New HTPC

[Rich A]

Update:

I had one instance a couple days ago where using ShowSqueeze to move the file didn't work. BUT the BTV log and Task Que data said it DID ?!

The show in question was still on the primary recording drive, and had not been moved to the secondary drive. Yet the log showed the move being completed and the task que showed the process completed successfully. Before I consider this a "BTV Bug" I'll have to see it happen again a couple times and have some more log details available.

1. Also still trying to figure out why all BTV recordings are fragmented into 4 to 6 fragments during the recording process?

2. Along with why the BTV copy/move process also fragments the show while being copied to a new drive?

I can live with #1 and 2. I guess it has always been this way. And it doesn't seem to create any big problems. After all, a 5 GB drive being put in just 4 or 6 different areas on the hard drive doesn't really put any more load on the drive system. IF it were fragmented into say 100 parts, well that could be a problem. Still, I'm shooting for perfection here. I think that # 2 (just the copy process) might be the easiest thing to modify ?

Lastly, in the quest for perfection, I wish I could have the two BTV post recording processes available to be reversed in execution. Right now, there is no way to stop the ShowSqueeze process from happening before the SmartChapter process. I can't tell you how much better it is, how cleaner the whole system is, and how much faster everything works when the order goes like:

Showsqueeze (only to copy/move)
SmartChapter Generation
Custom Post process stuff.

The only one of those three processes that can be changed is the Custom Post Process. But SS and SC can only be done in the order listed.

Before anyone says "Why not just use a Custom Post Process. file move script?", let me say this. The "power" in the ShowSqueeze move setting is that you can designate WHERE to move any show on a PER SHOW basis.

[Fickey]

Originally Posted by Rich A

...I had one instance a couple days ago where using ShowSqueeze to move the file didn't work. BUT the BTV log and Task Que data said it DID ?!

On rare occasions, I've seen completed recordings remain "hidden", I wonder how that would affect the move process.

1. Also still trying to figure out why all BTV recordings are fragmented into 4 to 6 fragments during the recording process?

2. Along with why the BTV copy/move process also fragments the show while being copied to a new drive?

I wonder if this purposeful fragmentation is to statistically reduce the odds of huge contiguous blocks on the HD, thereby reducing the odds of a recording/playback seek glitch? Or maybe it has some correlation to the buffer slice size/count?

[Rich A]

I don't think the show is being hidden. It's there exactly where it was originally recorded. The showsqueeze copy option works just like a showsqueeze compression option. It will do multiple processes depending on how many cores you have and if you have them enabled in BTV.

After examining the logs, I'm now pretty much convinced this is a bug. The showsqueeze copy utility IS copying the show BUT it's not "moving" it. Rather it is just copying a "temporary" file name of the show back to the orginal folder and then deleting the original show and renaming the new version to it's original name. I believe this is the way they manipulate the file names etc. so that you can do a "compression" showsqueeze and leave the new compressed file in it's original folder. They also give you an option to delete the original after show has been compressed.

What's happening is that every once in a while, the copy show process seems to forget where it was supposed to copy the file to. Take a look at these two log snippets.

This is the way it normally works -----------------------------------

[2009/4/23 22:02:07] Copying from: E:\CSI_ Crime Scene Investigation-(If I Had a Hammer)-2009-04-23-0.tp to: G:\Crime Drama\17a7a933-a942-467e-9b89-74f3545d8f7d.deleteme.tmp
[2009/4/23 22:03:57] Copy of E:\CSI_ Crime Scene Investigation-(If I Had a Hammer)-2009-04-23-0.tp completed.
[2009/4/23 22:04:01] DeletePostProcessor deleted E:\CSI_ Crime Scene Investigation-(If I Had a Hammer)-2009-04-23-0.tp
[2009/4/23 22:04:02] Library detected new media: G:\Crime Drama\CSI_ Crime Scene Investigation-(If I Had a Hammer)-2009-04-23-0.tp
[2009/4/23 22:04:04] Library detected a deleted file E:\CSI_ Crime Scene Investigation-(If I Had a Hammer)-2009-04-23-0.tp

Notice the CSI show was copied from E:\ to it's proper final destination G:\Crime Drama. That is the way that show is set up in it's own showsqueeze option.

Now look at this snippet. Smallville IS set up properly to be copied from the recording drive E: to it's final destination G:\Sci Fi. But instead, it just copied itself back to it's originating folder ! Remember that E:\ is where ALL my shows are recorded. That is where "Smallville" was first recorded. But look what the log shows

[2009/4/24 03:26:01] Copying from: E:\Smallville-(Stiletto)-2009-04-23-0.tp to: E:\97257936-96ba-4295-b269-4e64db21a1a1.deleteme.tmp
[2009/4/24 03:30:19] Copy of E:\Smallville-(Stiletto)-2009-04-23-0.tp completed.
[2009/4/24 03:30:25] DeletePostProcessor deleted E:\Smallville-(Stiletto)-2009-04-23-0.tp
[2009/4/24 03:30:26] Library detected new media: E:\Smallville-(Stiletto)-2009-04-23-0-1.tp
[2009/4/24 03:30:28] Library detected a deleted file E:\Smallville-(Stiletto)-2009-04-23-0.tp

Tonight is a big recording night. It'll be interesting to see which show it decides to "not" move.

[Rich A]

Last night I recorded five digital HD shows in a two hour time span and all recorded and moved correctly. Maybe it's like the watched pot that never boils. I'll ignore it and see what happens.

[Rich A]

Well it's been several days now and this problem (saying it moved show when it didn't) has not re-surfaced. It may have been something I did as I have been doing a lot of experimenting with this new system.

I'm pretty pleased with how this whole thing works. Better than ever before and as I've mentioned, my temporary recording drive is always at zero fragmentation when a new recording or session starts. Shows are moved off the temp drive within seconds of the recording finish.

Play back of the shows from their final destination drives has improved from the server as well as from link locations.

So far my work around for the smartchapter generation problem is to manually mark the finished shows for SmartChapter generation after they have been moved. I'm still looking at various forum posts and other areas for more info regarding post capture process ordering. Being able to set it like this:

1. ShowSqueeze
2. Custom Post Process
3. SmartChapter generation

would greatly enhance BTV for those of us running a lot of recordings and tuners. Think about it from a "power user" point of view. Normally if you have one separate physical recording drive then you could have at any point in time, (all at the same time)

Several shows being written (recording) to it.
Smartchapter being generated
Link machines accessing playback
Server accessing playback.

What I've done is to automatically remove EVERYTHING other than recording from the recording drive. Which takes a BIG load off it during high activity times. All you really need is two dedicated drives ... one for the main temporary recording drive and the other for the final destination/storage drive.

[ZetaVu]

I share your pain Rich, not on the rig set but on the dead htpc. My foxconn mobo just died Wednesday. Fortunately I had just built a new computer for my office and it was just a matter of moving all my capture cards over, swapping a few hard drives, and installing drivers and BTV. While I await the verdict on my RMA to Foxconn (it was under warranty, and they're pretty good about that) my current system is GIGABYTE GA-MA770-UD3 with a 5200 X2 and 2gb Corsair DDR2 800 memory. I upped the PS to an Antec SP-550 and a Sonata Ii case, otherwise same video card, tuners as my sig. My LCDSmartie is down since I don't have a parallel port on this board (debating on whether to get a pci card or usb adaptor, I've heard some usb adaptors don't work with the LCDsmartie).

When I built my system Isaved a ghost of the OS before installing antivirus , firewall, and a lot fo the non-htpc programs, so I was able to load that on my OS partition, install drievers, BTV and restore my latest backup of BTV settings before the crash. I had to run setup wizard to get it to recognize the cards and usb tuners, otherwise everything worked fine. This gave me the chance to install my new WD 1tb recording drive (even formatted in 64k clusters). Only thing I don't have is a separate HD for swap like you have. My new system has over 2.5TB.

It should be about 2 weeks to get my old board replaced, then I need to decide if I keep my newer system in the HTPC or bring it back to the office. I do a lot of video editting and conversion in the office, so i might keep the faster machine there (plus as long as they replace me with the same mobo I've already got my BTV OS set up.)

[Fickey]

Originally Posted by ZetaVu

...I upped the PS to an Antec SP-550...

Just FYI- Maybe Antec has improved them, but those Smart Power PSUs had a reputation for high & early failure rates. My SP-500 died within 6 months, Antec let me RMA it for an Earthwatts, running fine for years now.

[Rich A]

The final power supply going into my HTPC is an OCZ modular PS. It's the 500 watt Mod X Stream-Pro. Very efficient. (at least 82 percent) And very quiet.

[ZetaVu]

Originally Posted by Fickey

Just FYI- Maybe Antec has improved them, but those Smart Power PSUs had a reputation for high & early failure rates. My SP-500 died within 6 months, Antec let me RMA it for an Earthwatts, running fine for years now.

Actually my main htpc (currently down, MB sent back for service) uses and Earthmate 430, I was worried it might be undersized but so far has taken everything I threw at it. I had an Sp-450 (I think) that came with my Sonata II case, that started conking out (My other Foxconn MB would not come out of standby) I sent it back and they sent me the SP-550 as a replacement. However, that also doesn't work for my Foxconn, (onboard 6150), it apparently needs a bigger start load to wake (never tried the Earthmate on it). I put a Thermaltake Purepower 500 on there and it works fine. So, my first SP may have flaked, or my MB flaked. I like the Thermoltakes better and would probably buy more of those before another Antec.

[Rich A]

Originally Posted by ZetaVu

Actually my main htpc (currently down, MB sent back for service) uses and Earthmate 430, I was worried it might be undersized but so far has taken everything I threw at it. I had an Sp-450 (I think) that came with my Sonata II case, that started conking out (My other Foxconn MB would not come out of standby) I sent it back and they sent me the SP-550 as a replacement. However, that also doesn't work for my Foxconn, (onboard 6150), it apparently needs a bigger start load to wake (never tried the Earthmate on it). I put a Thermaltake Purepower 500 on there and it works fine. So, my first SP may have flaked, or my MB flaked. I like the Thermoltakes better and would probably buy more of those before another Antec.

The other day when I was reading this, I just happen to have had a new Antec power supply in our shop waiting to be shipped back for exchange. This was the latest of several we have had die on us over the last month. Some were destroying main boards and others were just dead.

The RMA information was interesting. In it, Antec said to not send back any cabling for a modular power supply. (which was fine with me) but the next statement made me go "hmmm". It said that the defective supply would be exchanged with a "non-modular" one. And if that was a problem to call them. In other words, they didn't want to replace the modular supplies with (new) modular supplies at all.

Anyway, Antec has been dropped from our preferred power supply suppliers at this time.

[redlion4]

/quickly hides this post from our HTPC, currently running happily along on an Antec EA430 (Earthwatts, might be different from ZV's). Also have an Antec EA380 in the kids build (Asus M3A78-EM mATX w/ 780G, AT3200 IGP, and AMD 4850e CPU).

[ZetaVu]

My Earthmate iEA430 s working fine, that's what my HTPC normally runs on, it was the smartpower that went south.

[RiderFan]

I have always bought Antec power supplies, and have never had a problem. My last build last week, I put in a Seasonic supply and it seems to run nice as well. Maybe I have been lucky so far, but I would continue to buy Antec.

P.S. My wife raised an eyebrow when I told her how sexy my new PS is. I resisted the urge to name her, lol. I think I am in way too deep with my PC love affair, lol.

[Rich A]

Originally Posted by redlion4

/quickly hides this post from our HTPC, currently running happily along on an Antec EA430 (Earthwatts, might be different from ZV's). Also have an Antec EA380 in the kids build (Asus M3A78-EM mATX w/ 780G, AT3200 IGP, and AMD 4850e CPU).

Sounds like the prudent thing to do. The ones we were having trouble with were all the ones with the modular cables. Don't recall if that is the EA430 or not? I'll check tomorrow. BTW, I was talking to one of the techs at the shop about this. One of the things that can happen (and did in our shop) was a power supply didn't crowbar and passed a high voltage to the main board frying it. So the main board and power supply were changed and the new main board again got fried. Come to find out the CPU was also damaged and the result was that this damaged CPU was capable of damaging the main board. Like Forest Gump was heard to say. S--t happens.

[redlion4]

Neither the EA380 or the EA430 are modular. I'm too cheap to splurge on those "fancies", haha.