I don't want to entirely rule out video display issues or some directx craziness but:

Try this: bypass all video fx (in the preview window). Render the problem section to a new track. If you still crash, it could be a corrupt source media file.

If you do not crash, it could be a wierd video plug-in bug.

If you make a region around what you think is the problem area, save as a new .veg, and send it to Drdropout@sonicfoundry.com I will take a look at it. I am not guaranteeing a fix, but by inspecting the project file I might be able to figure something out.

"Strange what you say. I can give a try to DXB 1.33 and see what gives, but if your son used same version as me and it worked ...
I am sure that I selected everything in DXB, rebooted as required, ... I received "work complete" message, no warnings.
How are you and your son pretty sure that you remove DX9 completely ? What tool did you use to check ?"

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Go to Start/Run and type in dxdiag hit enter and it will tell you which version you are running. It should say 8.1 You can continue along to check the files and run the diagnostics (of limited value BTW). Post back here if they show something you aren't sure about.
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
"Concerning the 30-sec "hole". It is not blank frames, it is a true 30-sec gap in the project, where "nothing" exists.
Spooky, isn't it ?"

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
I have had "holes" like that interrupt a Print To Tape. One time it interrupted Vegas and it stopped PTT and another time it interrupted my recording deck causing it to stop recording. Whether or not you want to call them "holes" or "black blank frames" they are an ongoing problem with Vegas which has been brought up many times and has YET to be resolved.:

http://www.sonicfoundry.com/Forums/ShowMessage.asp?MessageID=137083&Page=0

John

SonicEPM - Now THAT'S what I call Customer support.:-

"If you make a region around what you think is the problem area, save as a new .veg, and send it to Drdropout@sonicfoundry.com I will take a look at it."

A big :) from me.

Thanks for popping up here. This type of support we have come to expect in the UK. I'm sure it goes on in the US. Maybe SoFo should investigate the option of going for a type of Customer Satisfaction Award? I really do mean it! We should support OUR Customer Support Team with feedback on what you provide.

Thanks again,

Grazie

Hi. Good morning (11:00 here)
First things first, which means: answering my mail :)

To Craftech:
Concerning DXB: I was using DxDiag also. Strange. I will go and try DXB 1.33
The "hole" was created by me, trying to eliminate the supposed "bad" events which were, by themselves, creating the crash. But I was not being a straight shooter (please read below).
John, your posting seems duplicate. You can delete it, if you wish.

To SonicEPM:
Weird video plug-in bug it shall be (please read below).
I can re-create the problem in any way you want. I can send it to you (thanks !!), but please advise: with or without media ? Not avi's of course, but I could prepare some rotten mpeg's. Or I can send you the EDL printout, so that you can know the properties of each event.
Waiting for your advice.

To Frenchy:
The SoFo support line is just fine and I am glad that you have been helped in the past, but to be true, no-one else took advantage from the help except you. When I come here "screaming", four things happen: SoFo maybe listens and helps (see above), other good-willing people listen and help (*trying* to help is good enough for me). I take benefit from that help. Everybody (including you) takes (potential) benefit from that help. If you keep reading you will (maybe) be prepared when the same problem comes up to you, and you will not need to go to SoFo or to here or to scream ...
Maybe (why not) SoFo could create *another* Forum, called Vegas Support Line Forum, where we could all go "scream" for attention and where we could get attention from SoFo and everybody else and we would have a Support Line Repository there.
Who knows ? that might be a good idea. At least, it would leave this forum only for Creativity and Inventiveness threads.

To Grazie: I would bet that you ARE British, aren't you ? :)
Reading more: yes you are :)))
Concerning SonicEPM's pop: yes, I am also very happy and grateful. I have written above that I was sad with SoFo, but not anymore. Now, I am just very worried, because SoFo seems to be dwindling. Maybe to another thread ...

OK. Mail answered, let's go.

===================================================================================

Hey, I may have solved it !!! At least it seems so...

Wanna know how ? Like Grazie said, going so far we better learn something ...

Hold on tight. It took me more than 50 (!) Vegas reloadings to reach conclusion. But actually, I can say that I ended up enjoying the experience (I'm kinda scientist myself...)

The name of the game here is (1) brute force and (2) play with the "media offline" concept to exhaustion.

Report:

In each experiment, after Vegas loaded I tried to crash the system, zooming in and out, browsing left and right, etc. Either I got a crash or not.

=== put as many media offline as possible, renaming the media folders (just put one 'x' before the name). When Vegas loaded *everything* was offline. No crash.
=== reactivated each media folder one by one, each time reloading Vegas. On purpose, left to the end that folder containing the suspected clips. No crash, except when "that" folder was reactivated.
=== then, activated all the folders, but deactivated some of the clips around and including the suspected "zone". Reloaded. No crash.
=== then, without reloading, I started to add one by one of the offline clips. Using the explorer view, I opened each clip in the trimmer, selected video only, and using the right mouse button, added the clip as a take to the respective media offline. To help my task I had switched on the option "show active take names in events". Of course, if I had an offline event which was named (for example) Clip_035.avi , I added the one named xClip_035.avi (the 'x' was what put it offline in the first place). Whenever I got a crash, I started over again from the crashing point.
=== ok, I detected that Clip_028.avi was the culprit !!! By adding the video and the audio parts one at a time, I concluded that it was the video part of Clip_028 which was at stake.
=== So, I left this clip offline, reloaded Vegas and started making experiences. And then ... crash again. Damn !
=== well, restarting, reloading and experimenting, I discovered that Clip_031 was also wrong ... who would have guessed that ?? Two culprits !!
=== ok, back from start, two clips offline, reload, experiment, ... no crash !!! rock solid !!!
=== then, I started to look closer at both events. Event with Clip_028 had a velocity envelope and a Page Peel transition on the left. Event with Clip_031 had a velocity envelope, a Page Roll transition on the right, an Event Pan/Crop and an HSL effect. See the pattern ??? It should be the event envelope !!!
=== carefully annotated the settings of the envelopes, and removed them. Saved project (under a new name !!!). Reloaded with all media online (including Clips 28 and 31). No crash. Save.
=== moved Clips 28 and 31 to an isolated track and carefully reconstructed the velocity envelopes. No crash. Repositioned Clips in position. No crash. Save.

End of experiment. Phew !!! One hour, maybe (and 4 days before that !!!)

Just to make sure about my conclusions, I must say that suddenly I remembered that I had *not* erased my intermediate project saves (Grazie, I know that you're smiling now...) they were only in my "Garbage" folder. So, I reloaded one save that corresponded to about 80% of this project (part 1). Crash !

Which means that, really, everything was OK with Vegas when that save was made. I would never have been able to finish part 1 if that intermediate save was bad ...

I submit that *some* file *somewhere* must be doing *something* that collided with that peculiar velocity envelopes (but not the *other* envelopes in the project) and when I replicated the envelopes *exactly*, the disturbance was not reproduced. Logical, right ? "Velocity" , "float operation" ...

SonicEPM offered to look at the project (the crashing one, of course) and I willingly oblige. Please, EPM, say how you wish it (read above on this posting). And, I insist on sending the dump :)))

But where will that *something* be? The Windows folders ? The Vegas folders ? Not these, because I cleaned everything and reinstalled, but what if Vegas writes some new files inside the Windows folders *after* it is installed ? That will not be detected by the uninstallers, and later on those files may become altered and we can't know what/where they are ...

Then, what if some of these "new" files (or any other new files) added to the system folders created a "wrong" interface with Vegas (an API , as they say) ? Who can answer to that ?

Well, just to finish (and prove my point) I reinstalled DirectX 9 and Windows Media 9 . No crash !!!

So, what do I have now ? A working project, a grain of wisdom and a contender to the longest and most boring thread in this forum !!!

Maybe someone has learned something ... I sure did learn one or two things, some of which I will leave to separate threads. But that's another story.

So far, that's it for now. Thank'y'all for your friendly support and encouragement. And apologies for being a bore and writing too much.

Cheers ... and thanks !!!

=== then, I started to look closer at both events. Event with Clip_028 had a velocity envelope and a Page Peel transition on the left. Event with Clip_031 had a velocity envelope, a Page Roll transition on the right, an Event Pan/Crop and an HSL effect. See the pattern ??? It should be the event envelope !!!
=== carefully annotated the settings of the envelopes, and removed them. Saved project (under a new name !!!). Reloaded with all media online (including Clips 28 and 31). No crash. Save.
=== moved Clips 28 and 31 to an isolated track and carefully reconstructed the velocity envelopes. No crash. Repositioned Clips in position. No crash. Save.

End of experiment. Phew !!! One hour, maybe (and 4 days before that !!!)


++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Just out of curiosity, did the floating point exception occur when you zoomed in ONLY when the timeline was playing or did it occur when you zoomed in with the timeline standing still?

Sqblz, thou art a real trooper...

It kinda refreshing to hear someone go through such struggles with such good humor.

Having had a similar issue, I greatly appreciate your efforts to identify the problem and find a workaround. I am all too familiar with the time and tedium involved in tracking down this kind of thing, and I salute you for your perserverance.

Not boring reading to me, and others I'm sure.


Regards, MPH

Good sleuthing, and glad you are up and running!

It could be the velocity envelope but I doubt it. My money is on a problem with the source media files- I notice they are close together in name, so they probably came from the same tape and capture session...right? And I'll bet that if you replace the media in those problem events with a new take from some known-good media you won't crash...but I'm not asking you to do any more experimenting at this point.

Please send me the .veg file, that's all I need.

+++++++++++++++++++++++++++++++++++++++++++++
did the floating point exception occur when you zoomed in ONLY when the timeline was playing or did it occur when you zoomed in with the timeline standing still?
+++++++++++++++++++++++++++++++++++++++++++++
I never "played the timeline, but I "browsed" it left and right, meaning that I moved the cursor left or right a certain number of frames, which had the effect of moving scenes in the preview display. For some experiments I zoomed in so much that each cursor movement corresponded to a single frame.
The crash ocurred in the zoom and in the browse movements likewise.

+++++++++++++++++++++++++++++++++++++++++++++
Sqblz, thou art a real trooper...
+++++++++++++++++++++++++++++++++++++++++++++
MartyH, that coming from you is a true cumpliment. Man, *you* are the one, with all those tutorials and such ... Gee, many thanks !!!
As I wrote, I'm kinda scientist myself. I like organization and rationality. And I *never* accept something that puzzles me and I don't understand ... believe me, *IF* I get the necessary help I will catch the "how's" and the "why's" of this.

I used to make programs in Fortran and Assembly, not for work but just for the fun of it ... can you imagine ?
A species in extinction, I reckon ...

Ah Ah, SonicEPM, you are WRONG !!! not that I'm happy about it, it's just that I anticipated your idea :)))

You didn't imagine that I described above ALL the testing layout, did you ?? I *am* a bore just like that, I don't need to try harder ...

One of the things that I pondered was whether the media itself was spoiled. To test that, what I did was: reload Vegas with the "culprit" events offline (Clips 28 and 31). No crash. Then, pick *any other* clip from the Explorer (one of those that wasn't creating problems) and add it as Take to the offline event where originally there was Clip 28 or 31. Guess what ? CRASH !! In the moment that I dropped it...

If you read again you see that I cleaned the envelopes, added the clips and rebuilt the envelopes. What I am using now (without crash) are always the same clips (not recaptured). And yes, you are right, same tape, takes made 1-hour apart and captured sequencially.

I will send you one "crashable" veg soon, maybe tonite (my tonite, your martini hour). Thanks.

It is a bit puzzling. I tend to agree with SonicEPM. If you check my FIRST post:

"the only thing that comes to mind is a corrupt image file or an image file incorrectly displaying an output which Vegas and/or your Video Card cannot properly display."

I am happy you stuck it out and provided enough information for us to give you intelligent input and suggestions. Sometimes posters just get frustrated (understandably)and don't give enough information about systems, hardware, and setups
even when asked to. None of us "know-it-all", but collectively we all benefit.
It takes a lot of patience to enjoy the fun factor in computing. Glad to see you meet that requirement.

John

At any rate, we're all glad you got it solved!!!

For what it's worth and all... Maybe I look at things from a slightly different perspective, and my explaining is not at all up to par, but trying to learn from this with the rest here's my thoughts.

It's easy & perhaps logical to suspect the media files from a technical point of view, though this may not be dead on - probably isn't. With any video file there is a lot of room for what's legal & what's not - we've probably all experienced cases where a media file would open in one program and not another. I'm one of those who (whom?) often use other apps in the video processing chain, so I've come up with my share of puzzling situations where Vegas (& more recntly winmedia encoder9) didn't like a particular file format but only as it was created by another app..

So, it could have been a case where Vegas had to stretch a bit to start with, and the problem files were the one's where the link was weakest - where the chain broke.

In my experience, there's a difference between Vegas accepting a file, & likeing it. I've had to change project & codec settings on the run so-to-speak, opening a project with say settings #1, then changing those to #2 before render, and nothing else would work, please trust me. Otherwise Vegas would not display the frames. To me this says that Vegas latches onto at least some of it's original settings, continues to use them after things have changed. This could have also contributed to the problem discussed in this thread.

Perhaps some of the original stuff Vegas fixated on wasn't accurate any longer, or just wouldn't work? Not worth the trouble of forensics at this point, but it is possible that something about the problem files did change. Norton could have fixed a cluster for example, resulting in a very minor change that somehow Vegas detected because it didn't match the stored information, & then when it tried to manipulate the video based on faulty information, crash.

And, Vegas can save bad project files -- as with P/Shop, I save several versions during creation & this has saved me a few times. "Could be nothing more then this." Could have happened the very last time the project was saved.

As I said earlier, Vegas fixates on some details (otherwise wouldn't be able to force it by changing setting midstream). It keeps track of an awful lot of info for the undo features, and some of this gets written to the project file on save. Perhaps in cases where too much change happens, these original details in the project file either don't always get updated, or become muddled. I've gotten in a wishy washy mood where I couldn't decide just how I wanted the pan/crop settings, changed them *several* times, and had Vegas just get fed up with me, refusing to do any more until I went back to an earlier saved project version.

Don't know if this helped as a postmortem or not? I do think it's misleading to focus on windows or dx. Most often there really isn't that much in either that will cause terrible problems with anything. In my experience with windows & dx in general, & please believe me I'm not a true blue MS fan, problems usually arise because of other software that use windows &/or dx facilities, either because the code in the app was sloppy, or they couldn't envision how a line of code or registry entry here, could effect software at the opposite end of the spectrum.

It's not easy writing code now days, SOFO has moved over to writing in .NET, and no offense intended, I think that dealing with trying to get stuff working with windows & dx9 etc. has put MS code at the top of their lists of stuff most likely to cause the purchase of voodoo dolls.

mike

Craftech, you may have skipped my posting, made 30 min before yours.
About making fun out of it ... well, I wouldn't be here if I didn't. I am not a Pro, I do this because I like ... and challenges make my day.


Mikkie, your posting is refreshing, and I agreed to all those parts that I understood ;))) shame on me for the others ...

I concede that Vegas must deal with an awful lot of data (funny that, Vegas getting fed up with you) and I wonder where is the Undo data resident (some swap file, memory ...).
Which gives me one more argument... indeed it is a waste that Vegas loads, processes, keeps track of, unloads, and then JUNKS all that information. I could very well do with some sort of log file that indexed whatever Vegas was doing from beginning to end of a session. And then, guess what ? Crash, read log file, see where it went (maybe) ...

Maybe this is utopic, maybe not. The dumps are there for some reason (still unexplained to me) so the logfiles could be used as well (I have managed to disentangle Win98 more than once, just by looking at bootlog.txt)

SoFo, could we please have an Internal Option in Vegas v4.1 , to keep a session logfile (or successive session logfiles with increasing names) ?
Thanks either way, SoFo.

SonicEPM, just sent my veg now.

Thanks and good luck !! (to me also ...)

SQBLZ - A truly staggering piece of inductive/deductive reasoning and its application.

As to the most boring thread in this Forum - no! Thorough definatley! Boring not at all. - You should read my postings on the fan/s issues on the VF forum. Dell allows the fan/s to interrupt the BIOS and trip up the piping of dv out through the firewire - GO FIGURE! - But that is history . . .

Moving on. What have I gained from this? Hmmmm . . .

1. - A WISH - Instead of the enigmatic "dump log" appearing, my "wish" would be for SoFo to set aside some programme that would do a - well, what can I call it? - soemthing like a "Health Check" on the validity of the stuff in the timeline. I'm thinking along the lines of the "red line" appearing under my bad spilling in MS WORD - yes?

2. - IMPROVING VV3 - Do I think SOFo would countenance this? I'm really not sure. Of course I'm not suggesting SoFo even attempts to make a checklist of "mispellings". There are far to many combinations to contemplate. However, something like a "runtime" macro checker, slithering through the timeline, sniffing out potential rogue media, its settings, along with highlighting dangers. It's plain that I'm no code writer. But there again that might be an advantage. I don't know what I'm saying in terms of "is it possible?" - but rather "Make It So!" In a sense, THAT is exactly what you, Sqblz, have done - yes? You've carefully repeated a mechanistic approach to an issue - isn't this something a macro could do? I don't know I'm not a programmer. But where there's a will . . . and, IMHO, here's the rub.


3. - THE RUB! - It would take some transparent realisation, by our beloved SoFo, to admit that maybe, possibly they have produced such a complex, function-rich product that it might allow us - the users - produce anomalies which the s/w can't cope with. Hey we aint perfect. Personally, having used VF for nearly a year, and now just starting to delve into the value of VV, I'm very very fortunate to make use of an NLE which does deliver AND in a friendly way, an opportunity to turn my creative thoughts into movies - yes? That said, if SoFo WAS to take this thread as an opportunity to provide an EVEN better s/w package, I'm sure it would gain some further creditbility, which in turn could be reflected in increased T/O, leading to increased profits.

Apologies for the previous "dense" paragraph, but I had to attempt to give my feedback on the gigantic effort Sqblz made, and "sketch-out" my thoughts on potential improvements to Vegas, and come to that, Video Factory too.

Hey - well done Sqlbz! Well done everybody too! S - did I smile? . . . Yes I did!

Regards

Grazie

Wise words, Grazie.

I just cross my fingers for them to go beyond wishful thinking ...

sqblz: I received your project file and was not able to reproduce the problem on any of the machines I tried it on.