I can't say for sure that the two different RAM part numbers is causing BSODs. However, I've done some extensive research on the characteristics and (especially) the timings of the two RAM cards that you have. The two sites I used are here..
For part number M471A2K43
DB1-CWE
RAM reviews Samsung M471A2K43DB1-CWE 16GB, characteristics, performance
For part number M471A2K43
EB1-CWE
RAM reviews Samsung M471A2K43EB1-CWE 16GB, characteristics, performance
If you compare the two you will see that all the timings match, except for the latency measured during PassMark testing. Latency is the time it takes for the RAM to return a memory value after being selected, and since the timings of the two different part numbers (in clock cycles) are the same then the latency should be the same - but it's not. One RAM card had a measured latency of 37ns whilst the other was 55ns, to me (and I'm not a hardware expert) that is sufficient evidence that the two RAM cards are
NOT matched.
It would be nice if someone who is a hardware expert could comment, but I really would like to see how the laptop performs with just the original RAM card installed.
In the most recent dump, the MEMORY_MANAGEMENT on 14th April, there are no third-party drivers on the full call stack, that means that the only drivers being called were Microsoft drivers, which are considered flawless. That is a strong indication that the problem is hardware.
In this dump the call stack shows that address space cleanup was going on...
Code:
4: kd> knL
# Child-SP RetAddr Call Site
00 fffff50a`0404f058 fffff805`6a0bc49a nt!KeBugCheckEx
01 fffff50a`0404f060 fffff805`6a06d836 nt!MiDeleteVa+0x153a
02 fffff50a`0404f160 fffff805`6a06d94b nt!MiWalkPageTablesRecursively+0x776
03 fffff50a`0404f200 fffff805`6a06d94b nt!MiWalkPageTablesRecursively+0x88b
04 fffff50a`0404f2a0 fffff805`6a06d94b nt!MiWalkPageTablesRecursively+0x88b
05 fffff50a`0404f340 fffff805`6a06a94b nt!MiDeleteVad+0x88b
06 fffff50a`0404f3e0 fffff805`6a0bad31 nt!MiWalkPageTables+0x36b
07 fffff50a`0404f4e0 fffff805`6a07e830 nt!MiDeletePagablePteRange+0x4f1
08 fffff50a`0404f7f0 fffff805`6a48555f nt!MiDeleteVad+0x360
09 fffff50a`0404f900 fffff805`6a42441f nt!MiCleanVad+0x43
0a fffff50a`0404f930 fffff805`6a431550 nt!MmCleanProcessAddressSpace+0x137
0b fffff50a`0404f9b0 fffff805`6a70da92 nt!PspRundownSingleProcess+0x20c
0c fffff50a`0404fa40 fffff805`6a08e5c5 nt!PspProcessRundownWorkerSingle+0x32
0d fffff50a`0404fa70 fffff805`6a1265f5 nt!ExpWorkerThread+0x105
0e fffff50a`0404fb10 fffff805`6a204848 nt!PspSystemThreadStartup+0x55
0f fffff50a`0404fb60 00000000`00000000 nt!KiStartSystemThread+0x28
We can see calls to nt!MiCleanVad and nt!MiDeleteVad to remove an address space (the virtual memory in which a process runs), followed by calls to nt!MiDeletePagablePteRange, which deletes the page tables for this address space (these map virtual addresses to real memory addresses). Then things start to become unusual, we see another call to delete the address space, followed by multiple calls to re-check the page table structure, then a final call to delete the address space and then the bugcheck that caused the BSOD.
This then was a problem deleting an address space and the page tables associated with it. All that points straight at RAM - again.
I do appreciate that you need 32GB of RAM but all my spider-senses are telling me that you have mismatched RAM and that this is most likely contributing to your BSODs. The only way to conclusively prove that is to remove the new card and see whether it BSODs on the original one alone. I have no idea whether the competition you need the extra RAM for runs continuously, but if not would it be possible to remove the extra RAM at a time when the competition is not running?
It's going to be difficult to troubleshoot further with suspicion hanging over the RAM, with each new BSOD we'd always be wondering whether this is that RAM or whether it's something else...
One other possibility, and I have no idea what your financial situation might be, but
Crucial have compatible RAM for your laptop at $34.99 for a 16GB card. It may be wiser to scrap the two RAM cards you have and buy a matched pair of 16GB RAM cards from Crucial, or from another vendor.
It is possible that there are other issues on the laptop, I note that you said you had BSODs before you added the extra RAM, and we've already seen those fatal errors for the AX201 during a power transition, though that may be resolved by a driver update. I looked on the Dell support site for a suitable driver, but it seems that the G5-5500 came with a Killer wireless card, because they only have a Killer driver available. Have you replaced the wireless card as well?
Another bit of advice about drivers; laptops often have highly customised drivers, mainly for reasons of power saving. It's not uncommon to find that generic drivers (from Nvidia, Intel, etc.) do not function properly on laptops because they don't contain these customisations. It's always wiser with laptops to use only the drivers provided by the vendor's support site, and for you that means the
Dell support site. For example, the power transition failure message in your System log is typical of the sort of failure we'd expect if you don't use the customised Dell drivers.
