exxos's Atari, Amiga & retro forum

I am slightly confused over the CS0 and CS1 signals which are linked to A5.

According to the ATA spec, they are used to select upper or lower 8bit bank select in the drive. Fair enough.. but the CPU has a 16bit bus, so why doesn't the data transfer as a 16bit block ?!

The CPU would actually have to do 2 bus cycles I assume to change the A5 for upper and lower 8bits ? Would seem a bit pointless having a 16bit drive and interface, and yet its looking like its driven as a 8bit device :shrug:

EDIT:

I think half the confusion is coming from this site...
http://blog.retroleum.co.uk/electronics ... interface/

It says

When /CS0 goes low, registers 0-7 are accessed, and when /CS1 goes low 8-15 are accessed.

Then later says.

Tip: You typically dont need to access IDE registers 8 to 15 so /CS1 can just be pulled high via a resistor.

So I am assuming the transfer is actually 16bit and CS1 may not actually get used for anything...

EDIT2:

OK http://citeseerx.ist.psu.edu/viewdoc/do ... 1&type=pdf

Chip Select (CS0:1) - Selects the command block or control block registers.

Why couldn't some site or document have actually just said that :roll:

Do not mistake register number to data bit on the bus
Register 0 to 7 are 16 bit register, and so are register 8 to 15 and they are accessed as 16 bit registers even if only the lower 8 bit contain actual data for some register.
So CS0 and CS1 select the register bank, not the bus width.

Small update:

We have been playing with a DEV IDE board and have had the IDE interface being recognised using the PLD circuit we have been tweaking.


Now we are moving onto a more final version that will also have TOS 2.06 Rom onboard.

exxos really like that board! :bravo:

Sign me up already :D

Oooh, nice.

Oh my, that looks nice!

We have been doing a lot of testing and finding out some interesting things around the PIO modes and driver interactions with IDE devices at initialisation.

However the tests we are performing to check the interface and devices are working have been captured in this video.



This is our definition of a working IDE Interface and Device as test steps. If we can get to the bottom of the list then the IDE interface and Device is working. Definition of Device is SD IDE Adaptor with card or CF IDE Adaptor with card:

1. Boot up ST
2. Boots off of Floppy with HDDriver 10.14
3. Driver sees IDE interface and Device. Device is displayed correctly
4. Load HDDUtil from floppy
5. Partition Device into 9 partitions that are <500MB as TOS partitions
6. Reboot after partitioning
7. Boots off of Floppy with HDDriver 10.14
8. Driver sees IDE interface and Device. Device is displayed correctly
9. Partitions are recognised by the driver
10. Once at desktop go to preferences and change res to medium, install devices and make sure all partitions are present and save desktop
11. Open C: and create a folder called HDD and go into it
12. Open A: which is still the HDDriver 10.14 files and copy all files and folders from A: into C:\HDD
13. Go to C:\HDD and run HDDUtil to see if it loads correctly from HDD
14. Once in HDDUtil install HDDriver on C:
15. Go back to C: and copy C:\HDD to D:, E: & F:
16. Delete the folder HDD from D;, E: & F:
17. Reset ST and it should boot directly from Device and not floppy

Update: We have ROM.

This is the latest version of the board which has a few tweaks and inclusion of ROM so you can select which version of TOS to run.


We now need to send off to get a few boards made so we can test further before looking at the final version of this board.

Let me know if you need another Guinea pig.