REV 3 - REV 5 - The beginning (ST536)

[stephen_usher]

@exxos disable the ram test routine, it's not an essential and YAART does a better job. Itll also speed up boot times

But surely the crash is a symptom of a bigger issue? By just skipping the memory test you're masking the issue rather than fixing it.

[PhilC]

@exxos disable the ram test routine, it's not an essential and YAART does a better job. Itll also speed up boot times

But surely the crash is a symptom of a bigger issue? By just skipping the memory test you're masking the issue rather than fixing it.

Exactly, if it works without it, then the error is in the ram test routine, so you could insert coding to indicate exactly which point it's at in the routine, or waits for a key press at certain points etc

[exxos]

@exxos disable the ram test routine, it's not an essential and YAART does a better job. Itll also speed up boot times

I think what @stephen_usher says is key.. We can't blame the TTram test routine at this point, it might just be a trigger of a deeper problem.. in which case, other random software could cause resets just the same.. So removing the TTram test would be brushing the problem under the rug at this point...

[exxos]

I gated BUS ERROR but did not change anything.

I disabled the ROM enable (so doesn't run ROM from TTram) , let it get to TTram test, then ran my original ROM to TTram copy code (I removed the ROM copy code so it only verifies) and ...

IMG_4601.JPG

So the ROM shadow in TTram doesn't get corrupted.. its something else going on...

Theres only really the ROM shadow logic which is simple, and its ran overnight.. So its not like its unstable. But I can't figure out how it gets from testing TTram to crashing a few seconds later..

Oddly one of my latest ST536 boards doesn't run with the new firmware, it shows TTram corruption.. So that could likely be a clue..

EDIT:
It passes TTram test with fast slew.. I was using slow before..

The latested board (dual regulators etc) doesn't seem to have this odd ROM copy problem at all...

So :

REV 5 not ok (this one is @coonsgm ) board, So I don't know if there are other issues at play yet, or its just the board rev.
REV 5.50 (hacked up res mods) seems ok
REV 5.50 (no mods other than 3.3V wire) seems ok
REV 5.62 seems ok

EDIT2:

So @coonsgm board only seems to act up when TTram is enabled. But I just noticed it has Alience RAM.. so that's probably why..

[exxos]

So the difference between REV5.00 and REV5.50 , the most likely significant change is the addition of the clock patch.. I need to find the little patch boards I had done and replace the clock to see if that fixes the problem or not.. That will be fun to find....

[exxos]

The R5 board is perplexing.

Changed the SDRAM from alience to winbond.. resoldered everything 3 times.. Hacked on a 3.8V reg for the OSC.. nothing has made any difference.

Though now it hardly gets past checking rom CRC and resets. So its got worse as times gone on.. :roll:

IMG_4606.JPG

IMG_4607.JPG

IMG_4608.JPG

IMG_4609.JPG

IMG_4610.JPG

IMG_4611.JPG

Its going in the US cleaner as its just a sticky mess. But this board either has some bad via, or something changed on later boards which fixed the issue. But no idea what, as wasn't much difference from 5.00 to 5.50.. Mostly the IDE connector was moved and the OSC had its own reg.. but not much else past that from what I can see.

[exxos]

I think the more I work on this project, the more things make no sense :stars:

The rev 5.00 board.. boots up like this ..

IMG_4617.JPG

IMG_4618.JPG

and yet...

IMG_4616.JPG

I'm puzzled over the rev 5.50. It seems to lockup accessing floppy half the time. The later 5.62 board seems fine so far. No one has a lot more Better Grounding on the Board.. So are really just wonder am I just fighting PCB -related issues at this point.. And all this weirdness is just simply down to signal integrity.. Is very difficult to tell because sometimes it will work perfectly fine and I cannot seem to find a "trigger" to what changes.

[exxos]

Can you believe this.. All I did was jumper CDIS..

IMG_4619.JPG

IMG_4621.JPG

IMG_4622.JPG

[PhilC]

Have you tried a different CPU to rule that out?

[exxos]

Have you tried a different CPU to rule that out?

I Just swapped the CPU just to doubly make sure.. (was swapping CPU yesterday in fact) . Made no odds :(

I think this problem might be backing onto a similar thing when I thought 32bit ROM copy wasn't working.

My rev 5.50 seems to have fits of random oddness.. I've not seen the problem yet on the 5.62 board, but haven't done much testing with that yet.

There must be some odd timing issue somewhere, or the cache malfunctions without good enough caps.. no idea.