Grzyb wrote:Deksor wrote:As far as I know, ATA is mostly 16 bit ISA on a cable, so I believe XTA is more or less 8-bit ISA on a cable, so ATA should be backward compatible with XTA ?
ATA is more than 16-bit ISA on a cable, it's also the register set compatible with that WDC-based controller from IBM 5170.
XTA has different register set.
Well, yes and no. 😀 The original hardware for ATA-0 aka IDE or AT-Bus (not EIDE) is just a glue logic without any intelligence (just buffers/an address logic).
Here, the smart part resides in the PC/AT BIOS (speaks WD1003 language) and the HDD electronics itself (emulates WD100x logic).
That's in contrast to on-board ATA controllers found on PCI motherboards. They contain parts of the ATA logic.
The older MFM/RLL Shugart interfaces, on the other hand, incorperating a real WD100x interface chip (8-Bit boards often had an on-board BIOS).
Also, ESDI was an in-between of IDE and old Shugart MFM/RLL interfaces.
It contained about 2/3 of its intelligence on the controller and 1/3 on the fixed disk.
AT-Bus/ATA-0/IDE also was used for other tasks, such as interfacing EPROM programmers.
The drawback in comparison to full ISA is, that it is limited to a fixed I/O range and I/O ports (how was Multi-Word DMA done, then ? 😕 ).
Edit: Some typos fixed.
Edit: If you meant to say "ATA as a whole is more than 16-bit ISA on a cable", then you're fully correct, of course. My bad. 😊
"Time, it seems, doesn't flow. For some it's fast, for some it's slow.
In what to one race is no time at all, another race can rise and fall..." - The Minstrel
//My video channel//