Page 45 of 68
Re: Icky - Progress
Posted: 10 Oct 2018 23:56
by Icky
exxos wrote: 10 Oct 2018 23:52
Also could explain why on the ste the DMA port is delayed a clock cycle before data is allowed on bus...
You really did name this one well on the board Exxos - BAD DMA.
I have just tried part of rpineau suggestions where I took out DRQ as that was made easy by the U51 chip. That still did not recognise drives.
About to also try your suggestions Exxos
Re: Icky - Progress
Posted: 11 Oct 2018 00:01
by exxos
:)
Out of interest.. Lift cs pin 1 of the 1772 and tie to 5v.... Floppy won't work.. But see if that makes the hard drive boot...
Re: Icky - Progress
Posted: 11 Oct 2018 00:07
by rpineau
exxos wrote: 10 Oct 2018 23:52
Also could explain why on the ste the DMA port is delayed a clock cycle before data is allowed on bus...
Yea, I see the 74ls74 to delay the 74ls245 /G pin by 1 clock cycle after /FDCS goes hi could be the solution.
That and the fact that there are buffers isolating the DMA port from the DMA/FDC ..
Re: Icky - Progress
Posted: 11 Oct 2018 00:12
by exxos
rpineau wrote: 11 Oct 2018 00:07
Yea, I see the 74ls74 to delay the 74ls245 /G pin by 1 clock cycle after /FDCS goes hi could be the solution.
That and the fact that there are buffers isolating the DMA port from the DMA/FDC ..
Yep. I did jumper those buffers once and didn't see any difference... But my guess is Atari knew of the problem on stfm and added buffers on ste... But also note the route hdcs via delay as well...
I bet isolating 1772 from DMA bus when 1772 cs goes high might solve that problem also... I'm assuming 1772 is slow to release the bus after cs goes high... Or something like that..
Re: Icky - Progress
Posted: 11 Oct 2018 00:28
by rpineau
/HDCS is route via the 74LS244 , not the 245, so it's only delayed by the 74ls244 propagation time on the STE (/1G and /2G are grounded on the 74LS244).
Only the DMA data pins are going through the 74LS245 and delayed. I think you're right about the FDC being slow to release the bus after /CS goes high.
I wonder if the Ajax is faster and doesn't have that problem.
Re: Icky - Progress
Posted: 11 Oct 2018 00:32
by Icky
exxos wrote: 11 Oct 2018 00:01
:)
Out of interest.. Lift cs pin 1 of the 1772 and tie to 5v.... Floppy won't work.. But see if that makes the hard drive boot...
Nope doesn't boot HDD
Re: Icky - Progress
Posted: 11 Oct 2018 00:37
by exxos
Icky wrote: 11 Oct 2018 00:32
exxos wrote: 11 Oct 2018 00:01
Out of interest.. Lift cs pin 1 of the 1772 and tie to 5v.... Floppy won't work.. But see if that makes the hard drive boot...
Nope doesn't boot HDD
Damn... :( Well we solved a problem nobody has had yet anyway :p
Are you sure you haven't got a iffy socket or something daft ? Resolder it worth a try...
Re: Icky - Progress
Posted: 11 Oct 2018 00:38
by exxos
rpineau wrote: 11 Oct 2018 00:28
I wonder if the Ajax is faster and doesn't have that problem.
Interesting thought..
Re: Icky - Progress
Posted: 11 Oct 2018 07:33
by Icky
exxos wrote: 11 Oct 2018 00:37
Icky wrote: 11 Oct 2018 00:32
Nope doesn't boot HDD
Damn... :( Well we solved a problem nobody has had yet anyway :p
Are you sure you haven't got a iffy socket or something daft ? Resolder it worth a try...
I did think that too. Reason for my slow going last night is that I was going methodically through all the connections checking again for bad solder and making sure everything was connected correctly. Including going over the socket with more solder.
Re: Icky - Progress
Posted: 11 Oct 2018 08:37
by Icky
exxos wrote: 11 Oct 2018 00:38
rpineau wrote: 11 Oct 2018 00:28
I wonder if the Ajax is faster and doesn't have that problem.
Interesting thought..
Might be worth trying to get my hands on one of these to test