questions about java vs max


    Jun 19 2006 | 7:44 am
    I got a process divided in four parts : clock / timing maintenance, recording, buffer manipulation and playback. Buffer manipulation is in Java, all the remaining is max only. I would like to have the whole process into one single object (would be Java in this case) but i remember : Be carefull with that axe, Eugene (sorry, be carefull with timing in Java).
    - Clock subpatch is dealing with tempo, metros, beats, and counters (lots of) : would it be ok to be implemented in Java (i know, i know, browse the archives) ? - recording doesn't really matter (record~ is ok), but what about recording to disk with Java (just for info) ? - does a Java playback object reading a buffer~ would be efficient as a max only object ? - Simply : does a Java object dealing with MSP can stand against MSP only ?
    The whole things would be used as many as you need sound tracks, so it could make a lot...
    Thanks for your advices & cheers
    f.e -- f.e chanfrault | aka | personal computer music > >>>>>> http://www.personal-computer-music.com > >>>>>> |sublime music for a desperate people|

    • Jun 19 2006 | 3:46 pm
      f.e. there is a java based tool kit for doing this kinda thing. JMSL (java music specification language) you can find it at www.algomusic.com
      i recently worked with the main developer of JMSL on classes and max patches that would allow for easy JMSL control of MSP patches.
      -lcc
    • Jun 19 2006 | 9:45 pm
      Every time you cross from the Java world to the C++ world of MaxMSP code you pay a price--elsewhere in the list I have seen this pegged at 2.5 milliseconds in MacOS. I suspect it is similar on PCs.
      ISTR that Jitter Tutotrial Patch 38jPerformanceSetup.pat provides an example of how Java can be optimized--to a point--to get around this. It seems to be a System-level problem, not easily solved by tweaking MaxMSP code--ie.e, not likely to get better in the future--unless Cylcling74 has made some Great Discovery.
      So, the answer to your question is that there are definite performance limits on what Java can do. MaxsMSP and externals written in C++ will not have the same latency problem. Though it is relatively small, it can have a serious cumulative effect.
      -- Paul
    • Jun 24 2006 | 6:55 am
      so, choosing Java to write an external designed to do looping playback from a buffer~ is not an issue ? especially if you plan to you use dozen of it in a single patch ?
      f.e