exxos blog - random goings on

[exxos]

:D

IMG_0194.JPG

[Badwolf]

That really was quick!

BW

[exxos]

Unless I have mixed boards up ( don't think so) going back to the board I hacked some bits on to..

https://exxosforum.co.uk/forum/viewtopi ... 087#p93087

Assuming that board was the one that failed on bit11 after I turned it on in the morning.. Now seems to work correctly again :shrug:


But even more bizarre as well testing out the DFB1. A couple of random times it popped up RAM errors..

IMG_0210.JPG

Annoyingly I did not seem to capture a image of when it failed during that ram test previously so I don't know if it failed on the same bits or not :roll:

Restarting the Falcon it came up a different error elsewhere and then was fine again..

While it could be a bad connection does not explain why it totally failed on bit 11 then just leaving it for a couple of days and it magically healed itself... I don't really get what is going on now.. Problem is,I cannot remember if that board I hacked bits on was the one that actually failed with bit11 or not.. But all the other ones on my floor are definitely not the ones because chips are missing etc so either I have lost the board which failed or the Falcon as intermittent faults which would vary each power up.. I don't get it..

EDIT:

I do seem to have left the hacked board on because it is on top of the floppy drive which is generally where I put stuff not to mix it up.. That one board in this post as text on the chip so I think bit11 had failed on the hacked board.. But now it is working again.. crazy.

viewtopic.php?p=93271#p93271

[exxos]

I wish these bloody computers would make up their mind what faults they are going to have :roll:

IMG_0211.JPG

IMG_0212.JPG

IMG_0213.JPG

It is acting more like a bad connection but I don't get why would only be a bad connection on this one particular board. I did already re-solder everything relating to bit 11 on the board itself.

IMG_0214.JPG

It seems to be consistently failing on exactly the same addresses. This seems to get progressively worse from when I first plugged it in earlier.

[exxos]

Actually I have just realised something I think.. It says bad BITS not bad BIT... So just confirmed it to make 100% sure...

111110000000000
432109876543210
101010101010101 (5555)
101110101010101 (5D55)

So bit 11 is "stuck" high. But always on those exact same addresses.

IMG_0215.JPG

EDIT:

Slight address offset this time..

IMG_0216.JPG

[exxos]

My brain has gone to sleep :roll:

So these addresses are what it is failing on..

Code: Select all

326AE6
316AE6
306AE6

221111111111
1098765432109876543210
1100100110101011100110
1100010110101011100110
1100000110101011100110
    ***

How does this translate into the actual address in DRAM as there are only A1-A10 anyway.. One for @Badwolf as our resident DRAM expert :lol:


EDIT:

Ah it's not A0-A10 its MAD0-MAD10 was thinking its the main address bus but its not.

[exxos]

Hmm...

IMG_0220.JPG

IMG_0221.JPG

soooo.

Code: Select all

322AE6
320AE6

221111111111
1098765432109876543210
1100100010101011100110
1100100000101011100110
        *
A13

IMG_0222.JPG

IMG_0223.JPG

:pullhair:

:dizzy:

[exxos]

Seems I went down the wrong rabbit hole there.. I grabbed the DFB1 Because it was easy to put some resistors on it.. Only that was not contributing factor at all.. It was the fact the DFB1 is inserted!

It is repeatable every time. When it is the Falcon only it spews out more RAM errors. when the DFB1 Is inserted, it generally gives only 1-3 errors. Mostly 1 or 2.

So for some bizarre reason @Badwolf Your booster is partly solving my ST-RAM problem :shrug:

:dizzy:

[Badwolf]

How does this translate into the actual address in DRAM as there are only A1-A10 anyway.. One for @Badwolf as our resident DRAM expert :lol:

Wash your mouth out, young man!

So for some bizarre reason @Badwolf Your booster is partly solving my ST-RAM problem :shrug:

Credit duly accepted! :cheer:

How strange, though!

BW

[Badwolf]

I can't think of many effects it could have on the actual DRAM side of things. That's all hidden behind COMBEL and VIDEL. Load on the power rail? It's not exactly nearby either.

BW