lifewebb wrote:
I could tell you that I bought two 16GB modules from Amazon MX, thinking that it would make the computer even faster and they were sold under a brand called TimeTec which allegedly were the correct ones for this iMac 2019,1 - 27 inch screen. Compatible with Apple DDR4 2666MHz / 2667MHz for Mid 2020 iMac (20.1 / 20.2) / Mid 2019 iMac (19.1) 27-inch with Retina 5.
I have seen mixed results for people using the TimeTec memory & SSDs.
Plus, memory very rarely increases system performance unless your workloads require it It is very easy to determine if memory is hurting system performance. Use the computer for days without rebooting while doing your tasks making sure to also perform the heaviest workloads.
Then use Activity Monitor to check the memory usage. If the Memory Pressure graph is:
- red, then you need more memory.
- yellow, then you are borderline on having sufficient memory
- green, then check the Swap Used and Compressed Memory...if they are in the GB range, then you probably need more memory (especially if it is more than 2GBs).
The file system issues can easily cause performance issues, plus the Fusion Drive is a bit of a limiting factor. Of course 8GB of original memory is very limiting except for the most basic tasks. I don't know why Apple sells sub-par configurations without a clear warning that such a system is for only very basic usage since most average users won't realize it.
I discovered that the 16GB modules had been made by the same manufacturer that Apple had used for its 4GB modules - SKhynix.
While the memory chips may be the same, that is not the complete story. The supporting electronics on the memory stick must also be of high quality. Typically the minor players will skimp in order to save money so they can sell their products for less to make them more attractive. Plus those Skynix chips may not be 100% quality...they could be chips that did not pass all the rigorous tests other memory manufacturers require.
FYI, lots of manufacturers, even good ones may utilize partially bad chips by bypassing the bad sections in order not to lose money. So lets say a memory chip can support 4GB, but a 1GB portion is bad, they may use the remaining 3GB section. Usually that is not a problem if section used is 100% good.
It was a test of my own devising. Now the RAM consists of two 16GB blades which are correctly installed as the first two modules at the bottom of the tray. However, the day after doing this change, the computer again crashed and its report said 'Possible memory corruption'! So, I am no further on with whether the physical RAM is causing trouble.
The memory should be installed in the 2nd slot from the top and the bottom most slot IIRC. That should be the way the memory was installed from the factory. I've never seen memory in an iMac from the factory installed in slots next to each other.
Run MemTest86 to see if any issues are detected. Unfortunately the free version of the app is limited to just running the test in four loops each time, however, it can still have a good chance at detecting a memory related issue. With 32GB those four loops will likely take a day to complete (perhaps longer).
I would suggest reinstalling the original Apple memory back by themselves without the TimeTec modules to see if MemTest86 detects any problems and to see how the system performs. It is also possible the mix of Apple & TimeTec modules together is the problem....sometimes memory is not compatible with other memory. If you use both at the same time, then I would alternate the slots used by the Apple & TimeTec modules so matching pairs are in the DIMM0 slots and the other matching pair in the DIMM1 slots.
FYI, it can be very difficult to troubleshoot memory related issues unless the failure is extreme or very repeatable.