Author Topic: Regarding Comparing Intel to AMD- Stress Test & Performance Testing  (Read 384 times)

Offline TequilaChaser

  • AH Training Corps - Retired
  • Plutonium Member
  • *******
  • Posts: 10173
      • The Damned - founded by Ptero in 1988
I never thought I would see the day that a Performance Testing Software would openly admit that their Testing Software was built to optimize Intel components specifically

Quote
Software and workloads used in performance tests may have been optimized for performance only on Intel® microprocessors. Performance tests, such as SYSmark*, MobileMark*, and SPEC CPU are measured using specific computer systems, components, software, operations and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that product when combined with other products.

 Measured by SYSmark* 2014, a benchmark from BAPCo* consortium, which measures the performance of commonly used productivity applications, such as Microsoft Excel* and Adobe Acrobat*. On HD video playback, compared to 5-year-old system.

I like both Intel and AMD, I just have always figured and from doing test over the years, noticed on paper/results Intel Systems always look better for the most part, but have found in real life use that , that is not always true.....

Here is a Company that openly admits my thoughts....... I have never tried using their software, but think I might give them a try, just to see, when I get some time and money

TC
"When one considers just what they should say to a new pilot who is logging in Aces High, the mind becomes confused in the complex maze of info it is necessary for the new player to know. All of it is important; most of it vital; and all of it just too much for one brain to absorb in 1-2 lessons" TC

Offline Denniss

  • Nickel Member
  • ***
  • Posts: 607
Re: Regarding Comparing Intel to AMD- Stress Test & Performance Testing
« Reply #1 on: August 05, 2016, 03:17:59 AM »
Cudos to them they openly state this and shame on them doing this. But this trick was already known, AFAIR the german computermagazine c't (www.heise.de) once modified the CPU-ID string of an AMD CPU from Authentic AMD to Genuine Intel and the benchmark results raised a lot.
This behaviour/trick may have its origin in the Intel compiler they used as this always sets this CPU-ID trap.