- Aug 20, 2014
- 1
- 0
- 4,510
Hello,
In addition to my interests in all manner of computers, I also am quite interested in mathematics, and I recently discovered a highly recursive mathematical function called the Ackermann function.
As you can see in the article, there is a brief mention of this function being a useful tool for benchmarking computer performance, however when I looked up Ackermann-based benchmarks, only academic paper's about the subject seemed to show up in Google. As a result, I ended up programming my own. Once it was up-and-running, I tested in on both my computer (i7-2600k w/ GTX 580), and my roommate's (i7-4790k w/ ASUS Hero VII on-board graphics), and my roommate's computer scored about 25% higher. I'm not really sure if that accurately displays the difference in performance, because I can't seem to find consistent information on the GFLOP count of my roommate's processor.
Is this function an accurate way of measuring either CPU, GPU, or total computer performance?
In-case anyone wants to try this on their own computer, the .exe application and the source code in .h format are linked below. Ignore the broken "how it works" function. I have yet to get it to properly display the message that you can read in the source code, so if any programmers want to tell me how to fix it, that would be nice.
Application (EXE): https
/www.dropbox.com/s/bh5c72j395xz1m7/A44.exe?dl=0
Source (H): https/www.dropbox.com/s/guavn0c364w97hz/Form1.h?dl=0
In addition to my interests in all manner of computers, I also am quite interested in mathematics, and I recently discovered a highly recursive mathematical function called the Ackermann function.
As you can see in the article, there is a brief mention of this function being a useful tool for benchmarking computer performance, however when I looked up Ackermann-based benchmarks, only academic paper's about the subject seemed to show up in Google. As a result, I ended up programming my own. Once it was up-and-running, I tested in on both my computer (i7-2600k w/ GTX 580), and my roommate's (i7-4790k w/ ASUS Hero VII on-board graphics), and my roommate's computer scored about 25% higher. I'm not really sure if that accurately displays the difference in performance, because I can't seem to find consistent information on the GFLOP count of my roommate's processor.
Is this function an accurate way of measuring either CPU, GPU, or total computer performance?
In-case anyone wants to try this on their own computer, the .exe application and the source code in .h format are linked below. Ignore the broken "how it works" function. I have yet to get it to properly display the message that you can read in the source code, so if any programmers want to tell me how to fix it, that would be nice.
Application (EXE): https
/www.dropbox.com/s/bh5c72j395xz1m7/A44.exe?dl=0Source (H): https
/www.dropbox.com/s/guavn0c364w97hz/Form1.h?dl=0
Facebook
Twitter