Hi guys, On Thu, 2008-04-03 at 05:04 +0100, Ben Wheeler wrote: > On Wed, Apr 02, 2008 at 08:26:33PM -0700, Tom Jennings wrote: > > I realize this is a downer on the fun of adding features, and > > that I'm in no position to complain, but the inabiity to keep > > mixxx running for more than an hour or two without crashing > > kinda trumps new features for me. > > Oh absolutely. If this is happening to other people, I would > hope we aren't going to release 1.6.0 until it's fixed. > The GSoC projects meanwhile are new features for the version after that > AIUI, (could be 1.7.0 but if we get SVN scalable skins and a new more > efficient waveform view, I think mixxx deserves a promotion to 2.0 > personally. It already will have about twice the features of 1.4.2 :) >
I'm fully aware of the stability issues that beta2 and SVN have right now, and I've spent _a lot_ of time trying to figure it out. There's something very messed up happening somewhere in the code. I tried disabling the side chain, but that doesn't seem to have alleviated the stability problems for anyone, so I think the side chain code is fine. My next suspect is enginebufferscalest.cpp. Over the course of the 1.6.0 development period, we've had various problems with these EngineBufferScale* classes (some of which have caused insane stability problems). After beta1, I switched EngineBuffer over to using the SoundTouch code (EngineBufferScaleST) exclusively because the EngineBufferScaleLinear class seemed to be causing crashes. Also, people might recall that I fixed a bunch of stability issues that were plaguing Mixxx between last Summer and the Fall, which were caused by thread safety issues in EngineBufferScaleST. I wouldn't be surprised if there were more problems with these files. (The biggest problem is that these EngineBufferScale classes are poorly designed. For example, each subclass has code to "reverse" the audio when it's played backwards, something that is independent of the scaling method.) ... and for the record, I did manage to DJ for several hours (live at a gig) on Saturday night, although I realize that was risky. :) > > All versions of 1.6* rarely last more than 60 minutes before > > either an interface lockup (as reported) or core dump. > > Presume this is 60 mins of use, rather than just sitting playing > a very long MP3? > > > There's no way at present for me to use mixxx in a live > > environment because of this. Am I the only one experiencing this? > > Absolutely, stability has to be our top goal, without it no-one will > want to use mixxx. > Just to reassure everyone, there's no chance in hell we'll be releasing 1.6.0 with any of these stability problems. That being said, it would be ideal if we could get it out the door sometime before the end of May, if possible. As usual, any extra help hunting down these problems is much appreciated. :) Thanks for your patience, Albert ------------------------------------------------------------------------- Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace _______________________________________________ Mixxx-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/mixxx-devel
