Hallo Herr Eric Auer,
> As FreeDOS has the reputation of having slow disk I/O, 'reputation' meaning that Jack told you so... but indeed the FreeDOS kernel is about 2* slower then necessary *on rotating rust disks*; the reason is known (1), hasn't been fixed for the last few years, and given the current maintaining efforts will never be fixed. > I would like to get that quantified a bit... a) why ? b) why don't you try to quantify this for yourself, and only then ask us if we can reproduce this? just asking around if people could make some tests is about the opposite of a benchmark as bechmarks should be a) well defined b) easy to understand what they measure c) reproducable none of this is true for 'some tests' Tom (1) disk transfers that cross a 64k Boundary are split into 2-3 transfers due to DMA boundary crossing, even when this is only necessary for floppy disks. on rotating disks that don't catch this within the disk cache this really hurts. _______________________________________________ Freedos-user mailing list Freedos-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/freedos-user
