On Wed, Jan 25, 2012 at 11:19 PM, Freddie Chopin <freddie_cho...@op.pl> wrote: > W dniu 2012-01-25 23:03, Andreas Fritiofson pisze: >> >> What do you mean? What numbers? Performance-wise, there's no >> noticeable difference. Flash writing speed on the STM32F1x is not CPU >> bound. And the algorithm is exactly the same size (actually two bytes >> shorter thanks to the bug fix) so there's no difference in downloading >> speed. We're talking a few clock cycles penalty here, flash writing >> takes 50盜 (=400 cycles running off internal RC) per 16 bit. > > > You can test that and post the results here - why not? >
Because I like to do other things than run a bunch of tests that you can do yourself if you're interested in the numbers. For example finishing the ftdi driver rewrite that will _really_ make a difference for performance. But OK, these figures are from a build with the new loader: 64 MHz wrote 131072 bytes from file test.bin in 4.326505s (29.585 KiB/s) 8 MHz (why would you want to run at this speed anyway?) wrote 131072 bytes from file test.bin in 4.899374s (26.126 KiB/s) And some old figures I've saved from a recent test run: 64 MHz wrote 131072 bytes from file test.bin in 4.276374s (29.932 KiB/s) /Andreas ------------------------------------------------------------------------------ Keep Your Developer Skills Current with LearnDevNow! The most comprehensive online learning library for Microsoft developers is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3, Metro Style Apps, more. Free future releases when you subscribe now! http://p.sf.net/sfu/learndevnow-d2d _______________________________________________ OpenOCD-devel mailing list OpenOCD-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/openocd-devel
