Closed Lincutt closed 1 year ago
LINE 4602 add
$currentResultLineEntry.Line = $currentResultLineEntry.Line -replace ' \(thread . of .\)',''
can avoid this bug
Your fix is actually more elegant than mine, I'm uploading a new version soon.
0.9.1.0 still has this bug with set "FFTSize=All" in config.ini because 4k-240k has 2 lines in prime95's log file, but 256k-51200k has only 1.
replace LINE 4626
if ($settings.General.numberOfThreads -eq 1) {
to
if ($foundFFTSizeLines.Length -eq 1) {
and LINE 4631
elseif ($settings.General.numberOfThreads -gt 1) {
to
else {
can fix it.
because 4k-240k has 2 lines in prime95's log file, but 256k-51200k has only 1
Wat. I have to test this. O.o
Oh boy, for up to 240k it's explicitly using 2 threads, and for anything above it seems to use two "internal" threads, which don't show up in the log file.
I managed to pinpoint this to the new TortureHyperthreading
setting in prime.txt that was introduced in Prime95 30.7.
If I disable it, I can restore the old behavior of two worker threads and thus two log entry messages per FFT size, without the "thread 1 of x" addition. Going to add this in the next release.
when I set "runtimePerCore=auto" and "numberOfThreads=2" in config.ini script-corecycler.ps1 can't change to next core after all selected FFTs passed.
the small FFTs (4k-240k) passed results in P95 log will have "(thread * of 2)" in it. script-corecycler.ps1 can't parse them correctly.