Retina Mac External display problem

If my guess is correct, it would be a problem in the design of the output transceiver circuit on the RMP's motherboard, assuming the cable and receiver in the display are built to spec.

Could be an IC issue, could be the impedance control of the connector, could be the impedance control of the traces on the circuit board, could be poorly chosen passives, could be interference from nearby circuits, etc.

I wish I could give you a better answer but high speed signal integrity is one of the most obtuse and most black art aspects of electronics design, there's a fair amount of non-intuitive physics going on, e.g. signals going down a wire don't just go, they 'ricochet' off impedance differences. It's far too large a topic to treat in a HN comment and its whole system encompassing so 'what it could be' is a large set.

These types of things are not uncommon in hardware development, especially with high speed signaling. The interactivity of such systems can be high and hard to predict fully at times.

EDIT: Simply put - display outputs scan top to bottom, left to right. The fact that the pattern that triggers the issue appears to be horizontal means it would occur in the raw bit stream of the output in a periodic way. That means there is some low order frequency in the system caused by this pattern (maybe it repeats every 500khz or something like that). If there is something in the electronics of the output of the RMBp that resonates at that frequency (easier than you think), then it could possibly cause the described failure.

Sounds to me like your parent is suggesting it could be any and/or all of the above. Essentially, at very high speeds the whole link has to be very carefully balanced, so if any one piece is not built to a tight enough spec, things could fall apart.

I thought of a simpler, or at least more relatable, analogy:

Most people have been around a cell phone which was near an audio speaker and heard it make those awful snap and pop sounds.

It doesn't make sense initially that you would hear anything like that from a phone, after all they operate at very high frequency, 900 - 2400 Mhz, definitely not audible frequencies. But, the system works by breaking up radio time into chunks, several hundred chunks a second. The relative spacing of the chunks, a few hundred a second, creates noise that is easily picked up by the speaker and turned into audible sound.

In the same way with the laptop you've got this high frequency signal: HDMI / DVI but when there are particular patterns of bits being carried you can get lower frequency content, similar to the chunks being used by the cell phone. Just like the coil in the speaker gets excited by the radio chunks, you could have a problem where the output circuitry of the RMBp gets excited by these rare low frequency patterns leading to problems with the display.

Apple Cinema Display ‪23-inch DVI‬ (1920 x 1200) connected through a DVI/DP converter. It doesn't happen when I open the image in photoshop.

Ah fair enough, that made me suspicious in particular as that title seemed unrelated to the bug :-)

I wonder whether my HDMI -> DVI adapter is making a difference somehow?

awesome, thanks!

For what little it is worth, it doesn't repro on chromebook pixels.

Apple puts artificial restrictions on what tools you can use to write iPhone apps, and Macs are that allowed tool, so OS X isn't closed?

That is literally the dumbest argument I've ever read on HN.