ATTO SiliconExpress II - Clone an early Fast/Narrow SCSI II Card?

Trash80toG4

Active Tinkerer
Apr 1, 2022
998
292
63
Bermuda Triangle, NC USA
Remembered my early SCSI II card whilst slumbering in bed and tossed the notion of cloning it out to the gang at the MLA before I woke up enough to think better of it. So here goes:

P1000763.JPG

P1000768.JPG

Wondering if anyone here has any notions about what might transpire on what appears to be a fairly simple kinda thing? At first glance and after just a bit of research, seems to be mostly standard parts bodged together in the pre-ASIC era?

WAG would be that they used a 68000/pair of ROMS to do what was later distilled into ASICs?

Dunno, just gotta find out what you boffins over here think about this bit of craziness?
 

Trash80toG4

Active Tinkerer
Apr 1, 2022
998
292
63
Bermuda Triangle, NC USA
One of the magicians over at the MLA described the 86C05 as a NuBus Controller, but that didn't sound right to me. NuBus Controller is on the Mac's Logic Board or on the 030 PDS->NuBus Adapter in the cases of the IIsi and DuoDock.

86C05 is a multiple platform, programmable Bus Controller, but set up as a relatively simple MUX in this NuBus application, no?

Screenshot 2025-02-10 at 11-17-44 NCR_86C05_DMA_Master_Controller_Mar89.pdf.png



It appears to me that it's 16bit DMA interface would be hooked up to the 68000, which does the heavy lifting? Onboard 68000/ROM/SRAM sub-system boots on power up, allowing the Mac to boot from fixed disk on the Fast/Narrow SCSII interface.


Screenshot 2025-02-10 at 11-17-11 NCR_86C05_DMA_Master_Controller_Mar89.pdf.png


Thanks to @Bolle we have the formulas for the three Pals on what is apparently the BIOS PROM interface.

WAG would be that control interface on the fourth side of the diagram would be hooked up the the 68000 sub-system as well?



edit: forgot the point of this post. Methinks the multiple bus support of 86C05 is crazy overkill, if convenient, for it's function on this card? Way above my sorcerer's apprentice pay grade, but might the function set for this card be reproduced in FPGA without Herculean effort?
 
Last edited: