CPC-464 freezes on start-up with black background colour

[Tezz]

Thanks all, my first thought was to check it out with the Dandanator. Rob has kindly offered to send a diagnostics ROM. The cheapest I can see the Dandanator mini is roughly £39 including shipping from Spain. That way I can hopefully check out the ROM Insitu along with the RAM.

[Tezz]

It's actually £15 shipping to the UK so it works out no cheaper than the ones listed on eBay. I've not yet ordered one.

[Tezz]

I'm just getting back onto this today, I think the first step will be desoldering the ROM and checking it out making use of the TL866II+. I have one more 28 pin socket left so I'll have the board ready for a replacement or returning the original back if all looks good there 

[Tezz]

I've eventually read the ROM using my TL866II+ and all appears to fine with it. I compared my 32kb ROM dump against the standard BASIC/OS downloaded from the wiki using binary compare via windows FC.EXE. I can upload the dump if that's allowed on the forum. So, the issue with this board lies elsewhere, perhaps still a RAM promblem?

[Rabs]

Now you have the TL866II+ setup you could create a Lower Diagnostics ROM, to check the RAM.

[Tezz]

Now you have the TL866II+ setup you could create a Lower Diagnostics ROM, to check the RAM.

Definitely the next move!

[Tezz]

I've received my EPROMs and flashed Noel's diagnostics ROM. The CPC starts with a beep, does bit pattern, double beep then the diagnostics display. There's nothing from the keyboard input although as the diagnostics is in place of the firmware in the ROM I assume the diagnostics ends once it's passed through a RAM test and of course reports unknown cpc. I'm unfamiliar with it however, I should read the sources.

what should be the next move?

[Rabs]

Keyboard should still work and is that screen in black and white? How weird. RAM test has passed. AY sound chip is working as you heard the beep and must have been programmed via the 8255 PIO.

Ah ah you have no CRTC type being displayed.

[Tezz]

Keyboard should still work and is that screen in black and white? How weird. RAM test has passed. AY sound chip is working as you heard the beep and must have been programmed via the 8255 PIO.

Ah ah you have no CRTC type being displayed.

Hi, yes it does start in black and white just as the standard ROM did although again changing the TV source from and to SCART the colour output is there.

It'll certainly be interesting to finally discover what's been causing the issue!

[McArti0]

Keyboard was tested and not work? 

[Rabs]

A clue, I think, is that the Diagnostic ROM is not reporting the CRTC type.

You should be seeing something like this. See how the CRTC type is reported as 00. I think Diagnostic ROM is just executing a simple IN against the CRTC status register and yet it is hanging, or looks like it from your screen shot.

You cannot view this attachment.

[Rabs]

Looking at the Diagnostic ROM Code, it writes to Reg 12 of the CRTC before reading port B on the 8255 PIO waiting for VSYNC before then going on to read the status register and determine what type of CRTC it is. At least I think so. But VSYNC must be ok otherwise the screen would be corrupt. So is it stuck here? PIO? But at this point I am starting to guess...a lot...

Code Select Expand

GetCRTCType:
    ld    bc,#bc0c        ; select reg 12 (R/W)
    out    (c),c
    ld    bc,#bd00+%0110100    ; write a value
    out    (c),c

;    call    wVb
    ld     b,#f5            ; wait Vbl
@vbLoop1
    in    a,(c)
    rra
    jr    c,@vbLoop1
@vbLoop2
    in    a,(c)
    rra
    jr    nc,@vbLoop2

    ld    b,#be            ; read from status register
    in    a,(c)
    ld    d,a

[McArti0]

Or. Something wrong was happend with CRTC, A14 or nand IC110. When reading PA F4 reading FF from CRTC too.

[Tezz]

A clue, I think, is that the Diagnostic ROM is not reporting the CRTC type.

You should be seeing something like this. See how the CRTC type is reported as 00. I think Diagnostic ROM is just executing a simple IN against the CRTC status register and yet it is hanging, or looks like it from your screen shot.

You cannot view this attachment.

Ah yes, thanks for the image, that's helpful to see that it's hung at that point. That might explain the black and white output as it hangs with the standard ROM. I'll look to replacing the CRTC next. I have some 40 pin sockets. Is there a good place to buy UM6845R?

[Rabs]

I am not sure it is the CRTC? All it does is count VSYNC, HSYNC, Display Enable and refresh the RAM.

The registers which form the counters in the CRTC are being setup correctly, otherwise you would not have a display.

And there is not a lot between the CPU and CRTC in terms of logic.

[Tezz]

Keyboard was tested and not work?

Hi, yes I tested the keyboard on my long board and all was fine there.

[Rabs]

As @McArti0 indicated could be IC110 but equally could be IC112. If it was IC110 then I would expect the write to fail, as well as the read, and looks like that is not the case as the display is sort of working. But if IC112 failed? On a single gate, really? Guessing here without a scope. And then there is the black and white screen. What is all that about?

You cannot view this attachment.

You cannot view this attachment.

[McArti0]

No on other board but on DiagROM.

Check wire LK1, LK2 - 4. I should to change brand on screen and  50/60 Hz set. PB 8255 working or not.

[Tezz]

I am not sure it is the CRTC? All it does is count VSYNC, HSYNC, Display Enable and refresh the RAM.

The registers which form the counters in the CRTC are being setup correctly, otherwise you would not have a display.

And there is not a lot between the CPU and CRTC in terms of logic.

Ok, I'll take a look at the schematics for a better understanding. It would have been nice if it was down to a logic IC

[McArti0]

And then there is the black and white screen.

I think its 60 Hz and bad recognize NTSC.

[Rabs]

I have a test board with most of the BIG ICs socketed, except for IC112 and IC110    ,otherwise I could have simulated some gate failures for you to see if I get the same symptoms. 

If you are still stuck by the weekend, I can do that.

[Tezz]

I have a test board with most of the BIG ICs socketed, except for IC112 and IC110    ,otherwise I could have simulated some gate failures for you to see if I get the same symptoms. 

If you are still stuck by the weekend, I can do that.

That would've been great to know if the two ICs were socketed! 

[Rabs]

Still not sure because if the IORD was at fault, would the CPU just not read random data and not not crash or hang?

[Rabs]

On the off chance have you got or tried a spare Z80?

[McArti0]

Check wire LK1, LK2 - 4

Please check this!

LK1, LK2, LK3, LK4. or 8255 pin 19,20,21,22 to ground and turn on CPC. changing Brand on screen

Test Does this computer read any IO!!!