8-bit acorn hardware • Re: Atom: Speckling of red dots around the screen edges.

But doesn't the fact that the RGBtoHDMI produces no speckling or sync problems on my other ATOM suggest it isn't an RGBtoHDMI problem or is it common that it, even though two machines are the same make/model it'll work on one but not the other? I'm just asking out of curiosity as you're the expert.

Are both machines the same Issue PCB?

Even if they are, each machine is somewhat unique, due to different components, tolerances, and usage patterns.

So the optimal DAC settings on one machine may not be optimal on another.

Could you try to take some screen captures (Button 2) of the various forms of noise you are seeing?

Unfortunately, going back to the welcome banner screen, the purple bar is still there, though the vast majority of the red speckling was gone (and was mainly restricted to the top and bottom of the screen, predominantly the bottom).

Have another read of this post.

The cause of the purple bar I think is crosstalk on the 10-way IDC cable, between conductors 4 (INV connected to D7) and 3 (PA). The PA signal of the 6847 is very susceptible to crosstalk, as the three signal levels (1.0V, 1.5V and 2.0V) are only seperated by 0.5V. The problems is actually worse when a Noise Killer is fitted, because D7 is driven from a 74HCT573 which has much faster edges than the original 2114 parts.

What I found was that cutting conductor 4 in the cable, as close as possible to the IDC header, makes a significant difference:I've just tried that again with the cable I made yesterday, and that resolved the purple bar completely.

Dave

Statistics: Posted by hoglet — Thu Mar 07, 2024 2:21 pm

---