exxos's H5 A3 Build

[stephen_usher]

So, maybe it's a useful diagnostic if consistent?

What would it say initially normally? Are there just bits stuck high or low?

Surely Ejjwj is "Error"?

[JezC]

Any chance that one or more of the data bits are shorted or swapped?

Checking the ASCII codes might pinpoint it.
j 0x6A to r 0x72
w 0x77 to o 0x6F
etc.

Will see if I can spot any pattern later...unless you sort it in the next few hours.
Knowing the normal output should show it quite clearly.

Or even search in the ROM image for "Garbage before"?

[exxos]

By Klehpev R6 Teajy

I think that would say "By Stephan J Leary" so its almost the same..

By Klehpev R6 Teajy
By Stephan J Leary

Looks like some stuff is byte swapped or something, but nothing that big has been done on the H5, its pretty much based on the H4.

EDIT

Code: Select all

       76543210
T = 84  1010100 BAD
L = 76  1001100 OK

a = 97  1100001 OK
e = 101 1100101 OK

r = 114 1110010 OK
j = 106 1101010 BAD

y = 121 1111001 OK

Looks like bits 3 & 4 are inverted

[stephen_usher]

Maybe it's a self-encrypting data bus. ;-)

Say it's a feature... job done! ;-)

P.S. Yes, bits 4 and 5 look swapped. But that can only be in one direction, either writing to RAM or reading from it, not both otherwise it'd be self-correcting.

[exxos]

Removed the RGB/MONO jumper (must have been in mono before :roll: ) and do see a output now...

IMG_5716.JPG

IMG_5717.JPG

EDIT:

Diagrom does not find any RAM errors... Still getting garbage on the serial port though..

IRQ test results in a garbage red screen, can't seem to get out of it now.. but not sure what that test does anyway.. maybe my MFP is duff.

[exxos]

Now IRQ test seems to do something..

IMG_5718.JPG

I will have to connect up my H4 and see what the results are supposed to be...

EDIT: Same on the H4 so that seems OK..

[stephen_usher]

Could you have two data lines going to the MFP swapped?

[exxos]

Could you have two data lines going to the MFP swapped?

Don't see how unless the lib is wrong.

[exxos]

:WTF:

Capture.JPG

Bits 3 & 4 are swapped, wrong pins, but the lib shows the correct order just wrong pins :pullhair:

[Icky]

@exxos It’s always the libs that catch us out. Just like the regulator pins swap on the Sparkalphobia v1