STE video chip

[exxos]

I've done a quick video to explain all this in a nutshell.



Yes I have a cold. Yes I'm half asleep. Yes I need a better cam. And no I don't care

Though if I ever get my workshop straight. I may invest in a better recording setup.

[JezC]

Nice video - and, no I'm not being sarcastic!

Explains the problem & shows the difference very nicely.

So I need to check which version of the shifter is in my STEs & only test those with the 1712 shifters (if I have any)

[exxos]

So I need to check which version of the shifter is in my STEs & only test those with the 1712 shifters (if I have any)

It's worth checking the 588 just to verify the graphics only fail on the 1712.

[JezC]

So I need to check which version of the shifter is in my STEs & only test those with the 1712 shifters (if I have any)

It's worth checking the 588 just to verify the graphics only fail on the 1712.

Ok...I'll get ready to download & transfer the files over to floppy...

[chronicthehedgehog]

Mine's a 1712, I'll give it a test. I do get the odd weird error.

[chronicthehedgehog]

IMG_20231026_211826.jpg (22.86 KiB) Viewed 1436 times

IMG_20231026_211821.jpg (26.05 KiB) Viewed 1436 times

Definitely a bit of a mess

EDIT:

Audio crackle as well

[exxos]

Just done a bit more testing.

The "default" behaviour is this:

default.PNG (708.54 KiB) Viewed 1435 times

BUT. Turning the pot to a total of 50R solves that problem. But not much after (still turning the pot) it results in the "wobble" issue.

wobble.PNG (909.65 KiB) Viewed 1435 times

Going to 130R total the image is fine again.


Interesting the 32MHz and 16MHz are now synced to the rising edges for it to work best. Though in my original tests with the 588 shifter, they were synced on the falling edges.

IMG_1573.JPG (102.15 KiB) Viewed 1435 times

But I have a inverter on my test board now. So I guess I need to switch that back out to see if the syncs then change again or not...

EDIT:

IMG_1574.JPG (98.85 KiB) Viewed 1431 times

So with the buffer back in , the pot is still on the extreme 130R point, but the image seems fine, but I can't adjust anymore for rising edges to be in sync.

So I think ill switch to the inverter & buffer combo as as I did more tests with that in place.

Really needs more tests to figure out what the hell is going on with all this but someone else can do that work

[exxos]

Mine's a 1712, I'll give it a test. I do get the odd weird error.

Like what exactly ?

Definitely a bit of a mess

Looks like your not getting that expected diagonal lines but it does look more structured than mine

Audio crackle as well

viewtopic.php?p=78844#p78844

[chronicthehedgehog]

Like what exactly ?

Well it's probably just some general timing problem thinking about it, but about 50% of the times I load Lethal Xcess (the DBUG crack) the graphics are corrupt - quad vision like the bit planes are of sync is the only way I can describe it. It's strange that it's intermittent - especially after a cold boot

[exxos]

Well it's probably just some general timing problem thinking about it, but about 50% of the times I load Lethal Xcess (the DBUG crack) the graphics are corrupt - quad vision like the bit planes are of sync is the only way I can describe it. It's strange that it's intermittent - especially after a cold boot

Yeah sounds like the same problem. @Steve had problems with the we were demo, but don't think it was documented exactly what was wrong.

I guess we need more test programs to diagnose the issues. But not sure what stuff generally breaks. Maybe right timing demos. I guess you could try the closure demo as that pushes everything..