the cause.
Any ideas?
Given the following scenario:
3 tuners installed,
2 programs ending the same time,
1 program starting the same time that the other 2 are ending,
pre and postpads being used,
The programs are scheduled as one time recordings.
the result:
One of the earlier programs will be cut short by several minutes, and the cut short recording will end about 5 seconds before the minute mark. The NTFS last modified times on all files will be correct (having included the postpad), implying that the files were all closed at the correct time, and that one tuner stopped sending data several minutes too soon.
I haven't determined if the pre and postpad times affect the result. I use a prepad of 3 minutes and a postpad of 8 minutes. The cut short recordings are about 11 minutes short.
I have 3 nPVR boxes. Here are some notes I've compiled:
nPVR Box 1 (nwlp), Lenovo Thinkpad T61:
XP Pro 32bit, Core2Duo 2.0Ghz, 2GB RAM, 500GB SATA2 NCQ working, nPVR 2.4.3,
Box 1, tuner experiment A (this is the setup I'm currently running):
(2) Hauppauge HVR-950Q USB
notes: no problems
Box 1, tuner experiment B:
(3) Hauppauge HVR-950Q USB (via a D-Link DUB-H7 powered USB hub)
notes: recording cut short issue exists
Box 1, tuner experiment C:
(2) Hauppauge HVR-950Q USB
(1) Silicondust HDHomeRun (dual tuner)(via gig-e switch or dedicated USB LAN adapter)
notes: recording cut short issue exists, HDHomerun has ~50 bad frames per hour dedicated LAN or not and I tried it on 2 other machines too and with their command line recorder... probably the use of UDP at fault
-----
nPVR Box 2 (bsod), Sony VAIO PCV-7752:
XP Home 32bit, P4 2GHz, 512MB RAM, 200GB IDE, 300GB IDE recording drive, nPVR 2.4.3,
amazingly this ancient box is sufficient to record 3 streams, I named it BSOD because when I rebuilt it, all it would do was BSOD... hehe.. finally determined the RAM bus drivers were weak and it would only run stable with 1 stick..
Box 2, tuner experiment A:
(2) Hauppauge HVR-1150 PCI
notes: no problems
Box 2, tuner experiment B:
(2) Hauppauge HVR-1150
(1) Dvico Fusion5 HDTV PCI
notes: recording cut short issue exists, Fusion5 has ~50 bad frames per hour, also causes BSODs every few days (this is nothing new.. the other boxes this card has been in since 2006 have suffered BSODs caused by the Fusion5)
Box 2, tuner experiment C:
(2) Hauppauge HVR-1150 PCI
(1) Sabrent TV-PCIDG PCI
notes: recording cut short issue exists, Sabrent stream quality is as good as the Hauppauges and is half the price (wish I'd known that a few months ago...)
Box 2, tuner experiment D (this is the setup I'm currently running, I try to schedule such that it's only using 2 tuners max):
(3) Hauppauge HVR-1150 PCI
notes: recording cut short issue exists
-----
nPVR Box 3 (hdtv), HP dc7800pC8:
XP Pro 32bit, Core2Duo 2.33Ghz, 2GB RAM, 80GB SATA1, 500GB SATA2 with NCQ working recording drive, nPVR 2.5.5
Box 3, tuner experiment A (this is the setup I'm currently running):
(3) Hauppauge HVR-1150 PCI
notes: recording cut short issue exists, I have (3) HVR-1250 and (2) HVR-950Q I'd like to install for 8 tuners total after the cut short issue is resolved, I built this box this month and it's eventually going to be my main recording box but for now it's to help resolve this issue
-----
I'm receiving ATSC via a 15 ft Channel Master antenna (yeah a little overkill, but I used to live 70 miles away, plus I like the outdoor eye candy of a big antenna..hehe..). No amp as I'm only 11 miles from the TX farm. RG6 coax, Antronix splitters, PPC compression connectors.
The OS installs are all new and the boxes are only used for nPVR. I don't run antivirus on them. They are headless and I only use them to record, controlling them via NEWA from my main machine. I use UltraVNC if I need to get to the desktops. I image with Ghost 2003 after any significant changes and use it to revert to previously known good states if anything bizarre happens (and when removing tuners so no remnants of previous tuners remain). The XP installs have all Windows updates (including DirectX) and the only codec installed is FFDSHOW. I use Schedules Direct for program info.
I use a post processing app written with Delphi 4 to generate a batch file that renames via the file's alt. data stream (with a lookup table for the virtual chan num) and moves the file to my file server. I usually run the batch file via the Windows Task Scheduler only when nothing is recording (to reduce the chance of bad frames).
The only thing I haven't ruled out is the Hauppauge drivers. In all my experiments, there has always been at least 2 Hauppauge tuners present. I do have 2 more of the Sabrent cards on the way to experiment with a Hauppauge-less box.
Here are the logs for one of the cut short recordings. These are from Box 3.
[ATTACH=CONFIG]34308[/ATTACH]
Here are 3 programs I recorded, the NTFS last modified times, and the recording lengths (via Videoredo or MPC-HC):
2012-09-03 1600 13.1 WTVT DT Judge Judy.Anime Action Figures Missing.hdtv.ts
NTFS last modified time: 9/3/2012 16:38:01
Recording length: 00:40:57
2012-09-03 1600 3.1 WEDU HD Martha Speaks.Martha Runs the Store.hdtv.ts
NTFS last modified time: 9/3/2012 16:38:01
Recording length: 00:29:55 (this is short by about 11 minutes)
2012-09-03 1630 16.3 CREATE The American Woodshop.Sectional Plantation Table.hdtv.ts
NTFS last modified time: 9/3/2012 17:08:01
Recording length: 00:40:57
I'm thinking the cut short recording ending around the minute mark may be a decent clue as to the cause.
Any ideas?