I really can't use RC2 because everytime I click some song to play.. my whole comp freezes for like 30-60 seconds. After that it's normal again. Song s also start to play really slowly. Have to wait like 15 seconds song to start playing and after that 30-60 seconds before I can do anything with my comp.
Any ideas what might be wrong? I'll stick with RC1 meanwhile.
RC2 acting really slow
Moderator: Gurus
Using clean db RC2 scan.
I have also experience very slow response and problems with song playback with RC2.
-intermittent high CPU for MM...
-Very long start up time...sometimes...
-Lock-ups browsing the Nav Tree..eventually necessitating shut-down from Task Manager--NO Error messages generated from MM or windows.
**Many of these lockups have occured while trying to access or working with Files To Edit node..don't know if the problems center around these files or if it's just that I happen to be working there when the problems occur.
-Garbled playback...requiring MM restart...this has never occured in any version of MM3.
These lockups and slow resonse did not seem to occur with RC1.
No errors have occured that generate the emails.
Here is one error dialog that occurred.
Here is another that occured shortly after the one above...
I am sorry for the vague references...but that is the nature of the issues...recreating is not consitent.
As far as debug is concerned, I suppose I could do it, but I will wait for specific request. The instruction from rusty describes using a small set of tracks for the debug...not sure if this would defeat the purpose.
If I can isolate one specific, recreatable issue I will update. I just wanted to get on the record that RC2 is more problematic here than many versions back.
****update just generated an email level error and sent
I have also experience very slow response and problems with song playback with RC2.
-intermittent high CPU for MM...
-Very long start up time...sometimes...
-Lock-ups browsing the Nav Tree..eventually necessitating shut-down from Task Manager--NO Error messages generated from MM or windows.
**Many of these lockups have occured while trying to access or working with Files To Edit node..don't know if the problems center around these files or if it's just that I happen to be working there when the problems occur.
-Garbled playback...requiring MM restart...this has never occured in any version of MM3.
These lockups and slow resonse did not seem to occur with RC1.
No errors have occured that generate the emails.
Here is one error dialog that occurred.
Here is another that occured shortly after the one above...
I am sorry for the vague references...but that is the nature of the issues...recreating is not consitent.
As far as debug is concerned, I suppose I could do it, but I will wait for specific request. The instruction from rusty describes using a small set of tracks for the debug...not sure if this would defeat the purpose.
If I can isolate one specific, recreatable issue I will update. I just wanted to get on the record that RC2 is more problematic here than many versions back.
****update just generated an email level error and sent
Same here. Did a clean install of RC-2 and now get very slow behavior with AutoDJ enabled.
Specifics: as soon as one song is finished playing the CPU usage rises on both processors (Core2 Duo). The first processor goes to 60-70% and the second processor simultaneously rises but sometimes to a lower level (40-50%). Here's what I mean...
This lasts for about 20 seconds, until a new song appears to the Now Playing list, and then usage on both processors drops. The whole interface is slow during this period.
RC-1 did not exhibit this behavior. Tried reinstalling RC-1 and now get the same broken behavior. Perhaps there was an upgrade to one of the support .dll's (maybe SQLite?) that broke something?
Specifics: as soon as one song is finished playing the CPU usage rises on both processors (Core2 Duo). The first processor goes to 60-70% and the second processor simultaneously rises but sometimes to a lower level (40-50%). Here's what I mean...
This lasts for about 20 seconds, until a new song appears to the Now Playing list, and then usage on both processors drops. The whole interface is slow during this period.
RC-1 did not exhibit this behavior. Tried reinstalling RC-1 and now get the same broken behavior. Perhaps there was an upgrade to one of the support .dll's (maybe SQLite?) that broke something?
Jiri...I run streaming data and a lot of other stuff while working and using MM.
I have only lately begun using MM3 for simple playback instead of MM2 when I am working...
While it would be difficult to point to a specific version where the issue became noticable... MM3 definitely exhibits more CPU overhead than MM2.
This is most evident during playback at the "track change". I get noticable lag in my other programs using MM3...this lag has not really been noticable when using MM2.
On my system, using the Task manager as reference..MM3 appears to use double the CPU resources while playing a song and at song change MM3 maxes out the CPU for several seconds...the actual time depending on the total load on my system.
It noticably and adversely affects my other programs during these track changes, especially streaming data. My data streams will actually "hang" while MM3 is changing songs. Cursor control is momentarily lost, etc...
I assume there are added debug functions running in the MM3 releases that might add overhead...hopefully this, or something similar is the cause of the CPU resource drain.
Are the output Plug-ins different between the MM versions?
Could this be part of the added overhead?
I do notice that if I use the "Remote Speakers output" plug-in rather than the default "MM3 direct sound output" plug-in I get slightly even MORE CPU overhead from MM3 during song changes.
Thanks for any help understanding this...and hopefully a bit better performance will be possible. Please advise if there is any other info I can provide that might be of assistance.
I have only lately begun using MM3 for simple playback instead of MM2 when I am working...
While it would be difficult to point to a specific version where the issue became noticable... MM3 definitely exhibits more CPU overhead than MM2.
This is most evident during playback at the "track change". I get noticable lag in my other programs using MM3...this lag has not really been noticable when using MM2.
On my system, using the Task manager as reference..MM3 appears to use double the CPU resources while playing a song and at song change MM3 maxes out the CPU for several seconds...the actual time depending on the total load on my system.
It noticably and adversely affects my other programs during these track changes, especially streaming data. My data streams will actually "hang" while MM3 is changing songs. Cursor control is momentarily lost, etc...
I assume there are added debug functions running in the MM3 releases that might add overhead...hopefully this, or something similar is the cause of the CPU resource drain.
Are the output Plug-ins different between the MM versions?
Could this be part of the added overhead?
I do notice that if I use the "Remote Speakers output" plug-in rather than the default "MM3 direct sound output" plug-in I get slightly even MORE CPU overhead from MM3 during song changes.
Thanks for any help understanding this...and hopefully a bit better performance will be possible. Please advise if there is any other info I can provide that might be of assistance.
I don't see anything unusual on my machine, it could be related to some specific input plug-in, or maybe reading of tags from some slower (network?) drives, since MM3 needs artwork on track change. Also, MM3 does crossfading on track change, so it has to do some more decoding than MM2 - you can try to use out_DS.dll from MM2 in MM3 to see if it makes any difference.
Btw, there is a better way of measuring application performance: use Process Explorer from Microsoft (formerly from System Internals), find running MM process, click Properties and then Performance tab - you can see the total CPU time consumed by MM there and better report what causes problems.
Jiri
Btw, there is a better way of measuring application performance: use Process Explorer from Microsoft (formerly from System Internals), find running MM process, click Properties and then Performance tab - you can see the total CPU time consumed by MM there and better report what causes problems.
Jiri
Slow performance
If anyone can reproduce this, please send me:
a) your db
b) _specific_ stops that can replicate the performance problem
c) a debug log, with an indication of which action in the debug log triggered the problem
Thx.
-Rusty
a) your db
b) _specific_ stops that can replicate the performance problem
c) a debug log, with an indication of which action in the debug log triggered the problem
Thx.
-Rusty