@Background Pony #7F63  
That’s not a stretch at all, that’s exactly how it should be if it makes sense for said app to be threaded in the first place. It obviously depends on what you’re doing, but each thread needs at least a separate stack, and if you really want to have speedups anywhere near linear in number of cores, you need threads to share as little data as possible. For embarrassingly parallel problems, you will want to have twice as much memory available for twice as many cores available, otherwise you won’t be able to make full use of your cores.
 
But yeah, it depends on your programs rather than the CPU itself.

@Background Pony #7F63  
I wish I could upboat comments sometimes.

@plfx
 
Selection of cpu should have no effect on memory usage. I mean, if one of them has substantially more cores then maybe threaded apps spawning more threads could consume more ram, but that’s a stretch.
 
Ryzen vomiting RAM could be memed another way though:
 
“I’m Micron intolerant. I can’t consume it at high speeds or I get sick. Can you get me a Micron-free menu with wholesome Samsung B-die please?”

@plfx  
the reference is just a meme

SHiT, my RAM

Unused RAM is useless RAM.

Sounds more like a memory leak.

@plfx  
RAM and CPU usage often correlate. When a CPU is processing lots and lots of data very quickly, (which the new Ryzens seem to like to do a lot), most of that data will be quickly processed with the help of your system’s RAM. This means that its loaded on and off your RAM rapidly. This is done because RAM is easier for the CPU to access than other storage devices. Then, to ADD to this, after you CPU is done with its task, the RAM doesn’t stop. It continues to process the chunks of data in the event that the CPU needs it later, for fast access.

How can a CPU eat your ram.