Try to drill down on the CPU with a tool, 31% is high
how many cores do you have ?

Below is an example of a 12 core that is barely experiencing load :



it's an intel, 6 cores, but allows 12 threads..
you can see how the load is equally balanced

Wow, your setup sounds extremely intense. very impressive.
I'm using an AMD processor with 8 cores if that makes sense

Try to find out if one of the core's is not peaking out
could be a single thread that is overloaded and creates the bottle neck

interesting! the last two cores seem to be doing very little compared to the other ones. I know nothing about computers but assume they should be evenly distributed?

you can create a program that is 'mono core'
and it will peak out the core it is running on
if not supporting multi-core,
it will not exhaust the full power of the machine

one of my private modules, i converted the startup from
mono thread to multi-thread, the startup was recuced
from 40-50 minutes (loading all history) to sub 10 minutes
just to give you a feel for what that means, same
hardware, but more efficiently used..

sometimes you can not throw more power to it
if the program is not able to consume it

but analysis should help you understand what
the problem is, and how you can improve it

I'd say in general that it is probably not a graphics card issue. If you were doing some 3D renders and running a screen capture and playing a video game all at the same time, then it might be graphics card limited. From the system stats, it seems that it's a RAM and processor issue. Like rleplae was saying, some processes are single-core and there really isn't much you can do to get around that. Sierra Chart might have an option to enable multi-threading. It'd be strange though if the software defaulted to single core, especially because modern CPUs are increasingly multi-threaded.

To use an analogy, your computer is like a kitchen trying to prepare different meals for a restaurant. If your chef (CPU) is slow, you can only make simple things like PB&J. If your kitchen space (RAM) is tiny, your chef is going to be tripping over himself trying to reuse counter space (memory) to try to make a bunch of meals at the same time. If you had a larger kitchen (perhaps 16 GB of 3200 MHz RAM), your chef would have plenty of space to prepare each individual meal.

The GPU is sort of like a sous-chef. It does similar things that the CPU does. To continue the analogy, the GPU is like a sous-chef who is excellent at preparing fish, but is somewhat slower at preparing salads. The CPU will delegate graphics rendering to the GPU when the computational load is large, but 3% GPU usage means that it's not being stressed.

Overall, you should probably look into more and faster RAM, while monitoring your CPU for usage spikes.

One last thing to consider too is the RAM speed. You should be able to find it in the task manager pretty easily. If your current 8 GB stick is at 1800 MHz, for example, and you buy another 8 GB at 3000 MHz, your overall memory will be limited to the lowest speed present.

There are many other aspects to consider in general, and I'd suggest visiting https://www.reddit.com/r/buildapc to ask the tech wizards over there any other questions you might have.

Sent using the NexusFi mobile app