Re: I: STAR CASTLE PCB with a "little" problem

I would initially suspect a problem with the ROMs as the fault seems to
be specific to the starships. However, if any variables used in the
starship drawing routines are held in a specific place in RAM then there
could be a problem with the RAMs, try using RAM from another board. As
William says, the game would not work properly if one of the ALU logic
chips had failed and all the other vectors draw OK.

Chris

On 08/06/2011 04:00, William Boucher wrote:
> I thought that your video was very interesting and it certainly
> demonstrated the problem. The rotation of a ship or ring is a math
> intensive and relatively complex procedure that involves EPROM lookup
> tables (for both the basic object drawing data and Sin/Cos data), the
> ram chips (L14, M14, N14), the '181 ALU chips (L6, M6, N6), and many
> other ICs (assorted gates, latches, data selectors, etc.). Since so
> much of the general game operation also depends on many of the same
> chips that deal with all of the rotation calculations, I would not
> expect the game to run properly if one of them was failing. I would
> expect the game to crash. Since you say that you tried several chips
> (EPROMS, ALUs, RAMs) in another board and they worked then we'll
> consider them proven okay. This also eliminates the 34-conductor
> ribbon cable and the monitor as a source of trouble. The condition of
> the chip sockets on the problem board may remain an issue but like I
> said already, I would expect a crash if this were the case because
> such a fault would mess up a lot of things, not just rotation. If I
> were to assume that the sockets are okay, and since the game play is
> okay and the problem appears to only effect the picture, then I might
> suspect the video data output latch chips (74LS377 at M2, P2, S2).
>
> When I watched the video, I got the impression that the misplaced
> vertex was shifted mainly (or entirely) in the "Y" axis. I could be
> wrong about this since I only watched it a couple of times however for
> this moment I'll go with it. The 12-bit "Y" data is output as 3
> nibbles (4 bits each) where each nibble is latched by one of the
> 74LS377 chips M2, P2, S2 where M2 has the least significant nibble and
> S2 has the most significant nibble. Since the positional error is
> very small, I'd suspect a very low order magnitude bit, so I'd replace
> M2 first. You might want to check the condition of the J2 pins 1
> through 13, the traces leading to the pins, and the solder on the pins.
>
>
>
>
>
> William Boucher
> http://www.biltronix.com
>
>
> ----- Original Message ----- From: "vernimark" <vernimark@vernimark.com>
> To: <vectorlist@vectorlist.org>
> Sent: Tuesday, June 07, 2011 8:47 PM
> Subject: VECTOR: I: STAR CASTLE PCB with a "little" problem
>
>
>>
>> http://www.youtube.com/watch?v=n9h0c9qYkZM
>> (sorry for the bad quality)
>>
>> In some starship positions vecrots are drawn with one wrong vertex.
>> The same
>> happens with the castle and other starships. Any Idea?
>> 74181: ok
>> ROMs: ok
>> (on another PCB RAMs and ROMs work well)
>>
>> Thank you,
>> vernimark
>>
>>
>> ---------------------------------------------------------------------------
>>
>> ** Unsubscribe, subscribe, or view the archives at
>> http://www.vectorlist.org
>> ** Please direct other questions, comments, or problems to
>> chris@westnet.com
>
> ---------------------------------------------------------------------------
>
> ** Unsubscribe, subscribe, or view the archives at
> http://www.vectorlist.org
> ** Please direct other questions, comments, or problems to
> chris@westnet.com

---------------------------------------------------------------------------
** Unsubscribe, subscribe, or view the archives at http://www.vectorlist.org
** Please direct other questions, comments, or problems to chris@westnet.com
Received on Wed Jun 8 04:04:52 2011