On 2008-05-31 at 18:59:37 [+0200], SHINTA <shinta.0806@xxxxxxxxx> wrote: > > This code has two problems. > > One is taking long time in CAPEHeader::FindDescriptor(). 1 byte x 1 > mega times disk access seems to be slow, although this code worked well > in ZETA. Can you give some details, please. I just tested under VMware to fread() a ca. 1 MB file byte-wise, which took between 0.1 and 0.2 s. This is certainly not particularly fast, but not that bad either. When running the test on the host Linux it is only 20 - 30% faster. At any rate using the API this way is definitely not a good idea. Reading 64 byte chunks already speeds up the whole process by an order of magnitude. CU, Ingo