exxos's Atari, Amiga & retro forum

Thought I would start a new topic on this because I've just received @PhilC and @Steve STE536 for testing / debugging..

More to come later hopefully...

Looks like @Steve's board is up first.

Observations so far.. The cap on the bottom wasn't soldered. Missing ROM select jumper..

Oddly my TOSCOPY program sometimes passes, other times not. If EMUTOS gets to desktop, changing resolution causes an bus error..


So I will solder a 47pF cap on next and see what happens...

I've flashed "CURRENT5" firmware to make sure its using the latest "Stable" firmware.

EMUTOS let me change resolution and its now running YAARTTT...

OK so it passed several tests then started scrolling errors...


So something still amiss somewhere...

$FFBF = 1111 1111 1011 1111
$BFBF = 1011 1111 1011 1111

Just covering bases. I reflowed the PLD, RAM buffers, CPU.. but no change.

It always fails a few seconds into 4040 test. I half suspect the RAM is faulty... But surely it would have failed earlier...

$04BFFF00 is within the very last kilobyte of the 60MB TTRAM range....I think...

If it is the SDRAM.. which chip.... Or is both faulty on the same address (unlikely ?) Or some other problem....

I put my latest "somewhat working" firmware in, now its "stuck" on this for an awfully long time.. (or maybe I am simply massively impatient..) But the cursor is still flashing and so is the keyboard lights so it's not like it's completely locked up...


But I will disable my ROM shadow logic entirely in see if that improves anything......

@exxos put it in the ultrasonic cleaner for half an hour. With those buffers I found that it's possible to miss the smallest of whiskers of solder. I only spotted it after a dose of cleaning and the microscope at ridiculous resolution.

And if it's still got winbond ram, swap it for those Alliance ones you have?

PhilC wrote: 20 May 2025 14:18 @exxos put it in the ultrasonic cleaner for half an hour. With those buffers I found that it's possible to miss the smallest of whiskers of solder. I only spotted it after a dose of cleaning and the microscope at ridiculous resolution.

And if it's still got winbond ram, swap it for those Alliance ones you have?

Setting up my UScleaner takes ages as it's huge :p

I've just flashed CURRENT6 firmware and let it run again... It's taking about 10mins to do a test.. BUT I guess it's likely as I've not got FAST ROM at the moment...

I'll run MAPROM , might speed things up a bit...

EDIT:

Still taking ages... About 10mins per test aside from own address test which completes in half the time..
I've never really taken notice how long a full pass takes...

I'll test my own board later ... See if that takes the same kinda time..

Connected up my second test machine with my original STE536.. Its got to 4th test and only been going like 2mins.. So @Steve's board if running like 10x slow than it should be :shrug: But it wasn't earlier.. I wonder if me reflowing the PLD has shifted something.. in a not good way..

Will have to leave it on to see if it errors on the 4040 test.. which may take a couple hours at this rate..

It passed that time albeit very slowly...


The question is why is running incredibly slowly now...

FYI the capacitor *was* soldered, it probably broke off in my handling / packing / posting.