Hi, all.
After experiencing multiple BSODs (MEMORY_MANAGEMENT and PAGE_FAULT_IN_NONPAGED_AREA) with my new NUC7i5BNH, I've found through trial and error that the cause appears to be memory errors - but the errors only occur when both slots are populated.
My NUC's hardware consists of:
2 x Crucial CT16G4SFD8213 16GB 2133MHz SODIMMs
1 x Crucial CT1050MX300SSD4 1TB M.2 SSD
The NUC has been updated to the latest 0047 BIOS, and the BIOS is using default settings. (I had previously turned Hyper-Threading off with the 0046 BIOS in case that was the problem, but it didn't help; after updating to 0047 I reloaded the default settings but that made no difference either.)
I've tested each SODIMM individually in both the lower and upper memory slots, and the system works perfectly with either one as long as the other slot is empty; the NUC runs stably and repeated tests show no memory errors. Populating both slots and booting into Memtest86, however, results in memory test failures. Memtest86 also sometimes crashes midway through the test, showing "CPU Error: Could not start CPU 0".
At first the memory errors were extremely intermittent and I had difficulty reproducing them reliably, but now they occur nearly all the time when both slots are filled and a Memtest86 test is run. Are there any further steps I should try before considering an RMA?