balamw's advice is good. If the data is important, you can use dd to make a copy. You obviously need a drive anyway so you can start making regular backups. The proper command to start imaging, which almost certainly will capture a large percentage of the former OS X volume, but at least you won't miss a viable portion of the NTFS volume - if it's still valid at all.
Code:
dd if=/dev/disk0 of=/Volumes/<2nd_disk_name>/sectorcopy_disk_start782726216s.img skip=782726216 bs=256k
That probably scolls off to the right, make sure you get the whole thing.
This will skip the first 782726216 sectors of disk0, and start sector copying the rest of them, 256KB blocks at a time, to the end of the disk, and putting them into a file "sectorcopy_disk_start782726216s.img"
You can then do a search on this file:
Code:
hexdump -C sectorcopy_disk_start782726216s.img | grep "eb 52 90 4e 54 46 53 20"
This could take hours. And if you've made more than one NTFS volume on the disk without zeroing it, you'll get multiple hits for this string. Only one will be correct.
The hex values on the left side are 1 sector per x200 (that's hex 200, in decimal it's 512). So you'll have to get the hex value on the left for the line where "eb 52 90 4e...." appears. To get the offset in sectors:
a.) convert the hex value to decimal, divide by 512.
b.) divide by x200 in a hex calculator, convert to decimal.
The sector offset is what you will *add* to 782726216. The result is the start sector for the hypothetical NTFS volume. Equally important is the ending sector and that's like a whole other explanation.
If you're lost now, forget it. Send the disk to data recovery service if it's worth $2000-$4000. If not, write off the data and move on. In order for this to work you'l have to do this at least once, and possibly iterate twice the number of times you've formatted this disk NTFS since it was zero'd.
-------------
Darwin Boot is the Recovery HD. This is Lion? Or Mountain Lion you upgraded to?
-------------
And FYI, the above is about 1/2 of what's needed to recover this. Once you find the NTFS volume header, you still have to test if it's valid. And I only know a likely rudimentary of doing this which is to try and mount the volume read only (OS X does this by default btw, so long as you don't have NTFS writes enabled or 3rd party NTFS read/write software). If it mounts, get the data off ASAP. If not, then you have to compute the next offset, do the addition again, change the partition start sector, and reboot.
BTW, I notice some evidence you're not rebooting in between uses of gdisk, as it says you need to do. Not a good idea. You should reboot right after changing either partition table.
It's also possible that you find the proper NTFS volume start sector, but it needs repair (probably unlikely) before it will mount, even read only. In that case you've got a set of iterations where you try a non-destructive repair using a Windows installer disk, and chkdsk to see what it says about each of these. I think that's non-destructive. You'd have to research it. I've seen chkdsk see impossibly small NTFS volumes and say "yay! perfect!" and yet it's like 512KB which is definitely not a Windows partition of any sort.