tag:support.primatelabs.com,2011-01-31:/discussions/geekbench/27724-results-under-stress-to-take-underclocking-into-accountPrimate Labs: Discussion 2018-05-28T19:24:56Ztag:support.primatelabs.com,2011-01-31:Comment/453655452018-05-28T11:37:53Z2018-05-28T14:24:36ZResults under stress ( to take underclocking into account )<div><p>Restults of Geekbench benchmark is based on best results which means that it takes overclocking into account which means that this score you can have for a short time. And results are very misleading.</p>
<p>Problem is, that under stress CPU lower its frequency e.g. from 2.8GHz to even 1GHz.</p>
<p>It is not possibe to guess which CPU is better in long term usage.</p>
<p>Would it be possible to introduce Geekbench benchmark with long stress benchmark and add another score results.</p>
<p>Usually notebooks can have same score as desktop computers, but for shorter time.</p>
<p>Thank you.</p></div>skyfingertag:support.primatelabs.com,2011-01-31:Comment/453655452018-05-28T19:24:53Z2018-05-28T19:24:53ZResults under stress ( to take underclocking into account )<div><p>Hi,</p>
<p>Thanks for reaching out to us about this. I'll pass your suggestion along to my team and see if we can add more support for testing systems over a long period of time. Currently, with Geekbench 4 Pro, you can specify a number of iterations for each workload (or write a short script to run the entire benchmark via the command line several times in sequence), which may suit your needs as well.</p>
<p>All the best,<br>
Colin<br>
Primate Labs Inc.</p></div>Colin