Hello,

I'm starting to suspect Samsung Utility is doing something wrong, or maybe I am doing something wrong since this is the second F3_x drive in a row with strange behavior.

Things are as follows:

1. Both drives have weak head/s.
2. FW disables some heads upon startup, so reading with them is not possible on a normal start.
3. I applied the "initialize HDD with headmap changing in RAM". The utility patches the ROM and downloads it to the HDD's RAM. The logs shows some messages, like "Patching zones table..." or so.
4. A soft reset is issued.
5. Now I can read with all heads; weak heads output some UNC's.

Now, the strange thing:

1. HDD #1 has H0 and H1 OK. If I initialize the disk with H2, H3, H4 and/or H5 it will disable reading on such heads.
2. After applying the patch, I can read with H0, H1, H2, H3, H4 and H5.
3. Since Samsung zones are huge, I can read OK up to LBA ~75M, so I have the partition, the MFT, and lots of INDX and files/folders.
4. When I reach some LBA > ~75M, things start to get scrambled. Some MFT records point to garbage, or to different-than-expected locations, so all files recovered through the tree/explorer are corrupt. But I can read these LBA's!

5. HDD #2 has H0 weak and H1 OK (only 2 heads). The behavior is similar: I can initialize the drive with H1 only and then issue a soft reset and start reading on H0, but in this case, although I read quite OK (maybe UNC every few dozen thousands or so, with good 'islands'), I have no MBR, no partition start, no MFT start... and a deep scan will find INDX records and files, suggesting a NTFS FS on LBA ~203800 (with no header AFAICS).

Seems to me that in both cases I have a wrong translator affecting weak heads.

Am I doing something wrong? Is it coincidence? Is the 'patching zone tables' routine doing something wrong?

Cheers

OK what is happening is that the drive is reading from H1 when commanded to read from H2, but with the translator portion of H2. So when trying to read the first LBA mapped in H2, what I am reading is the first LBA mapped from H1, actually. That seems logical with a 0,1,1,1,1,1 head map, but the behavior is different in other manufacturers.

Now the new question is, how to read from H2, knowing I can access and read H2's SA?

Cheers

OK, news are I think I just need the H/T microcode for this drive in order to get access to H2 user data. Too bad Samsung is not sharing these codes.

Since I can read OK cylinders from 0 to 63 with H2 (zone 0), maybe there's a way a can fool the drive to make it believe SA occupies all cylinders from 0 to 200k or so. I'll try if that's the only thing left to do!