End of recorded show crashing SageTV w/350

[julesjohn]

I am seeing this happen every time I finish watching a recorded show. At the very end of a recorded show SageTV flashes, briefly shows the main menu and then crashes. Windows reboots as I have it set to reboot when there is a failure.

Comskip: I do have Comskip enabled but I don't believe that is related as it does the same thing on PBS recorded shows where I have COMSKIP disabled.

Delete Options: I do not have the end of show delete menu popup.

I am using the PVR 350 TV Out only to my TV. SageTV v2.1, and Cayars STV 17 in use. This happens with any STV really.

The current work around is to try and stop the recording during playback right prior to the last frame or end of the show. Any ideas? Thanks.

[pcuoco]

You've discovered the dreaded "end of file bug" that plagues everyone who has a PVR350. One way around it is what you're already doing. Others say that disabling hardware decoding only will stop it, but my poor old PIII 650 takes too much of a performance hit when I try that.

If you do a search for "end of file bug" or "EOF" you'll find more information than you'll probably care to read, and no definitive solution. Us PVR350 users have been trying to solve this since Sage 2 came out.

Not sure if it's a driver problem or not. It's more than likely something in the way that Sage writes to the 350's overlay.

Sorry I couldn't be more helpful.

[julesjohn]

Originally Posted by pcuoco

You've discovered the dreaded "end of file bug" that plagues everyone who has a PVR350. One way around it is what you're already doing. Others say that disabling hardware decoding only will stop it, but my poor old PIII 650 takes too much of a performance hit when I try that.

If you do a search for "end of file bug" or "EOF" you'll find more information than you'll probably care to read, and no definitive solution. Us PVR350 users have been trying to solve this since Sage 2 came out.

Not sure if it's a driver problem or not. It's more than likely something in the way that Sage writes to the 350's overlay.

Sorry I couldn't be more helpful.

This is helpful. It is just odd that even the Betas don't solve it either. I moved back to 2.1 even. Well what to do. Thanks.

[byopvr]

just to pile on... I have the same issue, using pvr350 with via epia m10k...

I'm in the "disable hardware decoding only" camp as well, much to my chagrin... as it kinda defeats the purpose of pairing the pvr350 with lower powered hardware (to be software decode concurrently with the hardware decoding)

*Shrug* luckily the m10k has just enough mustard to handle it, and the cheezy m10k cl266 mpeg accellerator helps a little...

Random thought.. I wonder... is there a super efficient or maybe even a faux-mpeg decoder that pvr350 users could run, as we aren't watching the software output... know what I mean? Kinda a cheezy workaround that would effect CPU usage less. Like a "pretend" -faux-decoder. Thoughts or low cpu-usage decoder suggestions (even if the playback quality is horrible)?

rampy

[jbromwich]

I wanted to know if Frey Technologies was working on a fix for this "EOF bug" so I emailed them to ask. Here is their response:

"We realize this is a big issue for people and we are working with Hauppauge on a fix for this. We are working on improving this and doing what we can. "
[March 6, 2005] - George Oms, Frey Technologies, LLC

Update:
July 29, 2005 -- See new info in message #22

[aberson]

on my old sage box, I had the EOF bug. I heard a rumor that it was related to AMD systems. When I switched to an Intel system, I was relieved that the bug did not reappear.

Tonight I upgraded the RAM on the system. It had a single 128MB DIMM in Slot 0, I added a 256MB DIMM in Slot 1. And the EOF bug returned! I tried swapping slots, no fix. The last thing to try would be the 256 DIMM without the 128MB, haven't tried that yet.

[Aside from the EOF bug, adding the RAM has REALLY improved performance - my RAM usage hovers around 150MB, so getting more than 128 physical totally cuts down on my swapfile usage.]

Could it really be the RAM causing EOF? Seems unlikely to me, but maybe. Maybe this piece of RAM isn't totally compatable with my system? (e.g. pc133 instead of pc100)

Adam

[1972]

I mailed them a month ago about the EOF problem. My reply was that they are working on it together with Hauppauge....

(Ok, a month has passed... still no fix, or will it only be fixed in 3.0? naahhhhh... i don't believe it :-( They problem is around more than a year. If Frey finds it important enough they already would have a fix.... or they realize it can't be fixed, due to the 350 hardware...)

And may the amount of RAM be a problem? I doubt it, I got 512 MB in my system and if the EOF occurs, the memory usage stays the same while the CPU gets 100% and locks or reboots the whole system.

[insomniac]

Tonight I upgraded the RAM on the system. It had a single 128MB DIMM in Slot 0, I added a 256MB DIMM in Slot 1. And the EOF bug returned! I tried swapping slots, no fix. The last thing to try would be the 256 DIMM without the 128MB, haven't tried that yet

I hate to start the goose chase, but....

ok, so if any of you have read any of my previous posts on the pvr350, you probably know that I have a perfectly working setup (sorry people). Like a dumba$$, I tempted fate, and decided to pull a 512mb chip out of my machine. "why would a headless, no resource using machine need 1GB of ram anyway..." moved it to my sons gaming machine where it would do more good

BAM! That night, EOF. WTF?!?! I dont get EOF's. Odd. I didnt make the connection yet. Stutters, etc all night. Next morning, EOF. errrrrr.. no connection...later that night, I see this post and kick myself. "why didnt I remember that I stole ram out of it?!"

Needless to say, my kid is upset, but my EOF is gone again.

My initial guess is RAM latency, not ram size. Try to change your ram latency to +1 of what its rated for and see what happens.

You will all have to forgive me, I will not tempt fate again to see if this would fix the issue ;P

I.

[insomniac]

forgot to mention, thats its been 12+ hours since adding back my memory. The next time I have to reboot (usually every month or so...), I will inspect the chips for chipsize, maker, latency, etc...

I.

[1972]

Originally Posted by insomniac

My initial guess is RAM latency, not ram size. Try to change your ram latency to +1 of what its rated for and see what happens.

Strange... but worth to try it

(As soon as I am able to connect a monitor & keyboard to my system I will try to change it in the BIOS. I'll let you know the results)

[insomniac]

still no EOF...definitely the memory...

At this point, it may be good if people that have a working pvr350 rig like myself, post their ram maker or at least as much info about the ram as possible. Although the only thing that I can imagine that will ultimately affect the EOF would be RAM latency set too low..

I.

[jphhughes]

I recently replaced two 256k sticks of Kingston Menory with one stick of CORSAIR ValueSelect 1GB 184-Pin DDR SDRAM Unbuffered DDR 400 (PC 3200) System Memory - After which I started to get the restart problem with EOF. Prior to the memory change the END of file problems would cause some delay but never restarted the computer. So when you talk about "RAM latency set too low" how do you adjust/change the latency of the RAM. The memory spec show the latency as cas 3. Don't have any idea what that means. thaks for any help.

[insomniac]

In Bios, you can change ram latency to a higher number. If yours is 3, thats probably as high as it can be adjusted, but you can take a look to see it it can go up to 4. In your case, I would suggest that you try to trade some other chips from some other computer and see if your problem goes away...never know...

I.

Originally Posted by jphhughes

I recently replaced two 256k sticks of Kingston Menory with one stick of CORSAIR ValueSelect 1GB 184-Pin DDR SDRAM Unbuffered DDR 400 (PC 3200) System Memory - After which I started to get the restart problem with EOF. Prior to the memory change the END of file problems would cause some delay but never restarted the computer. So when you talk about "RAM latency set too low" how do you adjust/change the latency of the RAM. The memory spec show the latency as cas 3. Don't have any idea what that means. thaks for any help.

[jadawin]

Originally Posted by insomniac

In Bios, you can change ram latency to a higher number. If yours is 3, thats probably as high as it can be adjusted, but you can take a look to see it it can go up to 4. In your case, I would suggest that you try to trade some other chips from some other computer and see if your problem goes away...never know...

I.

I am not a Windows kinda guy. Can anyone provide a step-by-step for checking/setting RAM latency?

Thanks

[aberson]

Originally Posted by aberson

Tonight I upgraded the RAM on the system. It had a single 128MB DIMM in Slot 0, I added a 256MB DIMM in Slot 1. And the EOF bug returned! I tried swapping slots, no fix.

A little more info on this. My system requires PC100 RAM, and although some people said PC133 was ok for it, all the RAM manufacturers said that it's no longer the case. So, I was careful to make the 256 stick a PC100. However, when I opened up the box and saw the existing 128 stick, it was labeled as PC133...

could the speed mismatch between the RAMs cause this sort of problem? Is it related to the latency setting?

Adam

[hubick]

Hrm, this is interesting.

I wonder if it has to do with MTRR's!

I recently went through this huge problem with my video card under Linux after adding RAM: http://www.rage3d.com/board/showthread.php?t=33821469

Linux let's you get at the details of what is going on, and I don't know how well MTRR's are handled by Windows, but the symptoms in this thread sound awefully familiar.

[rchady]

As everyone has said, this problem has been around for some time now. The answer, so far, has always been the same from Frey Technologies as well, 'We are working on it.' or 'It is an issue with Hauppauge drivers.' At least, these are the answers I received from support when I was testing out SageTV and ran in to this issue with it. I was told it was an issue with the Hauppauge drivers and there was nothing they could do about it at this time.

So, I tried GB-PVR, a free PVR software package to see if it had the same problem. Amazingly, it did not. Same Hauppauge drivers, NO EOF BUG! Thinking it might be something I changed in the setup, I recently switched back to SageTV to try it (SageTV is far superior in almost every other way to GB-PVR that I have seen, other than it not working properly with PVR350) and sure enough, SageTV crashes every time.

So, I do not believe this has anything to do with the Hauppauge drivers, but rather has something to do with how SageTV is accessing the file.

One thing I don't understand is a while back there was a STV that detected the EOF and attempted to stop playback just before it. It worked great apparently as I did this before I even knew about the EOF bug and never had an issue. I changed STVs and suddenly started seeing the EOF bug. Switched to the standard STV and it remained and now I don't remember which STV it was that tried to detect this.

Anyhow, I am hoping Frey Technologies will fix this issue once and for all. It is the only thing, in my opinion, that is marring the reputation of their product (at least for people that own PVR350's).

Robert

[jphhughes]

As Follow up to my post above I replaced tne 1GB stick of memory with the previously installed two 256K sticks and the problem of the systems restarting has gone away. So it seems to me that the type of memory being used has an effect on this problem. I have been living with this problem for a long time. If I recall correctly I believe their are 350 users that do not have this problem, If so can someone confirm that my recollection is correct. If everyone has the problem then it need to be fixed.

[Crashless]

Now I'm intrigued.

I've NEVER had a stable 350 system, and bounced around with different issues ranging from the GUI dying, but video working, to EOF BSODs. But all of a sudden, reading insomniac's post, I remember installing a second 512mb some time ago, and they seem to coincide with my current spat of BSODs...

I think I might try yanking that second stick to see if they go away. I've since ditched the 350's output, and gotten used to software decoding so I'm afraid to go back to the beautiful 350 only to have it fail me again. Maybe you're onto something with that latency thing. I wonder also, if it could be dual-channel related, as I also activated dual-channel mode with the second stick.

Hopefully I'll be able to play with this next week.

Edit:

I'm not at home and won't be until Tuesday, but here's a memory testing utility:

http://www.memtest86.com/

Maybe it will shed some light on this. Those who's problems are fixed by messing with memory should run this with both configurations and see if it throws up any errors.

edit2:

rchady

The STV was laurenglenn's STV. I've tried it and it didn't resolve my issue, but that doesn't mean it won't fix yours.

[carlgar]

Originally Posted by jphhughes

If I recall correctly I believe their are 350 users that do not have this problem, If so can someone confirm that my recollection is correct. If everyone has the problem then it need to be fixed.

I can confirm that with Hyper-Threading disabled I do not see any EOF issue.
However, with Hyper-Threading enabled I do see several problems that seem to be related and they often occur at the EOF. I just reported one of these problems today but it did not occur at EOF. The picture just froze, I was not even doing anything with the remote.