Atarian Computing's Raven Build

[agranlund]

Slightly concerning with bit12 on one of those chips sometimes ending up 0 instead of 1.
If you erase it, then run the dump command over and over a couple of times, will it always show the same result each time or can it become random on that bit even if you don't erase in between?

Basically wondering that even if you got what looked correct after a programming, it could still end up "randomly" wrong on reads.

The fault pattern appears so consistent with always the same bit within the same byte of each longword that I would probably take an extra look at D12 (and perhaps also D4 in case I got the byteorder wrong in my head) - on both chips because I don't know off the top of my head which one is 0,1,4,5,8,9.... and which one is 2,3,6,7,a,b....

[agranlund]

I've connected the usb-serial cable to the Raven with what I assume to be a null-modem cable. Does it have to be null-modem? If not, I need a gender changer as the usb cable is male.
I'm connecting at 115200bps 8N1 to the lower left port on the Raven.
I only have the ROM simm.

Unless the usb-serial thing on your PC side does something funky then yep you'll need a null-modem cable.
Only rom simm should be enough to get the first sign of life out of the serial port -- it should at least say hello and then get upset and yell at you for not having any ram

[Atarian Computing]

Thanks again @agranlund for bearing with me.

I think I've learnt a valuable lesson with these fine pitch smd components: Never, ever, rely just on visual inspection. It may look perfect but it might not be connected. Always do a physical inspection for every single pin with a needle.

While I haven't tried the ROM SIMM yet (too late today for that), there was a loose pin that looked perfect on the microscope. Unbelievable.

Next order of business, inspect all SMD ICs on the mainboard, pin by pin, tomorrow.

[Atarian Computing]

Sitrep

I have gone through all pins of all smd ICs on the main board. All but one were good.

After fixing the one pin on the ROM SIMM yesterday, the programmer is now behaving well and consistently well. It can program the full TOS with no shutdowns at all. Erasing only produces FFs all the time. So I think we're good on the ROM now.

I have also verified that my USB serial null-cable works. I hooked the cable to my TT and was able to read the test cartridge output on my terminal, albeit at 9600bps. So I'm confident that I should be getting whatever output there is from the Raven.

Except I'm not receiving anything. I do not have anything on the Eiffel socket yet. Should that matter @agranlund? If not, what would you suggest as the steps to check for signs of life? I do have a scope, in fact, a long-time proud owner of an HP 1661CS Logic Analyser/Oscilloscope. Not that I know how to use the LA properly yet but if it comes to it...

[Atarian Computing]

Hmm... no 24Mhz clock. Orientation is correct. Will investimagate...

[Atarian Computing]

Bad soldering job on the crystal.

Screenshot 2025-10-18 165314.png (8.04 KiB) Viewed 304 times

The Raven is trying to fly!!!

[PhilC]

Bad soldering job on the crystal.

Screenshot 2025-10-18 165314.png

The Raven is trying to fly!!!

Nearly there, well done

[Atarian Computing]

Nearly there, well done

Thanks!!

Installed a RAM simm. I'm getting endless scrolling of this:

Screenshot 2025-10-18 165843.png (52.23 KiB) Viewed 302 times

@agranlund got any hints on what that could be? Probably another shitty soldering job... I'll go check it and build the rest of the two simms.

[Atarian Computing]

Made another simm to test. Same behavior. Well, I'll leave it for today. Hopefully someone has some ideas to check first for tomorrow. Big win today though, so I'm happy.

[Atarian Computing]

Ok, this is weird. For shits and giggles I jumpered J102 & J104 to 1x and voila:

Code: Select all

RAVEN060

CPU:   M68060R6
SIMM0: 01000000
SIMM1: 00000000
SIMM2: 00000000
SIMM3: 00200000

Warmboot
InitHeap
InitUart
InitIkbd
InitMfp
InitMidi
InitI2C
InitRtc
InitCfg
InitVbr
InitMonitor
IkbdConnect
Unknown IKBD
InitAtari
InitVbr
InitMmu
InitEmu
InitTos
No TOS detected
StartMonitor

# Raven monitor 20251007 #

 d0: 00000000  d2: 00000000  d4: 00000000  d6: 01000000
 d1: 45544f53  d3: 00581e2a  d5: 00000000  d7: 60008000
 a0: 4000470c  a2: 40004a20  a4: 02000000  a6: 005fffe4
 a1: 00500c00  a3: 400054b8  a5: 20000020  a7: 005fff9c
 pc: 4000d74a  sr: 2710                   usp: 00000001
vbr: 00581a00  tc: 00008210 srp: 00581800 urp: 00581800
dt0: 00000000 dt1: 00000000 it0: 00000000 it1: 00000000
pcr: 04300601 bcr: 00000000 ccr: 00000000

Commands:
  x                 : exit monitor
  r                 : show registers
  pb [addr] {val}   : peek/poke byte
  pw [addr] {val}   : peek/poke word
  pl [addr] {val}   : peek/poke long
  d  [addr] {len}   : dump memory
  a  [addr] {len}   : disassemble
  rtc {clear/reset} : dump/clear/reset rtc
  vga {cmd} {opt}   : screen commands
  kbd {cmd} {opt}   : ikbd commands
  cfg {opt} {val}   : list/get/set option
  run [addr]        : call program at address
  flash             : flash rom image over serial
  reset             : reset computer

Awesome, but why only at 1x? The same issue would happen even without the 68150...