Liberator Catbox sig analysis problem. Moved from Vector

Ok let's hope I get all this right this time.

Sorry to the Vector guys. I've now got the raster list added.

Ok so I'm trying to use a C box to T/S my Liberator board, and have run into
a snag. I'm following the exact sequence as given in the manual.

I''ve cut and past the couple of responses I have gotten sofar, so you can
at least see what's been done.

Orignal post

Hi all, in my quest to learn more about using these tools, I've actually
come up with a problem, and could use a little explanation.

I'm working on a liberator board, and was using the atari cat box. The first
two tests in the liberator manual have you check address lines. Well those
checked out good. Next was the Data lines. This is where things fell apart,
and I'm not sure what I'm seeing.

Punch in address 0000, and write AA check the D0-D7 lines going to the ls245
well half the incoming lines are the wrong state, and the output of the
ls245 is non existent. Ie pins 2-9 have no output.

So I hook up my fluke 9010a thinking there may be something wrong with the
catbox, since this is the first time I've used it. Did the same basic setup
wddis grounded, and Φ0 Φ2 shorted together. Do a write to 0000 with AA and
use my logic probe to check. Same results.

Am I to assume Bad memory at location 0000.??

Reply from John

Yes.

Did you do the BUS Test first? That will show if any address or data lines
are misbehaving.

If BUS Test is OK, and you get this problem for RAM @ 0000h, then try the
next RAM set - say @ 0800h or 1000h (if RAM present - check Memory Map). If
other RAM all checks OK, or ANY RAM checks OK, then the RAM @ 0000h is
certainly suspect.

If the RAM is 4-bit, then which bits are locked will tell you which RAM to
replace.

John :-#)#

Response

Well I was following the seqeunce given in the manual, which is Address
lines, then Data lines, then Ram etc..

I'll run the bus sigs now, and see what I get.

I hooked up my fluke 9010 and did a bus check, I'm assuming that would be
the easiest/fastest way to do a busy check on the board. Results came back
good.

Now fi I read your email correctly, you want me to use a different memory
location. That makes sense.

However the memory maps have always confused me a bit.

The game is using AM9016EPC memory, which I'm having a problem finding a
datasheet for atm.

[MAME]

          Liberator Memory Map (for the main set, the other one is
rearranged)
(from the schematics/manual)

HEX R/W D7 D6 D5 D4 D3 D2 D2 D0 function
---------+-----+------------------------+------------------------
     0000 D D D D D D D D XCOORD
     0001 D D D D D D D D YCOORD
     0002 D D D BIT MODE DATA
---------+-----+------------------------+------------------------
     0003-033F D D D D D D D D Working RAM
     0340-3D3F D D D D D D D D Screen RAM
     3D40-3FFF D D D D D D D D Working RAM
---------+-----+------------------------+------------------------
     4000-403F R D D D D D D D D EARD* read from non-volatile
memory
---------+-----+------------------------+------------------------
     5000 R D coin AUX (CTRLD* set low)
     5000 R D coin LEFT (CTRLD* set low)
     5000 R D coin RIGHT (CTRLD* set low)
     5000 R D SLAM (CTRLD* set low)
     5000 R D SPARE (CTRLD* set low)
     5000 R D SPARE (CTRLD* set low)
     5000 R D COCKTAIL (CTRLD* set low)
     5000 R D SELF-TEST (CTRLD* set low)
     5000 R D D D D HDIR (CTRLD* set high)
     5000 R D D D D VDIR (CTRLD* set high)
---------+-----+------------------------+------------------------
     5001 R D SHIELD 2
     5001 R D SHIELD 1
     5001 R D FIRE 2
     5001 R D FIRE 1
     5001 R D SPARE (CTRLD* set low)
     5001 R D START 2
     5001 R D START 1
     5001 R D VBLANK
---------+-----+------------------------+------------------------
     6000-600F W D D D D base_ram*
     6200-621F W D D D D D D D D COLORAM*
     6400 W INTACK*
     6600 W D D D D EARCON
     6800 W D D D D D D D D STARTLG (planet frame)
     6A00 W WDOG*
---------+-----+------------------------+------------------------
     6C00 W D START LED 1
     6C01 W D START LED 2
     6C02 W D TBSWP*
     6C03 W D SPARE
     6C04 W D CTRLD*
     6C05 W D COINCNTRR
     6C06 W D COINCNTRL
     6C07 W D PLANET
---------+-----+------------------------+------------------------
     6E00-6E3F W D D D D D D D D EARWR*
     7000-701F D D D D D D D D IOS2* (Pokey 2)
     7800-781F D D D D D D D D IOS1* (Pokey 1)
     8000-EFFF R D D D D D D D D ROM
-----------------------------------------------------------------

Response John

According to this map 0000h and 0001h are x& ycoord and as such may not
store data, the RAM starts at 0003h through 3fff.

I would run the RAM test from 0003h to 07ffh, then 0800h to 0fffh, then
1000h to 17ffh and so on.

Does the game have aworking self test?

John :-#)#

Reply

Well it would if I could get the game to boot. It's currently stuck just
displaying a partial planet, and a yellow background.

http://lh3.ggpht.com/_s5vOp-dzA9Y/Swx7YffJh7I/AAAAAAAADkk/MKf2Jv6n7u8/s640/AFL%20016.jpg

Hence the starting of trying to follow the catbox troubleshooting guide in
the liberator manual.

ok just ran memory test using the fluke. going from 0003h-3fffh.. Test pass
with no problems. Very interesting to watch if clear the monitor as is reads
and writes to memory. It starts at the top and works it's way to the bottom
clearing out the static image that I sent the picture of earlier.

Thanks,

Kevin

Repsone from Mark

Kevin,

The signals you are trying to test D(7:0) come from the catbox/fluke itself,
so looking at them doesn't tell you much about your board. Another words,
you're looking at the output of the CB and could make the same measurement
with the board disconnected.

On the fluke, these signals will only be valid for one cycle, so you
couldn't probe them manually in the manner you are describing (the level
you're looking for would be long gone by the time you put your probe was on
the net).

If you have a fluke and a CB, which it sounds like you do, use the fluke. If
you have a fluke, the only reason to use the CB is for signature analysis.

Moving to a higher level, what is the purpose (for you) in doing this test?
There is no ram at that address (0x0), so you won't be able to read back
anything you write. If you're trying to run a ram test, you might try a
higher address (e.g. >= 0x4).

Also, this is a raster game and you are posting to the vectorlist.

Reply

Well I'm trying to do the signature analysis that is provided in the manual
using a CB. The manual states to write AA to 0000. I realize the signals
come from the CB, and I'm wondering why they are incorrect. The reason for
doing the test is to confirm that I don't have a dataline problem.

Thanks,

Kevin

So that's where I'm at.

---------------------------------------------------------------------------
** Unsubscribe, subscribe, or view the archives at http://www.vectorlist.org
** Please direct other questions, comments, or problems to chris@westnet.com
Received on Sat Nov 28 13:00:56 2009