If this is your first visit, be sure to
check out the FAQ by clicking the
link above. You may have to register
before you can post: click the register link above to proceed. To start viewing messages,
select the forum that you want to visit from the selection below.
There is, just look at the CPUMark in the graphs.
The results are a good reflection of the multi-threaded performance. (except for the single threaded graphs of course)
I might look at individual results as you suggest, but was hoping for something more expedient, like simply subtracting the ST score from the MT score and using that,but that's probably not a good idea.
Yes, there is a small component of the score made up of single threaded code. But most of the algorithms used in the CPUMark values are fully threaded. More fully threaded than most commercial applications in fact.
If this isn't enough download some baseline files in PerformanceTest and look at the individual results that make up the CPUMark (disregarding the single threaded test)
For my purposes, single-thread performance is not usually significant since most of my machines are running at 100%, 24/7. Perusing the CPU Benchmark charts, I see the combined score and the single-thread score are given. With this information, is it possible to use a formula to come to an approximation of the multi-thread performance of a CPU?
Edit: I think I inadvertently posted this in the wrong sub-forum, please move if appropriate.
Leave a comment: