STE video chip

[exxos]

Just screwing around now. I think this player uses the DAC.. It is struggling to play my tunes though :lol:

It totally freaked out with Dreamland..

IMG_0230.JPG

"Don't you want to" almost plays, albeit somewhat slowly :lol:

[fenarinarsa]

@fenarinarsa could you tweak your program to output all 1s to the DMA audio ? It may help me see what the timings are better to work out how much time there is to latch data. Or actually alternate 1 and 0 might also he useful.

Here you go. Note that I changed the blitter timings to fit the 50Hz frame cycle count so the background colors do not move on a STE with C300588 (haven't tested other setups yet).

czietz also pointed out it may be a ground bounce in the STE shifter.

It's great to see you people working on this issue this fast 👍

[exxos]

Here you go. Note that I changed the blitter timings to fit the 50Hz frame cycle count so the background colors do not move on a STE with C300588 (haven't tested other setups yet).

Thanks will have a look at it later!

[exxos]

@fenarinarsa I'm not sure if this one has an simply locked up or not. The background lines are still but there is nothing happening on the DAC at all.

[fenarinarsa]

@fenarinarsa I'm not sure if this one has an simply locked up or not. The background lines are still but there is nothing happening on the DAC at all.

If the lines are still then it's running, I know it may be a bit misleading :)

If the program crashes/locks up, the background would be full red or light blue.

I send %10101010 bytes to the DAC but maybe that's not what you expected?

[exxos]

I send %10101010 bytes to the DAC but maybe that's not what you expected?

hmm ok, I tried a few pins and they are all low.

I am just copying my tracker back to floppy as that should be pulsing at least :)

[exxos]

This is far more annoying than I first thought :lol: Driving 2x8bit dacs off the same 8bit bus, Cannot simply measure input and output a single latch :roll:

Anyway, this is CP and D3. Looks like data is valid for about 50ns, then CP, then data valid for 50ns more. So probably almost any latch would have plenty of time to latch data.

IMG_0231.JPG

IMG_0232.JPG

[exxos]

czietz also pointed out it may be a ground bounce in the STE shifter.

This is what he is showing. The top line is the D3 line glitching high as it latches.

IMG_0233.JPG

Looking again at low voltage levels..


ORIGINAL

IMG_0235.JPG

WITH GROUND WIRE

IMG_0236.JPG

The 2 bottom scopes are the GND points each end of the wire (original is without wire in place).

Going back to D3 is:


ORIGINAL

IMG_0237.JPG

WITH GROUND WIRE

IMG_0238.JPG

So it cracks constantly when the wire is not there and we get a peak low voltage of 1.68V. With the wire its more stable at 1.44V. Of course the time it spends at that level is also a factor. It generally seems to be around 20-40ns. So the "drop" of 0.24V is just enough to get it working. There are of course a lot less spikes as well in the image. Though with it being random I am just going for the overall peaks.

These problems are likely compounded by the Fast logic series as well. As mentioned many times before, they are not really good for "noisy environments" and going for a lower power slower chips such as the LS is generally a better choice.

[Steve]

Hey Chris what's the tracker in the second video called? The one that says 'digital audio'. I remember using that as a kid and thinking it was the coolest tracker ever.

[exxos]

Hey Chris what's the tracker in the second video called? The one that says 'digital audio'. I remember using that as a kid and thinking it was the coolest tracker ever.

http://www.atarimania.com/utility-atari ... 36960.html