Re: [Bug-ddrescue] bug on non-hardbs rounded start position or size!

[Дмитрий Костин]

Good time to you, Antonio!
Greatly thank you for the reply!
For now, your program helped me to rescue 327 Gb! The HDD has 2 partitions, #1 
sized about 160G with non-valuable films, and #5 (logical NTFS) sized 600G with 
plenty of needed data (about 500G). Windows can not mount it, neither check, 
because of BADs and slowness. Now I am using Ubuntu Linux 14.04 and have a free 
disk to save a copy to.
At the moment, the ddrescue program runs with such a string of options:
-c1 -d -vv -f /dev/sdb /dev/sdc -O -L -a4000 -r1 -K149s -i722000000s -R 
~/teralog.txt
The value 722000000s is in the "RESCUED" area, and the very end of disk too, so 
the program works fine. Also, I tried using ddrescueview to detect large 
skipped blocks and later run directly on them. Later, used ddrescuetools to 
read out the MFT and detect the used areas of the partition. So, I have rerun 
ddrescue for very-very many times (about a hundred) with different options. 
Without -d the HDD becomes almost stuck for minutes, even after Ctrl-C pressed.
I recommend to improve the program to correct -i and -s values to multipliers 
of hardbs if using /dev/ and '-d'. I am not so good in C++ to do it myself for 
now. Using binary values is not so good, as some data goes in decimal and using 
calculator every time is not good... :(
Removing -n and -N from the command line helped greatly, now speed statistics 
is much better.
Again, thanks for the program! (and for reading my story, too)