Arrrgh! (FastATA 4000 MK-VI 8.0 and issues with keyboard.)

[danbeaver]

Hmm,

OK, replace the CF connector (and card if it fails on a PC), or pm me your address and I'll send you a CF connector and 64MB CF card.
Assume the motherboard is not at fault.
Find another ATX-style to test to see if it is the Silverstone (which is a fornicating rip off if it is the Silverstone).

[danbeaver]

Well, I have the same card (ROM as you do,  and I do have a 300GB HDD I can get a "PIO 5" to work with, BUT the drive speed is less than 10 MB/s.  I have not tried an SSD, but it isn't all that important, well at least to me...

[danbeaver]

Now I didn't say it would make sense, I just thought you might squeak out another byte per second over sadly slow interface.  I don't know why UWSCSI was limited to just the Cyberstorm MK 3 & CSPPC.  35 MB/s is nothing to laugh off in the Classic Amiga field, and has to be bit easier to design than an accelerator.

[danbeaver]

I believe that if it seems fast, then it is fast.  I test stuff and report it just so it can be found on the Web, (not to make my phallus size seem larger).  However, a U320 15K rotation speed HDD will get 36 MB/s out of a CSPPC/CS MK 3.  Those numbers don't mean much in real world applications.  I wanted the FastATA 4000 to backup partitions to a CF card faster than the standard A4000T IDE.

[danbeaver]

Well, the bus and interface limit transfer speeds.  An accelerator with an onboard HDD controller bypasses the Zorro bus, just as their ram does.  In theory the CS MK 3 & CSPPC's local PCI bus for the G Rex 4000 and CVPPC does the same.

The Zorro III bus in theory could transfer 150 MB/s, but in reality the Buster chip limited it to about 13.5 MB/s.