PCWorld Forums

PCWorld Forums: Interpreting The Results Of Worldbench 6 - PCWorld Forums

Jump to content

Page 1 of 1
  • You cannot start a new topic
  • You cannot reply to this topic

Interpreting The Results Of Worldbench 6

#1 User is offline   trustwave 

  • Newbie
  • Pip
  • Group: Members
  • Posts: 3
  • Joined: 18-March 10

Posted 18 March 2010 - 07:27 AM

Hi,

I have been using the WorldBench 6 to do some performance analysis. But looking at the results I cannot figure if it was the time taken to perform each operation by the application or some kind of benchmark values. I do not know if it is better to get a higher or a lower value. It would be great if someone can help me with this.

Thanks
--
Vinzy
0

#2 User is offline   JamesM 

  • Manager of Benchmark Development
  • PipPip
  • Group: PC World Labs
  • Posts: 18
  • Joined: 18-August 06

Posted 18 March 2010 - 03:31 PM

WorldBench scores are in seconds for individual tests so LOWER numbers are better. The overall WorldBench Score is a weighted average of all scores where HIGHER is better and the baseline system scored 100. So an overall score of 150 is essentially 50 percent faster and a machine that scores 50 is about half as fast as the baseline system.
0

#3 User is offline   trustwave 

  • Newbie
  • Pip
  • Group: Members
  • Posts: 3
  • Joined: 18-March 10

Posted 19 March 2010 - 05:48 AM

Thanks a lot James for your reply. It would be really helpful if you can tell me what weights are assigned to each of the tests so that, even if one of the tests is incorrect I will be able to eliminate those results and use the others to determine the overall performance.

This happened with the WinZip test where the system was waiting in the License screen for a very long time and this inturn affected the credibility of the tests.

--
Vinzy

View PostJamesM, on 18 March 2010 - 03:31 PM, said:

WorldBench scores are in seconds for individual tests so LOWER numbers are better. The overall WorldBench Score is a weighted average of all scores where HIGHER is better and the baseline system scored 100. So an overall score of 150 is essentially 50 percent faster and a machine that scores 50 is about half as fast as the baseline system.

0

#4 User is offline   JamesM 

  • Manager of Benchmark Development
  • PipPip
  • Group: PC World Labs
  • Posts: 18
  • Joined: 18-August 06

Posted 22 March 2010 - 04:13 PM

View Posttrustwave, on 19 March 2010 - 05:48 AM, said:

Thanks a lot James for your reply. It would be really helpful if you can tell me what weights are assigned to each of the tests so that, even if one of the tests is incorrect I will be able to eliminate those results and use the others to determine the overall performance.

This happened with the WinZip test where the system was waiting in the License screen for a very long time and this inturn affected the credibility of the tests.

--
Vinzy

View PostJamesM, on 18 March 2010 - 03:31 PM, said:

WorldBench scores are in seconds for individual tests so LOWER numbers are better. The overall WorldBench Score is a weighted average of all scores where HIGHER is better and the baseline system scored 100. So an overall score of 150 is essentially 50 percent faster and a machine that scores 50 is about half as fast as the baseline system.




Hi Vinzy,
WinZip is 10 percent of score, Office is 20, 3ds max is 5 percent per test, all other tests in WorldBench 6 are 10 percent each.

Did it happen just once or on every run of Winzip because if you set it to run 3 passes it will look at results and try to determine if they are all within range of eachother. If one result is way off from the others it will run additional passes of that test.

James
0

Share this topic:


Page 1 of 1
  • You cannot start a new topic
  • You cannot reply to this topic

1 User(s) are reading this topic
0 members, 1 guests, 0 anonymous users