What was I thinking! Moon Cresta Bootleg Journey

N

NivagSwerdna

Active member
Feedback
1 (100%)
Credits
744CR
Looks like Sprites at certain Y values do not render correctly... (and maybe sprite colour is leaking into character colour)...

This is eight sprites all sprite 28 (hex) rendered at y=80 and y=A0, all colour 01

CaptureMooncrs2SpriteRender.PNG
 
N

NivagSwerdna

Active member
Feedback
1 (100%)
Credits
744CR
Bullets and Missile seem to render OK albeit in purple, red rather than white and yellow in Galaxian...

CaptureBulletsAndMissile.PNG


so it really looks like Sprite colour at certain positions is the problem... might have to think about this for a bit as it is far from obvious to me!
 
Macro

Macro

Active member
vacBacker
Feedback
4 (100%)
Credits
1,982CR
The colour changes could be down to iffy sprite ram (the 5 27LS01 chips in the corner of the PCB) if they are socketed then try swopping them around and see if the problem changes. These are not accessible by the CPU so are not covered by the ram test.
 
N

NivagSwerdna

Active member
Feedback
1 (100%)
Credits
744CR
Another weekend and another tinker... Socketed the Object RAM and tested the existing chips... seem OK... put in a couple of new ones for the time being... Removed all the Sprite RAM and tested individually... noticed that two had pins bent under and not contacting the sockets... swapped them around randomly as I put them back... Now the Raspberry Ripple effect is gone! Yay!

Screenshot_2022-02-20_16-05-44.png


Sprite rendering has got a lot worse but at least it is limited to Sprites only now...

Screenshot_2022-02-20_17-04-54.png


This should show eight sprites... I see only 4... and they are not rendering correctly either... one step at a time...

PS

Got Extron sync rock solid by increasing R65 to 270ohms

NivagSwerdna2022-02-20 17:07:13
 
N

NivagSwerdna

Active member
Feedback
1 (100%)
Credits
744CR
Discovered that main RAM was no longer passing and this was causing me issues when testing sprites..

Anyway the Brown Bodge wire that I never understood had come away, taking part of the track it was bodged too with it causing a break in the trace that gets from the CPU via the transceiver at 9B to the static RAM at 7N.

Removed and socketed 8D for good measure.

CaptureBrownWire.PNG


It sort of makes sense now. The brown bodge wire was to fix a broken trace marked A which they must have broken when installing the pull-up resistor pack that has mounted the decoder at 8C (which together produce the CS for the ROMs). They decided to attach the brown wire to the trace on the other side but didn't secure it with glue etc. and wear and tear have taken their toll.

Removed the bodge and replaced with a prettier bodge going from VIA to VIA.

CaptureGreen.PNG


Happier now and another bodge to add to the list of understood bodges.

Missiles and Shells are back...

Screenshot_2022-02-21_17-22-48.png


That's a relief. I thought I was losing the battle until I found the detached wire!

Missile and shell colours differ from MAME where they are White. Given that these are mixed by R58, R59 and R60 in Galaxian style I will probably leave them be.

Back to sprites...
 
N

NivagSwerdna

Active member
Feedback
1 (100%)
Credits
744CR
NivagSwerdna said:
Discovered that main RAM was no longer passing and this was causing me issues when testing sprites..

Anyway the Brown Bodge wire that I never understood had come away, taking part of the track it was bodged too with it causing a break in the trace that gets from the CPU via the transceiver at 9B to the static RAM at 7N.

Removed and socketed 8D for good measure.

CaptureBrownWire.PNG


It sort of makes sense now. The brown bodge wire was to fix a broken trace marked A which they must have broken when installing the pull-up resistor pack that has mounted the decoder at 8C (which together produce the CS for the ROMs). They decided to attach the brown wire to the trace on the other side but didn't secure it with glue etc. and wear and tear have taken their toll.

Removed the bodge and replaced with a prettier bodge going from VIA to VIA.

CaptureGreen.PNG


Happier now and another bodge to add to the list of understood bodges.

Missiles and Shells are back...

Screenshot_2022-02-21_17-22-48.png


That's a relief. I thought I was losing the battle until I found the detached wire!

Missile and shell colours differ from MAME where they are White. Given that these are mixed by R58, R59 and R60 in Galaxian style I will probably leave them be. [Thinking about it aren't they Yellow and White in Galaxian... OK... more thought required].

Back to sprites...
Click to expand...
 
N

NivagSwerdna

Active member
Feedback
1 (100%)
Credits
744CR
... in the previous photo it is 3L that is installed the wrong way around!

Removed and it is not surprisingly fried... socketed and put one in the other way around...

Capture3LTheOtherWayAround.PNG


Sprites are different... but we aren't there yet!

Screenshot_2022-02-23_18-43-42.png
 
Macro

Macro

Active member
vacBacker
Feedback
4 (100%)
Credits
1,982CR
I've seen it on several boards, but not all - it seems to affect moon cresta the most

there is a blanking signal generated from just before flyback to around 8 pixels down the new screen (which is why galaxian hardware can't draw sprites at the top and bottom of the screen - it may be that signal is not being generated correctly.
 
N

NivagSwerdna

Active member
Feedback
1 (100%)
Credits
744CR
My spare Sprite RAM is still to turn up in the post but my RAM replacement board arrived from China so gave it a blast...

CaptureAttackRAM.PNG


Seems to work fine...

CaptureAttackRAMTest.PNG


I made two variants for the two different board spacings I have... these are quite fiddly to make and hyper fiddly to install... not sure it is a viable product but let me know if there is interest.

I use edge pins for that authentic IC look and feel... and it even works in my dodgy single wipe sockets.
 
N

NivagSwerdna

Active member
Feedback
1 (100%)
Credits
744CR
Now the sprite RAM is replaced moving back to Shells and Missile generation... I think this should be the same as Galaxian... i.e. Yellow or White..

CaptureMissileAndShells.PNG


Clearly not as expected...

CaptureShellAndMissiles.PNG


on my Galaxian I see...

CaptureGalaxianShellsAndMissile.PNG


but on the Moon Cresta I see...

CaptureMCShellsAndMissile.PNG


Obviously no effort to make it look pretty... come on guys... line up those resistors...

Yellow Violet Brown Gold is 470 Ohm +/- 5% so that appears sensible on both boards..

Brown Black Brown Gold is 100 Ohm +/- 5% so that is as per the schematic on the Galaxian.

My colour blindness makes pesky resistor colour codes challenging... Brown Orange Brown? 130?

In any case the values are not far off 100... I would expect White and Yellow for Missiles & Shells?

The plot thickens.

NivagSwerdna2022-03-25 21:53:37
 
N

NivagSwerdna

Active member
Feedback
1 (100%)
Credits
744CR
Whilst I ponder replacing the 74LS368 which drives the colour output for shells and missiles I thought I would look at the spurious wires on the other 74LS368s...

CaptureInputJiggeryPokery.PNG


Looking at the schematics...

CaptureInputJiggeryPokery2.PNG


so this is a bit weird... Player 1 Left is wired to P2 Right, Player 1 Right is wired to P2 Left, and P1 Fire is wired to P2 Fire. Not sure why!
 
N

NivagSwerdna

Active member
Feedback
1 (100%)
Credits
744CR
Socketed and replaced 9D and got no change. Found what appears to be lack of continuity between one of the missile/shell resistors and 9D, so small I cannot really see the break, so repaired the trace.

CaptureBrokenMissileTrace.PNG


And now...

CaptureMissilesAndShells2.PNG


Perfect. We are back to the Galaxian White/Yellow colour scheme that I expected! Yay!
 
You must log in or register to reply here.
Top