Why didn't Seagate reintroduce dual-drive emulation?
Posted: August 1st, 2013, 1:54 amSome Seagate ATA drives from the mid-nineties had a feature which, when activated by the required jumper combination, split a single high-capacity physical drive into two smaller logical drives, one as the "master" and the other as "slave" (I put those terms in quotes because, contrary to popular belief, neither drive controls the other). It was intended as a short-term workaround for the 1,024 cylinder limit.
Of course, there were more barriers to come, so there was an opportunity to reintroduce further versions of the feature, but they missed it. (Of course, now with SATA it has little relevance, though a functionally similar feature could theoretically be provided on a SATA drive by emulating a "port multiplier", but that requires host support, which not all SATA ports have.)
Conceptually:
Device 0 ("master") - jumper 7-8
Device 1 ("slave") - no jumpers (or 5-7 which does nothing as both of those are ground)
Cable select - jumper 5-6
Force slave present - jumper 7-8 and 5-6
Limit capacity - jumper 1-2
Pins 3 and 4 are not for configuration, but are a serial terminal interface which is only to be accessed for data recovery. There are no other possibilities using the remaining pins, although, could they have programmed the controller to switch into split mode upon receiving the same signal it transmitted?
Of course, there were more barriers to come, so there was an opportunity to reintroduce further versions of the feature, but they missed it. (Of course, now with SATA it has little relevance, though a functionally similar feature could theoretically be provided on a SATA drive by emulating a "port multiplier", but that requires host support, which not all SATA ports have.)
Conceptually:
- 4,096 cylinder limit - split a 4.2GB drive into 2×2.1GB
- 6,322 cylinder limit - split a 6.5GB drive into 2×3.2GB
- Int13h address limit - split a 17GB drive into 2×8.4GB
- 65,536 cylinder limit - split a 60GB drive into 2×30GB
- 28-bit LBA limit - split a 250GB drive into 2×125GB
Device 0 ("master") - jumper 7-8
Device 1 ("slave") - no jumpers (or 5-7 which does nothing as both of those are ground)
Cable select - jumper 5-6
Force slave present - jumper 7-8 and 5-6
Limit capacity - jumper 1-2
Pins 3 and 4 are not for configuration, but are a serial terminal interface which is only to be accessed for data recovery. There are no other possibilities using the remaining pins, although, could they have programmed the controller to switch into split mode upon receiving the same signal it transmitted?