13
Also one last thing to add here, there's obviously not just a difference on numb...
source link: https://twitter.com/andreif7/status/1447214554816122888
Go to the source link to view the article. You can view the picture content, updated content and better typesetting reading experience. If the link is broken, please click the button below to view the snapshot at that time.
Don’t miss what’s happening
People on Twitter are the first to know.
Thread
See new Tweets
Conversation
I come back to the sampling issue and these two figures in the recent discussion: https://twitter.com/andreif7/status/1446925061387132936…
We're getting identical power figures for Linpack 5.5W vs 5481mW, both same score.
However GB5 is 5.1W vs 4632mW, much larger difference.
Quote Tweet
Andrei F.
@andreif7
· Oct 9
Replying to @Golden_Reviewer and @D3ntheman
Maybe your testing method is wrong here instead.
Linpack has a very flat and constant power profile, GeekBench has a very variated power profile that has a large range.
If you do not sample every peak and trough in the data, you cannot get to accurate averages.
There are obvious data concerns on figures >7W, however there also much larger methodology issues in the community here that explain why there are noticeable differences in numbers. You have to understand what you're measuring beyond just throwing out numbers.
Also one last thing to add here, there's obviously not just a difference on numbers due to the devices doing something else, but simply because there seems a disagreement on workload delineation, for example my data vs Geekerwan's video, some large discrepancies can be explained:
New to Twitter?
Sign up now to get your own personalized timeline!
Sign up with Apple
Relevant people
Recommend
About Joyk
Aggregate valuable and interesting links.
Joyk means Joy of geeK