End of recorded show crashing SageTV w/350

[Humanzee]

Originally Posted by insomniac

In sage.properties or client.properties file, change the setting: pvr350_demux_numbuffers to 16 -

I thought this demux numbuffers thing was supposed to be in the registry not in the properties file. I read that in a thread in a beta forum post by Jeffery.

http://forums.freytechnologies.com/f...ad.php?t=13130

I never found that setting in the registry and ended up adding it in before using that beta.

Is this what you changed to 8, in the registry? or has the setting migrated to the properties file now?

Originally Posted by aberson

My workaround suggestion is to always have something recording after a program ends.

I frequently have the opposite problem. Lets say that it is 5:15 and I am recording a show on the 350 that ends at 5:30. Lets also say that I just started watching a recording. Come 5:30 I get a lockup in the middle of watching a program. Sometimes a reboot. Any ideas here?

FWIW, I have the EOF problem even if I am recording when I reach then end of a file, though with 2 tuners in makes it more difficult to know which tuner is actually recording at the time.

[insomniac]

Maybe I have just been experiencing placebo effect then!

I emailed support days ago for an answer, but I guess they are just too busy to reply......................................................

Originally Posted by Humanzee

I thought this demux numbuffers thing was supposed to be in the registry not in the properties file. I read that in a thread in a beta forum post by Jeffery.

http://forums.freytechnologies.com/f...ad.php?t=13130

I never found that setting in the registry and ended up adding it in before using that beta.

Is this what you changed to 8, in the registry? or has the setting migrated to the properties file now?



I frequently have the opposite problem. Lets say that it is 5:15 and I am recording a show on the 350 that ends at 5:30. Lets also say that I just started watching a recording. Come 5:30 I get a lockup in the middle of watching a program. Sometimes a reboot. Any ideas here?

FWIW, I have the EOF problem even if I am recording when I reach then end of a file, though with 2 tuners in makes it more difficult to know which tuner is actually recording at the time.

[aberson]

Originally Posted by Humanzee

I frequently have the opposite problem. Lets say that it is 5:15 and I am recording a show on the 350 that ends at 5:30. Lets also say that I just started watching a recording. Come 5:30 I get a lockup in the middle of watching a program. Sometimes a reboot. Any ideas here?

This might be related to one of the 2 settings mentioned in my 01-29-2005, 05:22 PM post: seeker/fast_mux_switch or mmc/encoders/XXX/never_stop_encoding

[insomniac]

Originally Posted by Humanzee

Is this what you changed to 8, in the registry? or has the setting migrated to the properties file now?

not sure on this one. I found this setting in my properties file, so im now assuming that it was moved to properties file. It sure would be nice to hear back from support on this, but whatever....

[insomniac]

Just heard back from support. It has been moved to the properties file. I did not get a reply on what exactly this setting was supposed to effect. Hopefully someone can reply on this and clue me in.

I.

[rchady]

Well, as I had mentioned in the past I had been using GBPVR for some time with absolutely no EOF problems w/tvout on my PVR350. However, after some time I got a little frustrated with the lack of features so I decided to give MythTV a try. With very little fiddling, it is working great on my PVR350, including the UI, picture-in-picture, auto commercial skip, etc, etc. All the bells and whistles are working great and absolutely no issues with lockups, crashes, or EOF bugs.

With that, I guess this is a farewell to SageTV for me for good. I've tried getting this to work for years now and for something that Frey claims they officially support, I have seen very little effort to find a fix for this. This might just be my perception as I'm sure a lot of work was going on behind the scenes, but I had purchased my PVR350's due to SageTV's "support" for them and have been fairly bitter at the lack of truth in this (at least for the tvout).

Must say I'm still puzzled how 2 other pieces of software have managed to get around this problem (ok, so one is on another operating system, but still...).

Robert

[wongojack]

In response to Insomniac's fiddling with the reg to improve the EOF bug

Changing the numbuffers setting to 8 in the registry (sage 2.#) seemed to improve the EOF for me to the point where I had a long period of time where I could hit stop and Sage recovers (instead of rebooting)

The registry settings in the memory manager didn't improve performance for me.

[Rhino]

After nearly a year of working with Various drivers, registry hacks, stv's, mpeg decoder options, and sageTV versions, I am getting frustrated at having to work so hard to get my supported PVR350 card to work without hanging at EOF.
Thinking that things may have gotten better since my last update, I recently bit the bullet and upgraded to V5. To my dismay, things appear to be worse than with my previous version (v2.x)
I've followed many of the threads regarding the finger pointing between hauppauge and the folks at Sage and it seems that those talks are going nowhere.
Assuming that the problem is a driver issue, what can we do to get around this bug? Is there any work being done behind the scenes at Sage to provide functionality of stopping playback prior to EOF?

For the PVR350, and it's TV out, to be listed as a supported capture/playback device, I feel it necessitates a level of due diligence on Sage's part. I feel there is a need to insure that show stopping bugs, such as the EOF bug, should be dealt with in a timely manor, regardless of the source of the bug. Be it an internal or external entity, it is something that needs to be addressed. If that means inserting some "creative" work arounds in your own code, it's something that has to be done.
I currently sense that there is nothing being done to find a way around this bug, although I would love the opportunity to be proven wrong. Can someone please provide feedback on what is currently being done to mitigate the problems that have been discussed, in great length, on this forum?
I realize that this post may come off very critical of SageTV. That is not my purpose. I'm simply a frustrated user, left in the dark, who would like an answer a little more in depth than "we're looking into it."

Additionally, If anybody from sage is reading this... can you please provide a supported setup that you have internally verified to work properly? Surely you've run across a configuration internally that was less problematic that you could provide to your users.

Thanks for listening, I hope I can somehow help to find answers to these questions soon.
Ryan

[Opus4]

I merged your post here because it has about everything you will find about the 350. There isn't anything else SageTV can do to work around the driver issues at this point.

- Andy

[Rhino]

Originally Posted by Opus4

I merged your post here because it has about everything you will find about the 350. There isn't anything else SageTV can do to work around the driver issues at this point.

- Andy

Andy, I do thank you for the help, although this doesn't explain why Sage refers to the PVR 350 as supported when they're clearly not making any public effort to mitigate this issue in any way. Programatically speaking, it is simple to find the length of an mpg video. You could use this to stop video playback a few seconds before EOF.

While I do understand that this is thought to be a driver problem, there are other avenues that Sage could travel down to at least "mask" the problem so that it doesn't occur.

If this hardware is supported by sage, surely they have found some configuration that works to their liking. All I'm asking is that they clue us users in to what their lab config is. Surely they wouldn't certify that the PVR350 works w/TV out, and sell the combination on their online store, without finding at least one way around this bug. It's an extremely bad business practice and could possibly cause a lot of negative press.

Does the Linux version of Sage display this same problem?

[Opus4]

Originally Posted by Rhino

Programatically speaking, it is simple to find the length of an mpg video. You could use this to stop video playback a few seconds before EOF.

Sure could. I get the feeling of being stuck in an endless loop on that idea. Please read this entire thread instead of stopping before the end.

- Andy

[Rhino]

Originally Posted by Opus4

Sure could. I get the feeling of being stuck in an endless loop on that idea. Please read this entire thread instead of stopping before the end.

- Andy

I have read the thread in it's entirety, as well as many other EOF threads that have appeared over the last year I've been living with this issue. I'm simply using the premature stop as an example of a work around that could be put in place. Relying on users to independently research this fix via your forums and modify stv's should be a last resort. Additionally, many users are scared to due many modifications fearing that they'll mess things up worse. My point is that if this modification truly fixes the problem, why has it not yet been rolled into an official release? Could it not be made into an at least an advanced menu option on the official Sage release?

In my testing, I've come across the exact same problems everybody is describing. It appears that it's the overlay that is causing the problem. It is my understanding that Sage's definition of the problem lies within a function of the display driver. Hauppauge states that the driver works fine for their use and that it's Sage's Implementation that is not 100% correctc. These facts have been talked over until everyone was blue in the face. While each side does have a valid point, I feel that this description is an easy way out. Is there nothing that Hauppauge's engineers could help with? If this problem can not be addressed, why does sage continue to bundle their software with the PVR350 and claim it is supported?

In reading these forums, user manuals, and the Sagetv site, they all claim to support the PVR350, although their actions feel like I'm using an unsupported card that happens to work 75% of the time. When it works, it works great, although there doesn't appear to be any Sage response to the issue other than "unfortunately it doesn't work"
Maybe I'm looking in the wrong places, but I have not yet seen an official response from Sage stating any of these problems. At the very least you would think this issue would be included in a FAQ to new users. I've seen many people purchase the 350 thinking that the TV out will work only to be greatly dissapointed when greated with these issues.

If this is a supported card, please help us with these errors. Is there a specific 350 driver version that Sage has used for testing? What processor/chipset combination was used as a test bed? Is there a specific configuration that Sage has found to be less problematic? Are any enhancements planned for the future to help mask this bug?

I apologize if I seem to be a pain about this, I don't mean to be. As a consumer of an extremely nice, feature rich software package, I would just like to see even the slightest advance in the right direction. If I could get this issue taken care of, I feel that the software wouldn't only be nice, I'd be great.

[Opus4]

Originally Posted by Rhino

I have read the thread in it's entirety, as well as many other EOF threads that have appeared over the last year I've been living with this issue. I'm simply using the premature stop as an example of a work around that could be put in place. Relying on users to independently research this fix and modify stv's should be a last resort.

Then you read it w/o understanding what has been said many times. It isn't the work around that you _still_ seem to think it is. I'm tired of hearing "if they would just take the time to program SageTV to stop playback before the EOF..." For the last time:

SageTV already stops playback before the end of the file.

It appears that it's the overlay that is causing the problem. It is my understanding that Sage's definition of the problem lies within a function of the display driver. Hauppauge states that the driver works fine for their use and that it's Sage's Implementation that is not 100% correctc.

Hauppauge also doesn't display a UI via overlay on the video. That's the part of the driver that doesn't work. It works 100% correctly as long as you don't use what doesn't work.

Is there nothing that Hauppauge's engineers could help with?

Apparently not.

There are no plans to do anything else with this card. I have posted where the issue is & what settings you can try in SageTV. Other than that: this card's driver simply has issues on some systems and there is no fix.

These facts have been talked over until everyone was blue in the face.

Bingo. I'm sorry that I seem to be a little irritated, but the same suggestions have been made about the 350 for a long time and no one seems to pay attention to the fact that SageTV has already tried all of these things. There is an insurmountable problem in the driver, especially when using the UI on the 350. So, I think I've said the last thing I care to say about the 350, since the same suggestions about a 'fix' are repeatedly made regardless of what I say about those supposed fixes.

If I were you, I simply wouldn't use the card. My suggestion is to use a video card instead. You won't get the 350's problems and they can play many more file types.

And, if you have complaints about the way it is listed in the store, please contact SageTV directly.

- Andy

[insomniac]

Did you say that sagetv stops playback before the end of the file?

[Rhino]

Originally Posted by Opus4

Then you read it w/o understanding what has been said many times. It isn't the work around that you _still_ seem to think it is. I'm tired of hearing "if they would just take the time to program SageTV to stop playback before the EOF..." For the last time:

You're correct. For some reason I had never looked at your linked responses stating that this wasn't the issue. I had read the linked threads, but it was so long ago that I had completely forgotten about the discussion. What has been titled as the EOF bug is anything but. I understand that.

None of us here have knowledge of inner workings of Sage. I honestly wish I did. Hitting the stop button prior to the end of the recording is obviously handled differently than letting the software hit EOF without any interaction. One works, while the other ends in catastrophe. The fact that hitting stop appears to work without problem makes it seem that there is something that could be done. I guess it may take a re-engineering of the core functionality of the software, which will probably not be done. I understand that.

Originally Posted by Opus4

Hauppauge also doesn't display a UI via overlay on the video. That's the part of the driver that doesn't work. It works 100% correctly as long as you don't use what doesn't work.

You are correct here as well. This is what I meant in my previous post. The finger is being pointed squarely at Hauppauge. I'll start talks with Hauppauge. I just want an official explanation about why this doesn't work.

Originally Posted by Opus4

There are no plans to do anything else with this card. I have posted where the issue is & what settings you can try in SageTV. Other than that: this card's driver simply has issues on some systems and there is no fix.

After again reading through the thread, it appears that all your posts have indicated how to stop playback sooner than sage currently does.
I understand that this is probably a driver problem that can't be remedied by simply stopping sooner. If it's an issue with the driver conflicting with other devices, I understand. What I would like to see from sage is a combination of hardware components that have been shown to work properly.

Originally Posted by Opus4

There is an insurmountable problem in the driver, especially when using the UI on the 350. So, I think I've said the last thing I care to say about the 350, since the same suggestions about a 'fix' are repeatedly made regardless of what I say about those supposed fixes.

Knowing that this could be a driver problem, one of my original questions was in regards to the Linux version. Is the Linux version is affected by this problem? Is the windows driver used within the linux version through ndiswrapper or does it use a linux specific module such as ivtv?

Originally Posted by Opus4

If I were you, I simply wouldn't use the card. My suggestion is to use a video card instead. You won't get the 350's problems and they can play many more file types.

I completely understand where you're coming from here, and honestly, I've thought about it. My point in all of the PVR350 is claimed to be supported. In advertising material it's stated that the UI over the TV out is supported. Sage's actions don't appear to show any further support. This thread has been going on for more than a year without a single indication from Sage to new customers that there is a very good chance that things might not work as well as they hoped. While I understand that disclosing that fact isn't exactly the best marketing for them, it would prevent a lot of heartache for people in the long run.

Originally Posted by Opus4

And, if you have complaints about the way it is listed in the store, please contact SageTV directly.
- Andy

Duely noted, and will be done through other means.
Since this forum was linked from Sage's site, I was under the assumption that I was officially santioned by them and that I was voicing my complaint to Sage through an appropriate medium.

I'm honestly not just trying to ruffle feathers, I'm just trying to gather more information from Sage on what can be done to mitigate the errors some of us are seeing. Hopefully, with a little information I can swap a few hardware components or make a few software changes and be completely happy with a stable set up.

[Rhino]

Originally Posted by insomniac

Did you say that sagetv stops playback before the end of the file?

Somewhere I seem to be getting that impression as well

[Opus4]

After re-reading some stuff here, I realized that I can't figure out whether these changes have actually helped anyone or not:

Originally Posted by Opus4

1. Core change: "33. Play seam advance in the media player can now be set by the property videoframe/play_seam_advance. This might be useful in helping alleviate the PVR350 EOF crash bug."
   
   This can be used to force the recording to be ended earlier than it does now. It is the time, in milliseconds, prior to the end of the recording when SageTV will stop playback. I believe the default is 250, so you could try setting it to 1000 or 2000 to try stopping 1 or 2 seconds before the end of the file.
   
   
2. STV change: "30. Detailed Setup -> Customize now has an option to ask to delete a recording when playback ends for a) manual deletable shows only, b) all shows, or c) never"
   
   I think that one is obvious... but you can stop the EOF pop-up from asking if you would like to delete the recording.

If neither of these help, I have no idea what else could be done.

I said before that playback stops before the end of the file, but the default time may not be long enough. You can try setting the videoframe/play_seam_advance property to some higher number -- 5000 or so (for 5 seconds) should really make sure the end doesn't get hit.

Has anyone tried this?

- Andy

[insomniac]

Yeah, i tried that when it was made available. Didnt seem to make a difference. To me, the real problem seems to lie somewhere near how sage stops the playback. I always thought that it was maybe because sage tries to do too many things at once(e.g. overlay the screen, prompt for delete, order a beer, etc).

anyway....

Originally Posted by Opus4

After re-reading some stuff here, I realized that I can't figure out whether these changes have actually helped anyone or not:

I said before that playback stops before the end of the file, but the default time may not be long enough. You can try setting the videoframe/play_seam_advance property to some higher number -- 5000 or so (for 5 seconds) should really make sure the end doesn't get hit.

Has anyone tried this?

- Andy

[bhageman]

Seam_advance set to 5000 didn't help for me either.

[MadClown]

How about we put together a class action suit against
Sage for false advertisement and see if that gets the ball rolling on
a fix. This is just ridiculous now to have a 2 plus year old bug lingering.
Its not even a small bug. Its a major hinderous on the user. If they can't figure
it out than buy the solution from GP-PVR . They seem to have solved it.