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

[yurif74]

Hello, i've the same exact problem, but i doubt is the power supply, because the problem persist even if i detach all hardware and leave only the fastata mkvi controller and the hdd, with a mere 3630 cpu card..
However changing psu has fixed your problem?
Juri

[yurif74]

The incredible thing is that my sata-ide adapter has never work too, and i had to rplace the 80pin ide cable included in fast ata card... elbox never replied my emails! For sure this is the first&last item i bought from elbox!

[yurif74]

Yes i noticed that the ide-sata converter is a piece of chinese s**t, however it work perfectly on windows machine. Just for curiosity i tried another ide-sata converter i have with fastata, result: amiga 4000 doesn't even boot!
I discovered that the master channel 0 doesn't work anymore (de facto it has never worked)

[yurif74]

Hello i do not have a mediator pci, i'm using it on a standard a4000 daughterboard. As far as i know (reading on elbox site) v8.0 have better support for cf drives.

[yurif74]

@slaapliedje:
Hello! Any news?
Juri

[yurif74]

Do you use an adaptor to connect the psu to a4000 mainboard or have you simply connected +5,+12, -12, GND and FAIL lines?

[yurif74]

...When I push the power button the fan comes on, I see the IDE to CF Card thing light up for a second, as long as the CD-ROM light (so it gets at least a second of power) and then nothing.

what exactly mean "and then nothing"? the psu turn off? or it simply stays on with amiga black screen?

some new psu have a 24 pin connector instead 20 pin atx connector, afaik the amigakit adaptor has 20 pin connector, have you connected it correctly? (24 <-> 20 pin)

Juri

[yurif74]

at this point i suggest you to unplug the power connector from a4000 mb and check voltages on pins, original psu and atx psu (with adaptor)
maybe it have some sort of protection if there is no energy drain from the 3.3v it shut down, i guess.

[yurif74]

Amiga 4000 does NEED the pwr_ok signal (reported _fail on a4000 schematics)
I noticed that the atx psu pwr_ok line is "only" 3.66 volts.. afaik it should be a logic level 1 (5 volt)
this signal tell to the amiga that the psu is operating good.
your atx do not need the 3.3v to be connected at all, because it doesn't shut down when you tested voltage on a4000 power connector with hdd, fdd and cd connected. They required only +5v and +12v to operate.
juri

[yurif74]

Are there any news? did you tried a different psu with success??

Juri

[yurif74]

Different PSU, no success.  Though, at least I did get it to boot, but the FastATA still has the issue.

So it isn't psu related and i can confirm it!
i've a Cyberstorm MK1 with 68060@50mhz (no overclock) and i'm having your problem (can't go further PIO MODE 3) yesterday happened something really strange, a friend came to me with a standard A3640 to try it (his amiga have serious problems and he want to check if the CPU card was ok) tested ok, his A3640 cpu card was ok, so he asked me if i fixed my problem with fastata, and while talking i showed to him the problem, i setted to PIO MODE 5 and... nothing!! WHAT?? YES!! with a stock A3640 Fastata works PERFECTLY in PIO MODE 5!! (i'm getting a little bit under 6mb/s of transfer rate!) then i reinserted my CSMK1.. amiga locked.. after few second.. i had to re insert the A3640 to switch back to PIO MODE 3! now my amiga is working with the CSMK1 in pio mode 3... too bad!

[yurif74]

...I bet you anything it's incompatible with something on the Cyberstorm MK1 then, either the 060 library or just the way it works.  Anyone have any ideas on that?...

from elbox:
•                      Automatic                      processor type recognition
                    The FastATA 4000 MK-VI CF/SATA controller automatically detects                      type of processor and optimises and modifies some software                      routines in order to achieve the highest possible data transfer                      rate between the controller and the computer memory.

now afaik workbench recognize the 68060 as 68040 until setpatch is called... i've the fastata.driver command BEFORE the setpatch command..
it's very late here now, but tomorrow i've some free time and i will do some experiments.. then i let know

Juri

[yurif74]

I will somehow find a way to pay you if that is the fix!  I'll give it a shot hopefully in an hour or so.  Stuck at work 'til then.  Though if that's the fix, I'm going to be pissed off at Elbox.  I would think the installer would put FastATA.Driver after setpatch anyhow.

slaapliedje

slaapliedjeeeeeeeeee

put beer on ice! I placed fastata.driver after setpatch and cpu command and.. fixed!

[yurif74]

afrer some testing i had to switch back to PIO 4 because with pio mode 5 even if drivetest report "OK,no errors" i have a lot of checksum errors and locks.. with PIO 4 all appear to be ok. one little difference is that my A4000 mount a fastata mkvi with rom 7.0 (but i do not think that's the problem) have you tried to remove setpatch command? just for try, or another version of setpatch/fastata.driver
i've a lot of sh*it that loads on my amiga 4000: mmu, mui, poseidon, cybervision, modified cyberstorm libraries.. when i've some free time i will try to unload all, and put another hdd with clean workbench install and only fastata driver (ofcourse stock cyberstorm libraries from install floppy) i would try os3.9 too