Print to tape hiccups

southside_g wrote on 3/10/2003, 3:59 PM
I have just finished editing my first video project via Vegas Video 3.0c. Everything has worked absolutely dandy -- capturing, editing, previewing, rendering, etc. -- up until the last stage; printing to tape. The project is about 44 minutes long, and the printing and device control work fine except for an occasional "hiccup" or momentary pause/interruption on the A/V stream.

If I try another PTT, they will reappear, but not necessarily (and usualy not) in the same place as the did in the previous print attempt, telling me it is not a problem with the captured, pre-rendered or rendered audio/video files themselves. Typically the hiccups start around 5 to 7 minutes into the print and reappear every few minutes after that.

I have already tried defragging my media drive three different ways (as far as which files go where on the drive)using Norton Speed Disk, have cleaned my system tray and task list of everything except Explorer and Systray, have boosted my RAM from the 256MB I started the project with to 384MB (Kingston memory modules only). I have even tried hard-rendering my entire video to three large AVIs, created a new timeline containing only those AVIs and printing to tape from there. After all these modification, I have not even been able to touch the print hiccups, they appear just as frequently as they did from the beginning.

Here is the platform on which I am running the app. I know it is not exactly state-of-the art in some respects, but it should be plenty to handle the PTT function (especially considering I can capture, edit and render on this system brilliantly.):

Gateway 450MHz/PIII all-Intel motherboard with Phoenix/Gateway bios.
384MB of Kingston 100MHz SDRAM.
Win98SE OS.

ATA/EIDE:
Bus 1 Master: 6.5GB Maxtor System drive (OS and apps).
Bus 1 Slave: 8GB Maxtor data drive (data only, not used for multimedia).
Bus 2 Master: 80GB Maxtor Ultra (7200RPM, 8MB cache, VV project drive used for multimedia only).
Bus 2 Slave: TDK VeloCD 40/10/24x CD-R.

SCSI: Seagate Cheetah 9GB 10KRPM HD/Adaptec AHA9xxxUA UltraSCSI adapter (currently only used for backup/archiving)

A/V I/O: Adaptec DuoConnect 1394/USB card (firewire used for capture and print).

Video source/destination: Sony DCR-TRV310 Digital8 camcorder.

There are also a typical SoundBlaster64 audio card, NVIDIA video adapter and MOTU PCI-324 adapter (multichannel audio interface for MOTU 2408) filling up the remaining PCI slots (the MOTU card is currently inactive).

Can someone clue me in as to what might be getting in the way of my getting a clean print to tape sans hiccups?

TIA for any help or consideration,

Glen S.

Comments

BriceWilliams wrote on 3/10/2003, 5:13 PM
I had a simular problem. When ever i would ptt my camera would randomly "hicup". I replaced my capture board. My board was working just fine in my older computer, but when I upgraded my hardware I had same problem as you. After searching thread, I found other users had simular problems. The capture board would vary with the computer they were using. Search my username whbl173. I'm sure i listed the manufacture and the drivers. Had a response from Sonicfoundry as well. I'm pretty sure it was a SIIG board ($34.00)
TomG wrote on 3/10/2003, 5:41 PM
This is a nasty problem that has been knocked around quite a bit lately. The only way I have been able to get a clean burn (no hiccups) is to defrag (use Windows, not Norton) and place the pre-rendered files on a non-OS HD. I also had to upgrade my bios but run on a Dell P4 using 256MB Rambus. I'm pretty sure you are hearing disk thrashing when you experience the hiccups?

Good luck
TomG
RonE wrote on 3/10/2003, 7:45 PM
Interuptions are usually caused by activity on the PCI bus caused by either a card request or the OS on a a timer to request the activity. There is a good guide at Videoguys.com on setting up various OS for editing video. Essentially turn everything OFF that you do not need. No CD auto sensing, no clock on the taskbar, etc. YOu might try to see if it works with the Soundblaster card and CD rom physically disconnected from the system. The usuall culprits are video cards, sound cards and the CDROM that the system keeps checking.

Looking at your system I would also move the CDROM to slave on IDE1 and make sure that DMA is on for that channel to get the most out of the 80G 7200rpm drive.

ROnE
southside_g wrote on 3/10/2003, 7:50 PM
Tom, thanks for the reply. I have a feeling my problems are running a bit deeper. I have considered trying the native Win defrag (and I may still), but the fact that even when I specifically told Norton to put my VV project files and my pre-rendered and rendered rendered files ahead of everything else on the drive *and* not to reserve any current slack space that it had absolutely zero effect of the glitching tells me that defragging alone just ain't gonna cut it.

Plus as shown in my previous post, I not only have my media project on a seperate drive from my OS, but I have it on a seperate EIDE bus from the OS as well. Plus the media is on the intrinsically fastest HD on the system.

I did find a BIOS upgrade for my motherboard (Intel Tabor2(Rev1)) and am d/ling as I write this. If that does not work I am considering perhaps rearranging my EIDE chain so that my media drive and my CD burner do not share the same bus, or creating a separate user profile that eliminates virtually every driver on my system except the 1394 controller, and using that profile only when performing PTT, but those both seem like grasping at extreme measures. Hopefully the new BIOS will help (will try it sometime in the next 24 hours from this post.)

I hear nothing audible from my system when the hiccups hit. They are only momentary, typically maybe 2-3 frames in length (tho occasionally longer), but they are enough to keep it from being releasable to the client.

G.
southside_g wrote on 3/10/2003, 8:02 PM
Thanks for the reply!

I am seriously hoping it's not the board as I really need the dual 1394/USB capability. The good news is the board is using the native Win98 1394 drivers and that it seems to capture just dandy, so while I have not yet ruled out the board,I have that towards the bottom of my list right now. Also I have had previous posts onthis forum from users who are using the Adaptec board with VV3 and my camcorder sucessfully.

SIIG does also have a combo board that uses the identical drivers to the Adaptec, so that is a backup option. I will search your previous threads as well.

Overall, right now RonE and I seem to be on the same page as to where to go from here (see his reply to my original message). I am upgrading my mobo BIOS, rearranging my EIDE chains and trying to eliminate PCI drivers in a special Win user profile for PTT only.

Will let everybody know how it goes within 24 hours from this post.

southside_g wrote on 3/10/2003, 8:05 PM
Ron, Thanks for the reply;

Sounds like you and I are on pretty much the same page on both the EIDE rearrangement and on the PCI driver elimination (see my other replies in this thread). I have also found a BIOS upgrade for my motherboard which I will try.

I will let everyone know how this next round goes as soon as I know. ;)

G.
JackHughs wrote on 3/11/2003, 9:43 AM
Hi,

Your project drive and your CD drive are on the same IDE channel. If you have the CD set to auto detect (or whatever the right name for this function is), the machine will periodically poll the CD drive to see if there is something in the tray. This will interrupt the DV data stream.

Pardon me if someone else already mentioned this.

JackHughs
Grazie wrote on 3/11/2003, 11:08 AM
Jack, This I'm very interested in! PLease expand on what you know. I'm having this PTT hiccup. May not be related and it maybe. Hmmm . . "the machine will periodically poll the CD drive to see if there is something in the tray" - Please tell me more - seriously I'm really very interested.

Grazie
JackHughs wrote on 3/11/2003, 11:37 AM
For Windows 98, the proper term is "auto insert notification". Enabling this function allows Windows to immediately see a new disk in the CD drive. The quick test is to simply put a disk in your drive. If Explorer shows the disk, then auto insert notification is enabled.

The problem arises because Windows has to check to see if media is in the CD tray. On any given IDE channel, only one device can talk to the PCI bus at a time. So, when Windows queries the CD drive, the DV datastream from the hard drive is interrupted. Auto Insert Notification can be disabled from the "properties" box in Device Manager.

The minor downside of disabling Auto Insert Notification is that Windows will not know that a disk is in the drive until you tell it so. Therefore, application installers and other such things will not auto launch.

My personal preference is to have the capture/print-to-tape drive installed as the one and only device on a dedicated IDE channel. But then again, I also have a preference for both belt and braces.

JackHughs

TheShadow wrote on 3/11/2003, 11:37 AM
I have the same problem with Vegas 3.0c if I try to print to tape from the timeline. I tried the same type of cleanup but nothing helped. I then tried the Video Capture application and haven't had a hiccup since.
Grazie wrote on 3/11/2003, 1:15 PM
Jack, yes I do have "Auto Insert Notification" enabled. I'm running WinME. DOes this mean that my laptop is making a call out to the CD-ROM periodically? And that this is tripping up the DV-flow out through the firewire when PTT? I am very very interested in this. Is it possible for us to communicate away from this Forum?

My best regards

Grazie
southside_g wrote on 3/11/2003, 2:07 PM
Hey there Shadow,

I had looked into doing this myself, but I must have missed something, for in the PTT function of the capture app it appears I can only print individually rendered AVIs. Because of the size of my project it has to render into multiple AVIs (a minimum of 3 on my FAT32 system, which would still be at least 2 on an NTFS system). I could not see a way from the capture app to print multiple AVIs consecutively and seamlessly. What am I missing?

In the meantime, just an update on where I'm at. I did find and install a flash BIOS update for my motherboard. While -- perhaps -- this upgrade may have somewhat decreased the maximum severity of each hiccup, it has not had an effect on the quantity of hiccups.

As I mentioned in previous posts, the hiccups do not appear in the exact same location on each PTT attempt, but there does appear to be a general pattern in them; i.e. the first hiccup does not appear until approximately 5-7 minutes into the print and appears to be the first in a cluster of 3-4 hiccups spanning maybe 2 minutes. Then things may stay relatively stable for maybe another 5 minutes or so before another similar cluster hits.

I thought at first that this may be related to the transition between the end of one 2GB FAT32 render file to the beginning of another, but the timings do not really match up with that theory.

I am with the general consensus here that the problem is a bus stream interrupt, either PCI ir IDE and am working now to try solutions in one or both of those areas. I have a hunch that it is on the PCI side myself -- partically because of the timings and partially just gut feeling ;) -- so I am going to try that first.

News at 11...

G.

P.S. Thanks to everybody who has participated and responded to this thread thus far. Your help in both detailed info and in simply providing a mental whiteboard has be very ehlpful and greatly appreciated. Once I get this damn project out the door, I'll come back and contribute back to the forum in any threads to which I feel compitant as a form of thnaks and payback for the help.
jetdv wrote on 3/11/2003, 2:41 PM
Why would it still be at least 2 on an NTFS system. NTFS could fill your entire drive with one file (the limit is in the terabytes). Also, the Capture program CAN PTT multiple AVIs.
JackHughs wrote on 3/11/2003, 4:16 PM
Hullo Grazie,

The easy thing is to just disable Auto Insert Notification and make a test PTT. Just click on the check-box and its gone until you call it up again. If it works, great. If not, well it didn't take much time and it didn't cost anything - so no harm done.

It's certainly OK to e-mail me directly if you like. However, I freely admit that I know nothing more on this subject other than what I'm posting here.

I hope it works.

JackHughs


Grazie wrote on 3/12/2003, 1:01 AM
Understood.

This PTT hiccup has been reported on different set-ups and different machines. It doesn't appear at the same point in the T/L and does appear in "pairs" of instances.

Putting this all together, it would indicate an intermittent "call-out" tripping up the DV flow. Others have greater power machines than I with the latest drivers BIOS and OS. There is just too much - in terms of experimentation - "control" examples for it not to be a "generic" condition.

I'll give what you suggest a go. I hope it makes a difference. It remains a very depressing position to be in.

Grazie
mbryant wrote on 3/12/2003, 7:27 AM
If anyone is using Windows 2000 (as I am) and wants to disable this "Auto Insert Notification" function, it is more complex, but I found out how.

To disable AIN on Win2k, Click the Start button followed by selecting "Run". Type in "gpedit.msc" and click OK. Click and expand "Computer Configuration", followed by "Administrative Template", followed by "System". Double click "Disable Autoplay" and then select "Enable" followed by clicking "OK".

Grazie - I've seen some of your PTT trouble posts on the VF forum (I'm a VF user, thinking about Vegas so browsing here). I too have PTT problems with VF, and while I haven't proved it completely I think the "fangui" problem was what I was seeing (thanks for your posts on this). Are you now seeing additional PTT problems with Vegas?

Mark B
southside_g wrote on 3/12/2003, 7:48 AM
I was thinking of an older NT4 spec that limited the maximum file size to 4GB. My mistake.
Sab wrote on 3/12/2003, 8:01 AM
Hi,

I have also experienced this problem from time to time. Interestingly enough, the only way to cure it on our machines is by transferring the entire project to the C or system drive!. This goes completely against the proper thought process of keeping all things video on a seperate drive. The only reason I even tried this is because we have no problems of any type on our laptop system.

Additionally, the problem seems more apparent if additional audio such as a voiceover or music bed is used. It seems to work flawlessly if the audio originates from the C drive, along with the pre rendered files. This is NOT the way it should be. In fact it's a real pain not being able to shuttle projects between systems because everything is not all in one place.

Our systems use WinXP Pro and Win2000 Pro on either FIC or Gigabyte mainboards, both using the AMD761 series chipset, 512 megs of DDR, Audigy (drivers only), nVidia 64meg video cards, Pioneer A03 and A04, and several interchangeable drives.

Once again, I'm not advocating putting everything or at least the audio and rendered files on the C or system drive. I'm simply saying it works flawlessly for us and we had to try something or we couldn't PTT at all.

Mike
RonE wrote on 3/12/2003, 8:54 AM
SOund Blaster Cards are know for taking up PCI time that causes all sorts of problems they are not a good card to use in a video or audio editing set up. As I suggested in an earlier post try removing ( PHYSICALLY ) the SB card and any non essential cards on the PCI bus, dissable the CD ROM etc and see if it makes a difference. Then you will know were to look for a solution.

ROnE
Grazie wrote on 3/12/2003, 9:54 AM
Mark - Howdy - GET V4! It is the business!

PTT problem has returned. I'll try an AVI out through VF, just to keep you and me on beam - but I bet it will not be the fault of V4. Fangui helps a bit, but I can still get B/S with the fan/s NOT switching - AAARrrgghhh!!!

. . ..

Grazie
Grazie wrote on 3/12/2003, 9:56 AM
Jack - AIN = NADA! Nothing, no solution here - but thanks for the thought - AAAArrrggghhh.....

Grazie
TheShadow wrote on 3/12/2003, 11:09 AM
Hi Glen,

Printing multiple avi files to tape using the video capture application is no problem. You can load in as many avi files that you want. There is a order field that you can adjust to put them in the correct order. The app will print them seamlessy so you can't even tell that there were multiple files. Please give it a try, it was the solution to my problems.
mbryant wrote on 3/12/2003, 11:11 AM
Grazie,

Thanks, let me know how you get on.

When I had this problem (with VF), I found I did NOT have the problem if I used an old copy I had of "Videowave" to do the PTT. From all the posts in both the VF and VV forums about PTT problems it seems to me that both SF programs are more "sensitive" to other PC activity (fans or whatever) than other apps. So while I expect that VV4 is no worse than VF in this area, it concerns me if it is no better. I also don't understand why the problems are only on PTT, while capture is OK...both are equally CPU intensive.

For me, using Videowave to PTT is workaround; I hate using it as I'd really rather delete Videowave from my PC. I could use the same workaround with VV4 I guess, but given the cost of the software I don't feel like I should have to. To put it bluntly, I don't want to pay $499 for something that doesn't allow me to do the whole job (and a clean PTT is core to this).

Mark B
DGrob wrote on 3/12/2003, 12:29 PM
Just me again Grazie. Still no progress on my Dell either. Looking at videoguys.com comments. They seem to point to windows XP drivers as a potential source of PTT difficulties. I'm still researching bios updates and drivers, all the while copying my rendered .avi-s to C and PTT from there. Grob

BTW, arrrrrghhh!