Jump to content


  • Content Count

  • Joined

  • Last visited

Everything posted by Dave

  1. AlexR, Not the case, I was compiling a multi-source file project (as I previously stated). With my test system (Win2k machine AMD XP2200 with 256MB RAM) compiling your code (full rebuild) compilation times are as follows: V6.96 - 21s V6.97RC3 - 24s V6.97RC1 - 24s V6.97Beta 3 - would not link code V6.97Beta 2 - 22s So the greater delay you are experiencing is presumably something to do with your environment. Consider turning off any virus checkers you may have running and re-running the test, maybe there is some runtime virus scanning that is slowing things down now that we remove the dependencies of the compiler and linker on Microsoft runtime libraries. Regards Dave
  2. Alex, My findings are as follows for compiling a large multi source file project compiled on a Win2k machine AMD XP2200 with 256MB RAM: Target: PIC16F877A V6.97 RC3 - 48s compile and link V6.96 - 44s compile and link Target: PIC18F4620 V6.97 RC3 - 65s compile and link V6.96 - 61s compile and link So it does look as though V6.97RC3 is slighly very slower, but nothing like what you are describing. Please repeat your tests. Also, if you are able to, send your project to support@sourceboost.com and then we may be able to reproduce your experiences. Regards Dave
  3. Rick, Best you post a sample of what does work, even if that is assembler. Best to post a listing file so we can see code addresses. Regards Dave
  4. Maybe you should use -rb 0x1000 as location 0xFFF is used by the bootloader. Regards Dave
  5. Thanks Dave. Apart from having a quick play I haven't made the jump and used Novo on a real project yet so the following may seem like ridiculous questions. With priorities enabled I assume that 15 levels are always available since there seems to be no way that you can choose the maximum number. Before the re-alignment of the documentation to the code it seemed obvious (at least to me) that you would always use 0 as the highest level and then as many levels as required with increasing numbers and decreasing priority (including equal priority). Since 14 is now the highest priority what should you do if you only require say 3 levels? Does using 14, 13 and 12 have any impact on the code or memory compared to using 2, 1 and 0 in high to low order? Is there any problem in using say 14, 5, and 0? Is it simply the order that is important rather than the absolute values? The actual value makes no difference, a task with higher value priority has precendence over a task with lower priority value. Regards Dave
  6. Yes the Novo RTOS documentation has been updated. Regards Dave
  7. This has not been forgotten (well not yet). Microchip change the magic number in their coff file as it has some changes. Need to determine what has change and see what's needed to make it work. Regards Dave
  8. Pass a pointer to the middle of the array rather than the start: char txt[20]; char* ptr; ptr = txt; // pointer points to start of array ptr = &txt[0]; // pointer points to start of array - same as above ptr = & txt[10]; // pointer points to middle of array Regards Dave
  9. ra68gi, Please try V6.97 Beta3 as it has what you want in it. Regards Dave
  10. Potentially yes, but only the software delay functions and anything that uses them, which is not much. Regards Dave
  11. John, Some entries were missed. Attached is the header file that will be in the next release. Regards Dave PIC16F1936.zip
  12. BoostC doesn't support native floating point. You have to use some floating point functions. You can find the information in the floating point library help file. Regards Dave
  13. I suggest you take a look here: http://www.matrixmultimedia.com/mmforums/v...?f=3&t=7160 Regards Dave
  14. Ettore, Thanks for that, it highlighted a whole chunck of missing definitions. Regards Dave
  15. Did you get the PWM output when you used the USB 5V ? Regards Dave
  16. Yes the documentation does appear to be wrong, I just done the same simple test. I will check some other test samples I have.I'm just surprised that this is the first time anyone seems to have noticed this. Regards Dave
  17. Very glad you find it useful :-) Yes that is correct.You can confirm this by creating two tasks, one with a higher priority than the other. In each task simply call Sys_Yield(); in an infinite loop. Set a break point on both Sys_Yields(); function calls. You should find that only the high priority one gets call. Please try this as when I tried something strange happened (sometimes the low priority Sys_Yield(); task got called), but I was using a new release candidate so all may have not been well, even so I was a little surprised. Regards Dave
  18. Reynard means 36 clock ticks (or around 9 pic instructions), and in fact he is right. I measure exactly this in SourceBoost IDE debugger, but I forgot that the port has to be updated twice one time for off and the other for on. So the code goes around the loop at a rate of 1133.333KHz and the port changes state at this rate, but it has to change state twice for one square wave cycle. Regards Dave
  19. meththas, The code generated should give around double the rate you are seeing, you can check this yourself in the SourceBoost IDE debugger by stepping the code and looking at clock ticks changing at the bottom of the screen. Regards Dave
  20. Would be good if you posted the code you are using, maybe you aren't being very efficient. Regards Dave
  21. The path should point to the folder where the compiler lives. What fails exactlty?Did you rename the existing file or rename it? Using notepad should be fine. Regards Dave
  22. TomF, I guess we need you to provide a small but complete project that demonstates the issue. Regards Dave
  23. Are you saying that the code simple example above does not work? I see no reason why it should not. Regards Dave
  24. Sounds like you are picking up the files from a different location. Have you installed different versions of the SourceBoost package on the same machine in different locations? Check in SourceBoost IDE menu settings->options->Compiler&linker Directory. Make sure the directory is correct for the version of compiler you want to run. This information is stored as part of the project. Regards Dave
  • Create New...