AMD vs Intel for FEA: Build a Computer

  • Thread starter KyleGranger
  • Start date
  • Tags
    Fea
In summary, a graphics card might speed up your design if it can be made to fit, but it is not guaranteed.
  • #1
KyleGranger
37
2
TL;DR Summary
I'm building a computer that I'll be using for FEA
I'm building a computer that I'll be using primarily for FEA. No games. I'm leaning towards the Ryzen 9 5950x. I have a budget up to $1000-ish +200 maybe for the CPU. That will hurt seeing how I'm considering a $750 CPU (5950x), but I'd be willing to if need be and the gain is significant. Unless someone has other suggestions? The threadripper in that range that I've found is the 2970WX. Any suggestions are greatly appreciated.
 
Computer science news on Phys.org
  • #2
Can your graphics card add more power to your design ie can FEA be run on a graphics card?
 
  • #3
Intel and AMD are very similar general purpose CPUs so it shouldn't make any difference. What's likely to be more important is processor speed.

As @jedishrfu suggested, a GPU will give you more power if it can be made to fit your algorithms.
 
  • Like
Likes jedishrfu
  • #4
jedishrfu said:
Can your graphics card add more power to your design ie can FEA be run on a graphics card?
The software shows comparisons of how the GPU can speed up simulation so yes. I think I'll save up for a better graphics card to get more bang for my buck.
 
  • #5
Just be aware that many "professional" software packages won't be able to take advantage of consumer GPUs. They are often designed so that you need something like an NVIdia Quadro for it to work, Note that it is not necessarily about the HW (consumer GPUs are often very fast), but about the drivers.

Note also, that some packages (e,g, COMSOL) will be able to use the GPU when editing etc (which is nice, especilly when working with 3D graphics); but they won;t necessarily be able to use the GPU to actually speed up the calculation,. Hence, you need to read the fine print to understand what "acceleration" actually means for the package you are using.
 
  • Like
  • Informative
Likes CalcNerd, KyleGranger, pbuk and 2 others

1. What is the main difference between AMD and Intel processors for FEA?

The main difference between AMD and Intel processors for FEA (Finite Element Analysis) is the architecture. AMD processors use a multi-core architecture, which allows for better parallel processing and faster calculations. Intel processors, on the other hand, use a single-core architecture, which may be better for single-threaded applications.

2. Which processor is more suitable for FEA tasks?

Both AMD and Intel processors have their advantages and disadvantages for FEA tasks. However, in general, AMD processors tend to be more cost-effective for FEA as they offer better multi-core performance at a lower price point. Intel processors, on the other hand, may be more suitable for complex and large-scale FEA simulations.

3. Does the clock speed of the processor affect FEA performance?

Yes, the clock speed of a processor can affect FEA performance. However, it is not the only factor to consider. The number of cores, cache size, and architecture also play a significant role in determining the overall performance of a processor for FEA tasks.

4. How much RAM is recommended for FEA simulations?

The amount of RAM recommended for FEA simulations depends on the complexity of the simulation. In general, it is recommended to have at least 16GB of RAM for basic FEA tasks. However, for more complex simulations, 32GB or more may be necessary to ensure smooth and efficient processing.

5. Are there any specific requirements for the graphics card when building a computer for FEA?

FEA simulations primarily rely on the processor, so a dedicated graphics card is not a crucial component. However, for advanced simulations that involve visualizations or 3D models, a graphics card with good processing power can help improve the overall performance and speed up the simulation process.

Similar threads

  • Computing and Technology
Replies
7
Views
2K
Replies
18
Views
4K
Replies
2
Views
3K
  • Computing and Technology
Replies
2
Views
6K
  • DIY Projects
Replies
7
Views
3K
  • DIY Projects
Replies
23
Views
4K
  • DIY Projects
Replies
7
Views
2K
  • Computing and Technology
Replies
16
Views
85K
  • Computing and Technology
Replies
2
Views
4K
Back
Top