Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Sentient]

I am having a problem with MJ7.2|PLS| going unstable after I rip and encode a CD.

It will successfully rip a CD using digital secure and encode it using the external encoder pointed to lame 3.91.  I have it set to rip and encode simultaneously, and the first CD records beautifully.

However, after recording it if I go to play the recorded tracks they skip.  What is more when they skip the whole MJ gui pauses too, the counter stops, the visualization freezes, etc.  I also get the same skipping when trying to play the CD tracks that were just recorded from the DVD-ROM or another unrecorded disc inserted to insure I was actually playing from the drive.

There is more to this instability.  If I attempt to record another CD, along about the second or third song it would page fault recommend closing all programs and restart the computer.  Sometimes it would also lock up the computer solid and I would have to power off to recover.

I have a temporary workaround which consists of shutting down and restart MJ after each rip.  If I do this I can record a CD without any crashes.  Also after a restart of MJ I do not see any skipping.  The tracks previously recorded play very nicely.

I tried several other workarounds thinking that PC settings like screensaver, monitor and disk drive shutoff, and standby mode.  All are disabled so as not to interfere with MJ operation.  MJ still displays this behavior repeatably.  

Here are what I think are the relevant MJ settings I am using:
rip - digital secure
encode - external Lame 3.91 command line: "--alt-preset fast standard"
both id3v1 and id3v2 tags are on.
MJ and CDDB dial out to att.worldnet.net to get CD info when disk is inserted.
this may matter, att uses QOS software that has annoyed the firewall on my other computer)

My PC Configuration:
Performance Test:
Refresh Time: 32443 ms
(note: a rip was active while generating this)
Resource Info:
System resource: 58
GDI resource: 86
User resource: 58

CPU Info:
Number of CPUs: 1
Type: Genuine Intel Pentium II
Speed: 398 MHz
MMX Support: YES
(my true speed is 333MHz, the rip may have confused your tool)
Memory Info:
Memory Load: 64%
Total Physical Memory: 196 MB
Free Physical Memory: 70 MB

OS Info:
Type: Microsoft Windows 98 SE

Misc Info:
Internet Explorer 5

Any suggestions are greatly appreciated.  If this is a known bug, is there a fix scheduled?  I only paid for MJ7.2 last week and fear that a bug fix scheduled later will leave me out in the cold.

Sentient

[Sentient]

I need help troubleshooting this, is it MJ?  Has anyone else seen this behavior?
If MJ is stable, is anyone aware of interactions with the PC/OS that might be responsible?

Thank you.

[Doof]

Have you tried encoding to a different format to see if the behavior repeats? Like with WMA or APE or something? If it doesn't, then it's probably the LAME encoder causing the problem. Other than that, I'm not sure what could be happening.

There are no more bug fixes scheduled for 7.2 that I'm aware of (I'm not an employee). They're working on getting 8.0 finished. You may also want to try upgrading to 8.0.230 and see how you like it. You're license will work in 8.0 (but no further). But if you do, back up your library first, (not the actual mp3 files just the MJ database) because once you go to 8, you can't go back to 7.2 without starting over.

[Sentient]

I tried using APE as suggested with the same results.  The first error was an invalid page fault caused by Media Jukebox in KERNEL32.DLL at 0167:bff7b5c1.  When cleared all subsequent page faults were at 0167:bff88396.  Here is the Dr. Watson report on one of the repeating faults.

The application overflowed its stack.  This is typically caused by unrestricted recursion or an exception inside an exception handler.

Module Name: KERNEL32.DLL
Description: Win32 Kernel core component
Version: 4.10.2222
Product: Microsoft(R) Windows(R) Operating System
Manufacturer: Microsoft Corporation

Application Name: Media jukebox.exe
Description: Media Jukebox
Version: 7.2.280
Product: Media Jukebox
Manufacturer: JRiver, Inc.

It looks to me like a real problem in MJ.

I bought MJ assuming it was a stable product.  I don't want to introduce myself to all the problems that a beta version would have in addition.  From other threads I see that version 8 would be far less stable right now.  What I would like to find out from JRiver here is the following:
1. Is this a an MJ problem that is already known about?
2. Will the stable version 8.x fix it?
3. If neither of the above, What other issues with hardware or op system would cause MJ to infinitely recurse or fail to address exceptions before the next occurs?

Are there DVD-Rom drives that give secure ripping trouble?  Are there known driver issues that could cause this?  My DVD_ROM is listed in device manager is ATAPI DVD-ROM 16X, Firmware revision 231A, Auto insert and DMA are ON as specified by the Cyberlink Power DVD software that came with it.

If all answers turn out negatively, then I will have to assume that MJ is simply incompatible with my hardware or software, if that turns out to be the case who do I write for a refund?

I would prefer to get it working, so far I like MJ quite a bit.  I prefer an integrated solution to individual tools, but the integrated program needs to work as well as the individual tools.

Before giving up I will try the DVD-ROM with DMA off, if it still plays movies correctly and fixes the problem I will let you know.  I will also try ripping off the Sony CD-RW just as another test.  Suggestions are welcome.

Sentient

[Doof]

Well considering the vast majority of users who are using 7.2 without the experiencing the problem you're reporting, I'd say it's not an MJ problem. It may be an incompatibility between MJ and something else that's specific to your system, though.

[Sentient]

Quote:
Well considering the vast majority of users who are using 7.2 without the experiencing the problem you're reporting, I'd say it's not an MJ problem. It may be an incompatibility between MJ and something else that's specific to your system, though.

I strongly suspect that the "vast majority" of MJ users use the fastest ripping method rather than secure just as the vast majority are more interested in small file size than archive quality.  The Dr. Watson report indicates that indeed something in MJ is failing, whether that something is a bug or just a program struggling mightily but unsuccesfully with some other issue is what I am here to ask of a JRiver employee or developer.

Since you are at least willing to listen to this thread, would you try an experiment?  Set your ripper to secure mode and try ripping 3 or even 4 CDs in a row in a single MJ session using ape or an external encoder that runs in a DOS box.  Afterwards, let me know if you see any strange behavior such as program pauses or skips in playback.  Perhaps very few people have tried this combination.  I thank you in advance for trying.

Sentient

[Doof]

I can't do as you ask because I'm not using MJ 7.2 (I'm also not running Win98SE). Well, I could, but it would be comparing apples to oranges and we really want to compare apples to apples. But when I was using 7.2, I did exactly what you're doing (secure rip LAME encoder - I've since moved to EAC and APE |PLS| CUE sheet since that's REAL archive quality) and I never experienced the problems you describe.

But since you mention it... have you tried ripping in a mode other than Secure? Try Digital Large Buffer as that's probably the setting the majority of the users out there were using (at least based on what I've read in here).

Also, have you tried turning off the rip and encode simultaneously option?

It's pretty obvious that something happened with MJ. The real question is, was it MJ that caused it? Or was it something else on your system? And to be honest, if it IS MJ causing the problem, it won't get fixed anyway because they're no longer fixing bugs for 7.2.

[Sentient]

Thanks anyway.  Since no one has replied saying it was a known 7.2 bug fixed in 8, I suspect that if it is a secure rip bug, it would still do it in 8.  It is good to know you had this combination work for you, that indicates that it may be an interaction to which MJ is susceptible.  That is at least some meaningful feedback.  It should just work, my setup is very clean, I just installed the operating system a couple of weeks ago and aside from Worldnet, office 2000, a scanner, Nero and RJ1, there is nothing else on the machine.  I just can't imagine what would be interacting.

I will add the digital large buffer ripping test you suggest to tonight's regimen.  I am up to three dimensions now, 2 drives, DMA on/off, and rip method.  It may be time to do a DOE.  I have already tried turning off simultaneous rip and encode, see my first post, it didn't help.  Another test I just thought of is to disable all autostart items, just in case one of them interacts.  

I thank you for your suggetions and ideas, Doof.  Since this is the only Jukebox out with Secure ripping, I really would like it to work.

Sentient

[Doof]

What about updating the drivers for the IDE controllers on the motherboard?

Definitely try disabling everything that's running at startup.

Hmm... what mode do you have your CD's set up for, ASPI or IDE? Whichever one it is, try the other. I had a problem just listening to CD's in MJ8 because I was trying to use ASPI, but when I switched to IDE (actually I let MJ autodetect it and it chose IDE) the problem went away. If you are using ASPI (like if MJ autodetects and chooses that option) you may also want to try installing the latest ASPI update that's available from Roxio. Somewhere here in Interact is a link to it. If you do a search you should be able to find it.

[Sentient]

Those are some good ideas, thank you.  The IDE drivers were installed by windows 98SE.  I have an Intel 440LX MB in this Dell box with the latest BIOS they offered before moving on to PIIIs, I will see if Intel has some later drivers to download, but everything else has always worked on this machine beautifully, and still does.  I won't rule this out, but it will be fairly low on the list.

I do not have ASPI loaded, everything autodetected to IDE from the very beginning.  I can add trying ASPI to my list of things to try though.  I thought ASPI was primarily for SCSI CD/DVD-ROM access.

My startup items are few and very generic.  Things like office startup, point32, InCD, Systray and the like.  The only one that I have qualms about is MDM, the machine diagnostic manager.  I don't know which app installed it, but it is MS; although a part of visual studio rather than windows.  I wonder about this one since I never installed Visual Studio on that machine.  

It is going to be a long troubleshooting session tonight.

I will post my results tomorrow.  Thanks again for your suggestions and encouragement.

Sentient

[Sentient]

I didn't get far last night.  It took too long.  My first experiment was to disable all startup items.  All I left running were taskmanager and the system tray.  Unfortunately it produced no joy, MJ was unstable after ripping 2 CDs and crashed midway through the third.  I took note of the fact that when the MJ interface freezes and songs skip, the mouse also freezes and then jumps to a new location.  This happened as part of previous instability, but I didn't pay much attention to it.  It looks like MJ got about 4 to 5 tracks farther than average, but that is within the range of variability I have seen trying other workarounds.

Next I tried ripping from the Sony CD-RW.  This morning, after ripping 2 CDs, the second of which was started before bed, I did not see the same level of stop and go behavior.  Playing was smooth except when selecting MJ menu items, and the mouse didn't freeze,  or playback skip like it did during the first experiment.  I did have the visualization and EQ go AWOL just before bed when I right clicked on it unchecked the show EQ and Vis item, then rechecked it to have it not reappear.  This morning I tried turning on and off various options, screens, tabs and views to bring them back.  When I tried to select a tab from the big window MJ crashed.  I don't know if this is related to my ripping problems or not.  I presume not, and decided to repeat this experiment cleanly.  I restarted MJ, started another CD recording from the Sony and went to work.  I did note that the Sony behaves very differently than the DVD-ROM did.  On a track with a scratch, the DVD-ROM progressed steadily but spent much time paused at the end before going to the next track.  With the Sony, the disk, which was spun up to a very fast speed, slowed way down and ripping slowed to a crawl before speeding back up to either finish the track and for subsequent tracks.

Tonight and tomorrow I will try both drives, with 3 CDs each, with DMA off.  If any of this sets of a recognition bell or helps developers improve MJ then this may be worth it.

Sentient

[JohnT]

Sentient-

Thanks for your extensive testing of MJ. Although there was no specific fix for the problem you describe, MJ version 8 has a completely rewritten secure mode so it may work better for you. It would be interesting though if you could verify first if it's the secure mode or just ripping in general (from the DVD-rom drive) that is causing the problem. Did you try ripping in "digital large buffer" mode from that drive?

John T.
JRiver, Inc.

[Sentient]

It has taken a long while but I have a better understanding of what is going wrong now.

I have done lots of testing while compressing my CD collection onto my computer's hard drive.  The tests I performed and their results are here for anyone interested in a longwinded and detailed description.  I will summarize below.

To restate the problem: The symptom is fraction of a second pauses in windows operation (even the mouse halts), skips in playback of CDs or known good mp3 files, and finally a crash of MJ7.2 while ripping a CD.  The symptom occurs only after ripping at least two CDs in a row using my EPO DVD-ROM drive on my refurbished Dell Pentium II box.

I have determined that it is not Media Jukebox settings, Windows settings, or driver issues.  It is not encoder dependent, or a conflict with other running programs.  It seems to be related just to ripping from this DVD-ROM.  Ripping from my Sony CD-RW drive is not an option since it does not have accurate streams output and secure rips just don't happen with it.  However the Sony does not seem to cause MJ to fail the way the EPO does.  I can say that MJ is much more sensitive to the problem in secure mode than it is in digital large buffer mode, but that is little solace since it is hight accuracy rips that I desire.

I also compared MJ against EAC, a secure ripper that performed about 3-4 time better than MJ but still succumbed to the evils of the EPO.  It's symptoms were milder, and didn't effect windows performance the way MJ does, but it still started having playback problems after ripping 12 to 14 CDs in a row.

The tests I have not performed yet have to do with ASPI.  For this I need your assistance.  How do I uninstall the ASPI that is built into Windows 98 Second edition?  If I do so, will it damage Windows?  If I can do so, I will try both programs without any ASPI interface to see if that is part of the problem.  Second, I hear there is a newer ASPI to install, I will search Roxio for it, but would like any advice relating to installing it and it's compatibility with things like Nero and these OEM ATAPI CD drives.  At some point I may even try 8 if someone can tell me it is starting to get stable.

So far I seem to have extracted one moral from this longwinded story and it is DON'T SHOP AT FRY'S.

Sentient

[Sentient]

You said:

Hmm... what mode do you have your CD's set up for, ASPI or IDE? Whichever one it is, try the other. I had a problem just listening to CD's in MJ8 because I was trying to use ASPI, but when I switched to IDE (actually I let MJ autodetect it and it chose IDE) the problem went away. If you are using ASPI (like if MJ autodetects and chooses that option) you may also want to try installing the latest ASPI update that's available from Roxio. Somewhere here in Interact is a link to it. If you do a search you should be able to find it.

Where in the process does MJ autodetect it, and how do I go about changing it?

Thanks in advance,

Sentient

[Jeff]

Just to add a little more data.  I see pretty much the same behavior as reported by Sentient.  After I rip a CD, MJ becomes unstable.  It often won't exit properly and won't
play without skipping.  After I rip 2 or maybe 3 cds, it usually crashes in the middle
(sometimes I have to reboot at this point).  I'm using the Digital Large buffer mode.

I'm also running Windows 98, using an IBM thinkpad laptop.  I'm using the built-in
Toshiba CD rom drive.

[Sentient]

Jeff,

I have another thread going on switching ASPI on and off.  I found out there that win98 keeps it ON always.  Anyway I was told to try the latest ASPI layer software from adaptec as a possible fix for this problem we are both seeing.  I am doing that and have a rip going at home right now.  Later today I will check it and start another rip if I can.  You may want to try upgrading your aSPI too, in any case I will post the results here.

Sentient

[Sentient]

Doof suggested:

...you may also want to try installing the latest ASPI update that's available from Roxio. Somewhere here in Interact is a link to it. If you do a search you should be able to find it.

I went to the adaptec site, downloaded and ran aspichk.  It reported that my windows 95 (really 98se) had an older file on it that needed fixing.  So I downloaded version 4.70 and ran its aspichk, which reported that my win98 system was just fine.  I decided to install the new ASPI anyway and then set about ripping CDs.  I ripped 2 last night, the second one started when I went to bed. I ripped two more this morning and at lunch.  At the end of 4 rips there were no skips in playback from CD or mp3 files.  There were no pauses in the windows interface either.  

While it looks like the ASPI upgrade certainly improved things a lot, and may have fixed them completely, I still crashed MJ before I could start ripping disk 5.  I changed a display column setting to show the file dates, resized several columns while playing back an mp3 file and then attempted to create a new category.  That was too much and it died.  

Thank you all for suggestions and help.  Jeff, try updating your ASPI and let me know how it works for you.

Sentient

[Dirk-Jan]

I'm experiencing a similar problem. MJ crashes after it has encoded a track or two. Tried all codes and several drives. What is ASPI? I hope J River will fix this because i paid the $25 for the encoding-functionality.

[JimH]

Dirk-Jan, please read the FAQ here to learn about ASPI.

[Dirk-Jan]

OK, I did. Had nothing to do with my problem. I did find out that Windows Media encoding does work well. But every possible mp3-encoder still hangs MJ.

[Sentient]

Dirk-Jan

When I finally updated the ASPI layer on my machine my problem went away completely.  It seems MJ is a canary in a coal mine when it comes to ASPI problems, but that can be good when you want to find and fix underlying problems.  I too could rip better with other programs than with MJ while I had the problem, but even the best of other programs eventually succumbed to the ASPI problem.  One program, EAC, eventually started experiencing instability after ripping 12 disks in a row.  I will lay odds that if your problem is ASPI on Win98SE with OEM drive hardware,  windows media player will also get weird after a few more CDs are ripped in a row.

Your last message implies that you believe it is an encoder problem, rather than a ripping problem.  You said:

But every possible mp3-encoder still hangs MJ.

 Have you tried non-mp3 encoders in MJ?  When I tried APE I still had the problem before fixing ASPI, thus ruling out MP3 encoders as a problem.  Have you done that experiment?  Why don't you spell out the details, like I did above, and post your machine's help information too.  Maybe I can help you sort it out.  

Sentient

[Dirk-Jan]

I did try non mp3-encoders: that's what I referred to with 'windows media': not the player but the format. I have the type in the cd-settings pointing to IDE and I am using Windows 2000, so I guess the ASPI-discussion is not relevant?

Here are the details, I don't know how to check some, but here is what I know:
- version 7.2.280
- copy mode: digitally large buffer (tried digital secure too)
- encoder: Lame MP3 (tried all other mp3-encoders too, only the non mp3-encoder windows media works)
- Pentium PC 600 Mhz
- 196 MB Ram, free: 35 mb

Here is another hint: just found out that version 8 works perfectly. That is no help for me however: first I don't want to use a beta-version, second I share the library with a server that has no mmx, so i cannot use version 8 .

Thanks for your help so far!

[Dirk-Jan]

Just updated ASPI from version 4.6 to 4.7 and changed the settings to ASPI. Again MJ crashes when the encoding of the first track starts.

[Sentient]

Ok, updating ASPI didn't help you.  I read here that it matters a lot on windows xp and 2000 so if that wasn't it it has to be elsewhere.  You keep saying that you crash when using mp3.  Does the first track finish ripping before it crashes?  If you turn off rip and encode simultaneosuly will it rip all tracks and then crash when it starts encoding the first one?  If that is the case, it is truly an encoder problem.  For that I would try a few things.  Which versions of lame are you using?  Is it the external encoder?  If so what settings and command line?  If you are using internal mp3 encoders based on lame, are they the ones downloaded from JRiver and are they specifically for version 7.2.  If you try to use a version 8 encoder plugin with version 7.2 it might have problems.  Although, if MJ is well coded it will recognize you are downloading a plugin for a later version and refuse to use it.  If you are using external mp3 like lame 3.91 or 3.92 does the DOS box appear and start encoding before it crashes?  Try looking at a lot of things like that and progress logically, ruling out possibilities one at a time.  Be sure not to change more than one thing at a time too.  If you do you will never figure out which is the problem.

Just so you have something to compare to, here is what I have and do.
I am using the lame.exe from the mitiok site.  I have chosen the external encoder plugin in MJ and pointed it at that lame.exe file.  I set command line control and give it the --alt-preset fast standard %in %out or is it %1 %2.  (Anyway, the example command line in the external plugin reads %in %out options, which won't work.  Always make sure the command line follows the rules of the specific external encoder to avoid problems)
When I rip it copies the first song, and then starts copying the second.  While copying the second a DOS box appears and lame encodes the first.  As each track's rip finishes MJ moves on to the next track usually finishing several tracks ahead of lame.  Then lame, one new dos box at a time, encodes each track until it finishes.

I hope this helps.  Good luck.

[Dirk-Jan]

Hi Sentient, I tried an external Lame 3.92 encoder and it works!!! It's just the internal mp3-encoders that hang MJ. Other than that I already tried every setting-combination possible, it all makes no difference. FYI: it was always the first, the second or the third track that caused the crash and always right at the moment when the encoding-part started. Still curious why the internal codecs don't work but if it ain't broken, don't fix it So I won't touch it anymore now. Well I am happy now and I appreciate all the help you gave.