Garbled Screen only in Graphics Modes (Text Works Fine)

[NickH93]

I got sick of the issues with the last Apple //e I had and when the motherboard started falling apart, I bought a new one. I have a nice Rev B board now.

Anyways, when I boot up, I get the Apple //e screen, I can see it perfectly. I can program in BASIC, do whatever. Anytime color graphics are involved though, the screen becomes a mess of colours. Tried replacing the HAL chip, and both HAL chips I have heat up when it is in graphics modes. Tried 3 different video roms (two from my two boards, one from an enhancement kit), and I tried swapping any other IC I could and I have no progress. Self test gives me system OK. or occasionally will give me a RAM error. I've tried replacing all of the RAM, that makes no difference either.

Any ideas?

Edited January 24, 2017 by NickH93

[NickH93]

I am only replying to save anyone who may reply this: There is one chip that was recommended to replace for someone who had a similar problem. I couldn't replace it from my old motherboard because it was soldered on and that chip basically fell apart when I tried to desolder it. Now you see why I replaced that motherboard. It was all that corroded.

Anyways, it is chip LS374, and i have some coming in the mail. I will update when I try the replacement. In the mean time, any help would be appreciated if there are any other ideas!

 

Thanks

-Nick

[Iamgroot]

Doing an HGR at the prompt,

Then TEXT

Then CALL -151

Then 2000.2100

 

The HGR clears this area to all zeroes.

The hex dump will show if all the values are indeed zeroes or random numbers and will verify the RAM in that area.

The text screen is stored in a different area of memory so would not be affected.

[NickH93]

Doing an HGR at the prompt,

Then TEXT

Then CALL -151

Then 2000.2100

 

The HGR clears this area to all zeroes.

The hex dump will show if all the values are indeed zeroes or random numbers and will verify the RAM in that area.

The text screen is stored in a different area of memory so would not be affected.

I will post a picture of what the screen looks like in HGR mode when my phone has more charge to it and I can actually run the camera.

 

The hex dump showed all zeroes after doing this test. I also found a similar test where I set all data in the graphics parts of the memory (2000-5FFF) to FF and cycle through the modes and see if the data remains FF, which it did.

Edited January 25, 2017 by NickH93

[NickH93]

This is how the screen looks when I type HGR

[NickH93]

I don't mean to keep posting, but replacing the aforementioned LS374 chip changed nothing at all. So now I am not sure where to go.

[NickH93]

Sorry to keep updating, but I've found some things to better describe the behavior. First off, the LS244 buffer chips, I replaced those which did nothing either.

Either way, I've learned that pattern I pictured is when you show Hires black. Lores black also has it's own pattern. Running a color test on the master diagnostic disk, each color has it's own garbled pattern that is always the same for each color. It's wrong, but each color has it's own pattern and it is always the same wrong pattern.

Black has its own, yellow has it's own, etc. So something is consistently getting messed up when trying to display color. Does this help narrow down possibilities? Thanks!

[thorr]

It could be your display not liking the signal. Can you try hooking it up to a VCR through the video in port and then to the TV?

[NickH93]

It could be your display not liking the signal. Can you try hooking it up to a VCR through the video in port and then to the TV?

No luck there. Thanks though.

 

I replaced an LS10 chip on the board with the old one from the old Apple II, which I thought I did. Turns out I didn't. It flips the symptoms. Bad text, good graphics. So I'm starting to think both of those chips are bad from each board. I ordered two from mouser. At 79 cents a pop I'll try it out.

[thorr]

I think I may have stumbled onto your situation on my computer last night. The issue has to do with 80 columns mode and the second graphics page. I don't have time at the moment to tell you exactly how to do it but for starters, turn the computer off and on, and stay in 40 columns mode and see if it helps. Then try loading or plotting graphics to both graphics pages. Also, try hitting Esc then Control Q. I have a low-res graphics program I wrote and it ended up doing exactly what you described after messing around with 80 columns mode and the only fix was to turn the computer off and on again to fix it even though I went back to 40 columns mode first. Also, since you have a IIe, you could try removing the 80 columns card and see what happens. On my Apple IIc, there is a button that I was messing with and it didn't seem to do anything at the time but I was messing with it before I started having the problem for 80/40 columns and maybe that is why I am having the problem and need to push it again.

[thorr]

Just an update. I can easily replicate the issue by typing these commands on my Apple IIc or even on the AppleWin emulator. Start the computer and control reset to stop the disk.

 

10 GR

20 COLOR = 4 : PLOT 10,10 (Add more lines and colors if you wish to see the patterns for each color)

30 PRINT CHR$(4)"PR#3" : PRINT CHR$(0); : POKE 49246,0 (This is the command to start double low-res graphics.)

 

RUN

 

Hope this helps!

[NickH93]

Thanks everyone. I'm happy to say I finally fixed it through a strange thought process.

Early on, I replaced the HAL chip which made the text garbled and so I thought my replacement was no good either. I put the old one back in.

Fast forward to now, I replaced an LS10 chip on the board which ALSO caused the text to be garbled up and the graphics to work with messed up colors.

For some reason I thought "Replacing the HAL chip also gave these symptoms. Let's replace both chips that made the text garbled at the same time." and sure enough, that did it! Full color, full text. Now have my first fully functional enhanced //e

Sorry for all the posting. Maybe this thread will help someone some day.

Summary: New HAL and new LS10 chip was the fix.

[thorr]

Glad you got it working. Just curious, did you try the code I posted above? Does it show the same issues as you were having? Here is some better code that would show more on the screen:

 

10 GR

20 FOR X=1 TO 15

30 COLOR = X : HLIN 1,39 AT X

40 NEXT X

50 PRINT CHR$(4)"PR#3" : PRINT CHR$(0); : POKE 49246,0

RUN