Skip to content

Instantly share code, notes, and snippets.

@stevepond
Created September 12, 2012 21:24
Show Gist options
  • Select an option

  • Save stevepond/3710034 to your computer and use it in GitHub Desktop.

Select an option

Save stevepond/3710034 to your computer and use it in GitHub Desktop.

##Swipe Benchmarking - Round 2##

###Baseline###

  • Same HTML Markup
  • Same CSS
  • Loaded 200 copies of a single carousel item into carousel
  • Each carousel used a start button to trigger the test
  • The carousel advances automatically every second
  • Tests both show auto-advance and manual swiping

###Targets###

  • Google Nexus 1

###Environment###

  • Device Only

##Libraries##

  • Strategy1: iScroll
  • Strategy2: iScroll with peripheral blanked out Cards out of "peripheral", area beyond three center cards; their content is literally removed and stored into memory. When in view, the cached dom is reinserted into the cards
  • Strategy3: iScroll with peripheral blanked out + visibility is set to hidden Same as above, with an additional visibility set to hidden on the "blanked" cards

##Results##

##Observerations## Strategy 1 performed very poorly. Strategies 2 and 3 were neck and neck and performed a little better. Strategy 3 was the best, but just barely noticeably better over 2.

##Reccomendation## Go ahead and apply Strategy 3. Use UA to detect < Ice cream sandwich, and disable swipe; enable for >=ice cream sandwich.

##Estimate## 8 hrs

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment