Hey, if there's any NASA folk here or if you can drop a line to them... I'm kind of MIFFED at the regularity with which this issue is reported in the press as a "repeating pattern of ones and zeroes" and no one it seems, has any desire to divulge the actual pattern that is being received.
If we're not in the Cold War surrounded by State Secrets, it is time to crowd-source this mystery and that means provide actual dumps of the repeating data along with examples of recent correct data... and perhaps a complete dump of instruction and data memory to the extent of which it is known... so armchair enthusiasts (who aren't all 'armchair' as some of them might have actually worked on Voyager or be familiar with its instruction set) can spot any real pattern or peg the origin of the repeated data. Of course the objective is to create a fault in emulation that at least partly mirrors reality, so we can devise strategies against it.
When I was a tech fixing S-100 bus memory cards and to a great extent today for some systems, identifying faulty chips from diagnostics or observed behavior was the same as fixing the problem, because they were in sockets and could easily be replaced. I even had a hilarious fault in memory-map days where spelling of a word processing document was flipping on the display (bit 1) but printed properly.
This sounds as if a conditional instruction has suffered a bit flip changing the condition and trapping it in a loop from which it should escape. The nature of the 'gibberish' (contemptible term in this context) might yield clues to the nature of the problem or to knowledgeable persons, where the data came from if it is a bit flip in a pointer. Whether crowd sourced persons are capable to actually solve the mystery or not, they might at least offer good suggestions for devising diagnostics to narrow down the cause.
We owe Voyager all our efforts, as the most inaccessible computer system in human history.
If we're not in the Cold War surrounded by State Secrets, it is time to crowd-source this mystery and that means provide actual dumps of the repeating data along with examples of recent correct data... and perhaps a complete dump of instruction and data memory to the extent of which it is known... so armchair enthusiasts (who aren't all 'armchair' as some of them might have actually worked on Voyager or be familiar with its instruction set) can spot any real pattern or peg the origin of the repeated data. Of course the objective is to create a fault in emulation that at least partly mirrors reality, so we can devise strategies against it.
When I was a tech fixing S-100 bus memory cards and to a great extent today for some systems, identifying faulty chips from diagnostics or observed behavior was the same as fixing the problem, because they were in sockets and could easily be replaced. I even had a hilarious fault in memory-map days where spelling of a word processing document was flipping on the display (bit 1) but printed properly.
This sounds as if a conditional instruction has suffered a bit flip changing the condition and trapping it in a loop from which it should escape. The nature of the 'gibberish' (contemptible term in this context) might yield clues to the nature of the problem or to knowledgeable persons, where the data came from if it is a bit flip in a pointer. Whether crowd sourced persons are capable to actually solve the mystery or not, they might at least offer good suggestions for devising diagnostics to narrow down the cause.
We owe Voyager all our efforts, as the most inaccessible computer system in human history.